USF Libraries
USF Digital Collections

The stress response, psychoeducational interventions and assisted reproduction technology treatment outcomes

MISSING IMAGE

Material Information

Title:
The stress response, psychoeducational interventions and assisted reproduction technology treatment outcomes a meta-analytic review
Physical Description:
Book
Language:
English
Creator:
Mumford, Karen Rose
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
infertility
depression
research synthesis
in vitro fertilitzation (IVF)
meta-analysis
Dissertations, Academic -- Measurement and Evaluation -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: The psychological impacts of infertility have been well documented in the literature, providing evidence to support that at least some women who confront infertility are at risk for heightened distress and depressive symptoms. In response to this accumulated evidence, it has been argued that psychoeducational interventions may provide an important component to the treatment of infertility. While several theoretical models postulate the effects of stress on infertility treatment outcomes, results of these investigations have led to conflicting conclusions. However, a synthesis of the accumulated data examining the effects of stress on ART treatment outcomes was nonexistent until the conduct of this study. Therefore, the purpose of this study was to investigate the impact of stress on ART treatment outcomes and to determine whether psychoeducational interventions mitigate the impact of stress experienced by women during an ART treatment program.Two hypotheses were tested: 1. Increased levels of stress will be associated with a lower likelihood of Assisted Reproductive Technology (ART) treatment success, and 2. Psychoeducational interventions will mitigate the effects of stress experienced during Assisted Reproductive Technology (ART) treatment. A meta-analysis analyzing the results for each hypothesis was tested through a hierarchical linear regression model. A total of 13 studies, representing 43 effect sizes, were included in the analysis investigating the relationship between stress and ART treatment outcomes. Results of the HLM regression model suggest that stress has a small negative association with ART treatment outcomes (d=0.2012, p < .05). The analysis investigating the relationship between psychoeducational interventions and stress included a total of 4 studies, representing 12 effect sizes.Empirical evidence gathered through this analysis revealed that the effect of psychoeducational interventions on the stress experienced by women participating in an ART treatment program were not statistically significant (d=0.3071, p > .05). However, because this analysis was based on such a small sample of studies, generalizations regarding the efficacy of psychoeducational interventions cannot be made. Therefore, research aimed at investigating the impacts of a variety of programs should continue in an effort to provide more conclusive information.
Thesis:
Thesis (Ph.D.)--University of South Florida, 2004.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Karen Rose Mumford.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 314 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001498259
oclc - 57710637
notis - AJU6864
usfldc doi - E14-SFE0000495
usfldc handle - e14.495
System ID:
SFS0025186:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001498259
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 041209s2004 flua sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000495
035
(OCoLC)57710637
9
AJU6864
b SE
SFE0000495
040
FHM
c FHM
1 100
Mumford, Karen Rose.
4 245
The stress response, psychoeducational interventions and assisted reproduction technology treatment outcomes
h [electronic resource] :
a meta-analytic review /
by Karen Rose Mumford.
260
[Tampa, Fla.] :
University of South Florida,
2004.
502
Thesis (Ph.D.)--University of South Florida, 2004.
504
Includes bibliographical references.
500
Includes vita.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 314 pages.
520
ABSTRACT: The psychological impacts of infertility have been well documented in the literature, providing evidence to support that at least some women who confront infertility are at risk for heightened distress and depressive symptoms. In response to this accumulated evidence, it has been argued that psychoeducational interventions may provide an important component to the treatment of infertility. While several theoretical models postulate the effects of stress on infertility treatment outcomes, results of these investigations have led to conflicting conclusions. However, a synthesis of the accumulated data examining the effects of stress on ART treatment outcomes was nonexistent until the conduct of this study. Therefore, the purpose of this study was to investigate the impact of stress on ART treatment outcomes and to determine whether psychoeducational interventions mitigate the impact of stress experienced by women during an ART treatment program.Two hypotheses were tested: 1. Increased levels of stress will be associated with a lower likelihood of Assisted Reproductive Technology (ART) treatment success, and 2. Psychoeducational interventions will mitigate the effects of stress experienced during Assisted Reproductive Technology (ART) treatment. A meta-analysis analyzing the results for each hypothesis was tested through a hierarchical linear regression model. A total of 13 studies, representing 43 effect sizes, were included in the analysis investigating the relationship between stress and ART treatment outcomes. Results of the HLM regression model suggest that stress has a small negative association with ART treatment outcomes (d=0.2012, p < .05). The analysis investigating the relationship between psychoeducational interventions and stress included a total of 4 studies, representing 12 effect sizes.Empirical evidence gathered through this analysis revealed that the effect of psychoeducational interventions on the stress experienced by women participating in an ART treatment program were not statistically significant (d=0.3071, p > .05). However, because this analysis was based on such a small sample of studies, generalizations regarding the efficacy of psychoeducational interventions cannot be made. Therefore, research aimed at investigating the impacts of a variety of programs should continue in an effort to provide more conclusive information.
590
Co-adviser: Ferron, John
Co-adviser: Kromrey, Jeff
653
infertility.
depression.
research synthesis.
in vitro fertilitzation (IVF).
meta-analysis.
0 690
Dissertations, Academic
z USF
x Measurement and Evaluation
Doctoral.
090
LB3051 (ONLINE)
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.495



PAGE 1

The Stress Response, Psychoeducational Interventions and Assisted Reproduction Technology Treatment Ou tcomes: A Meta-Analytic Review by Karen Rose Mumford A dissertation submitted in partial fulfillment of the requirement s for the degree of Doctor of Philosophy Department of Measurement and Research College of Education University of South Florida Co-Major Professor: John Ferron, Ph.D. Co-Major Professor: Jeff Kromrey, Ph.D. Cynthia Parshall, Ph.D. Karen M. Perrin, Ph.D. Date of Approval: November 9, 2004 Keywords: infertility, depression, research synthesis, in vitro fertilization (IVF), meta-analysis Copyright 2004, Karen R. Mumford

PAGE 2

DEDICATION This work is dedicated to my family who has educated me in the truly worthwhile things in life. To Mark, my husband, to whom I owe so much. It is through his patience, love, and support that this achievement was possible. To my parents, Robert P. and Margaret Hutcheson, who have given guidance and encouragement throughout my lif e, as a token of love, gr atitude, and respect. To my sisters, Sandra and Rebecca, whose fr iendship and companionship I cherish. In loving memory to my grandmother, Viola Rose, whose example through life and humor has inspired me daily to bec ome a better daughter, sister, wife, and woman.

PAGE 3

ACKNOWLEDGEMENTS I am extremely grateful to a number of individuals for the support provided to me throughout the process of comple ting this dissertation. I thank my husband, Mark, for supporting me through my entire program of study and for giving me the most precious commodi ty throughout this endeavor, the time to study, conduct my research, and write. I am appreciative to my family, friends, and colleagues for their continued intere st and who have assisted me in the completion of this research in one way or another. I am deeply appreciative to Dr. Alice Domar and Dr. Jackie Boivin, wh o provided me with not only requested information and data for this research, but also offered me assistance, supplemental information, and support. I am particularly grateful to my committee members, Dr. Kay Perrin and Dr. Cynthia Parshall, for the care with which they reviewed this study and for t heir guidance. I am especially indebted to my major professors, Dr. John Ferron and Dr. Jeff Kromrey, who have spent innumerable hours reviewing my work and continuously challenged me. Their knowledge about statistics and research is absolutely encyclopedic and impressive to me. The leadership and mentoring they have extended to me throughout my program of studies will serve as an example to me throughout my professional career. If I have seen farther than other s, it is because I stood on the shoulders of giants. -Isaac Newton

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES.................................................................................................vi LIST OF FI GURES...............................................................................................ix ABSTRACT..........................................................................................................xi CHAPTER ONE: INTRODUCT ION.....................................................................1 Stress.........................................................................................................1 Infertilit y.....................................................................................................3 Social Context of Infert ility...............................................................4 The Stress of Infertilit y....................................................................6 The Role of Psychoeducational Interventions in Infertility Health Care......9 Stress and Infertilit y Resear ch.................................................................10 Meta-Analysis in M edical Res earch.........................................................14 Problem St atement ..................................................................................15 Research Purpose and Ques tions ...........................................................16 Limitations of the Stud y............................................................................17 Professional Signific ance of t he Study .....................................................18 Definitions of Terms.................................................................................19 CHAPTER TWO: LITE RATURE R EVIEW .........................................................24 Introducti on..............................................................................................24 The Infertilit y Probl em..............................................................................25 Definition, Prevalence and Trends of Infertility in the United States.........................................................................................25 Physical Causes of Infert ility .........................................................32 Physical Infertility Tr eatment Op tions ............................................37 Hormonal Tr eatment ..........................................................37

PAGE 5

ii Surgical Tr eatment .............................................................42 Artificial In seminatio n.........................................................44 Assisted Reproductive Technol ogy....................................48 ART Success Rates.................................................52 Associated Costs and In surance fo r ART................54 Legal and Ethical Considerat ions for Infertile Couples and AR T.................................................59 Conceptualization and O perational Classificat ions of Stress...................61 Theories of Stress .........................................................................64 Measurements of Stre ss...............................................................77 Stress Res earch ............................................................................85 Stress and Infe rtility .................................................................................88 Emotional Aspects of Infert ility ......................................................91 Conceptualization of Stress in Infertility Re search......................101 The Stress and Infertilit y Relation ship......................................... 107 The Role of Psychoeducational Inte rventions in Infertility...........108 Summary of Research Examining the Relationship Between Stress and Infertilit y..................................................................116 Methodological Issues.................................................................126 Meta-Analysis in Research ....................................................................129 Definition, Development and Use of Meta-Analysis in Secondary Research ................................................................129 Meta-Analysis in Medi cal Resear ch............................................131 Summary...............................................................................................132 CHAPTER THREE: METHOD .........................................................................134 Problem Stat ement................................................................................ 134 Research Purpose, Ques tions and Hy potheses ....................................134 Method ...................................................................................................136 Formulation of the Re search Pr oblem......................................... 136 Identification of Studies ...............................................................139

PAGE 6

iii Evaluation of Data .......................................................................145 Coding the Quality of the Studi es.....................................145 Coding the St udies........................................................... 146 Analysis and Inte rpretati on..........................................................149 Effect Si ze........................................................................149 Incomplete Reporting of Result s......................................152 Combining Estimates of Effect Size.................................153 Identification of Outlie rs....................................................158 Statistical A nalysis ............................................................159 Interpreting the Results ...............................................................162 Generalizabil ity............................................................................163 CHAPTER FOUR: RESULTS ..........................................................................164 Problem Stat ement................................................................................ 164 Research Purpose and Questi ons......................................................... 165 Collection and Evaluati on of Studi es.....................................................166 Literature Review ........................................................................166 Identification of Studi es for Inclusion in the Meta-Analysis..........167 Coding the Characteristi cs of Included Studies...........................174 Findings .................................................................................................176 Hypothesis One...........................................................................176 Hypothesis Two........................................................................... 194 CHAPTER FIVE: CONCLUSION S..................................................................202 Purpose of Re search .............................................................................202 Overview of Method ...............................................................................203 Summary of Findings .............................................................................210 Limitations of Study ...............................................................................212 Implicatio ns............................................................................................213 Implications for Fu rther Res earch.......................................................... 215 Closing Rema rks...................................................................................218

PAGE 7

iv REFERENC ES.................................................................................................219 BIBLIOGRAPH Y...............................................................................................236 APPENDICE S..................................................................................................241 Appendix A: Electronic Databases Employed in Computerized Data Search ........................................................................................242 Appendix B: Qualit y Review Form........................................................243 Appendix C: Study Characterist ics Important to Coding .......................245 Appendix D: Study Coding Shee t.........................................................246 Appendix E: C oding Manual .................................................................259 Appendix F: Empirical Studies Located for Meta -Analysis....................264 Empirical Studies Meeting Inclus ion Criteria for Hypothesis One..........................................................................................264 Empirical Studies Excluded fr om Meta-Analysis for Hypothesis One........................................................................265 Empirical Studies Meeting Inclus ion Criteria for Hypothesis Two..........................................................................................267 Empirical Studies Excluded fr om Meta-Analysis for Hypothesis Two ........................................................................268 Appendix G: Conceptual Defi nitions of C onstructs...............................270 Appendix H: SAS Code for Comb ining Effect Sizes .............................272 Appendix I: SAS Code for Calculating SAMD Statistic for Hypothesis One..................................................................................276 Appendix J: Scree Plot Identifying Outliers for Hypothesis One...........278 Appendix K: SAS Code for HLM Regression Analysis for Hypothesis One..................................................................................283 Appendix L: SAS Code for Null Model HLM Analysis for Hypothesis One..................................................................................285 Appendix M: SAS Code for Moderat or Analysis for Hypothesis One – Duration of Infert ility................................................................. 287 Appendix N: SAS Code for Moderator Analysis for Hypothesis

PAGE 8

v One – Country of Study Or igin............................................................ 289 Appendix O: SAS Code for Combining Effect Sizes for Hypothesis Tw o.....................................................................................................291 Appendix P: SAS Code for Computing SAMD Statistic for Hypothesis Two..................................................................................294 Appendix Q: Scree Plot Identifying Outliers fo r Hypothesis Two..........296 Appendix R: SAS Code for HLM Regression Analysis for Hypothesis Two..................................................................................298 Appendix S: SAS Code for Null Model HLM Regression Analysis for Hypothes is Two .............................................................................299 ABOUT THE AUTH OR...........................................................................E nd Page

PAGE 9

vi LIST OF TABLES Table 1 Percent distribution of wom en 15-44 years of age by fecundity status, according to parity and age: United States 1982 to 1995...................31 Table 2 Effect size computat ion....................................................................... 152 Table 3 Evaluation of the quality of indi vidual st udies..................................... 243 Table 4 Studies reporting characteristi cs of sample for hypothesis one..........169 Table 5 Demographic information for studi es reporting the effects of stress on ART treatm ent outcome s..........................................................................170 Table 6 Studies reporting characteristi cs of sample for hypothesis two..........172 Table 7 Demographic information for st udies reporing the effects of psychoeducational interventions on stress experienced by women participating in an AR T treatment program ....................................................172 Table 8 Study c oding agreem ent..................................................................... 176 Table 9 Studies reporting the effects of stress as it relates to ART treatment outcomes .......................................................................................177 Table 10 Test for outliers among effect sizes for studies reporting the effects of stre ss on ART treat ment................................................................182 Table 11 Test of homogeneity of effect sizes including outliers for hypothes is one ..............................................................................................185 Table 12 Test of homogeneity of effect sizes excluding outliers for hypothes is one ..............................................................................................186 Table 13 Hierarchical random effe cts analysis for hy pothesis one..................188

PAGE 10

vii Table 14 Hierarchical random effects variance analysis for hypothesis one ................................................................................................................189 Table 15 Hierarchical random effects null model analysis for hypothesis one ................................................................................................................190 Table 16 Hierarchical random effects null model variance analysis for hypothes is one ..............................................................................................190 Table 17 Hierarchical random effe cts analysis of moderators for hypothes is one ..............................................................................................191 Table 18 Hierarchical random effects va riance analysis of moderators for hypothes is one ..............................................................................................192 Table 19 Hierarchical random effects null model analysis for studies reporting information on the duration of infertility for hy pothesis one.............194 Table 20 Hierarchical random effects null model variance analysis for studies reporting information on the dur ation of infertility for hypothesis one ................................................................................................................194 Table 21 Studies reporting the effect s of psychoeducational interventions as they relate to stress experie nced during ART treatm ent regimens...........195 Table 22 Test for outliers among effe ct sizes for hypot hesis two....................197 Table 23 Test of homogeneity of effe ct sizes for hypot hesis two.....................198 Table 24 Hierarchical random effect s analysis for hypot hesis tw o..................199 Table 25 Hierarchical random effects variance analysis for hypothesis tw o.................................................................................................................200 Table 26 Hierarchical random effects null model analysis for hypothesis tw o.................................................................................................................201

PAGE 11

viii Table 27 Hierarchical random effects null model variance analysis for hypothesis two...............................................................................................201

PAGE 12

ix LIST OF FIGURES Figure 1 Dot Plot for studies repo rting the effects of stress on ART treatment outcomes .......................................................................................180 Figure 2 Scree plot for studies repor ting the effects of acute stress measured at baselin e/pre-treatment on ART tr eatment outcomes................278 Figure 3 Scree plot for studies repor ting the effects of chronic stress measured at baselin e/pre-treatment on ART tr eatment outcomes................278 Figure 4 Scree plot for studies repor ting the effects of depression measured at baselin e/pre-treatment on ART tr eatment outcomes................279 Figure 5 Scree plot for studies repor ting the effects of acute stress measured during the follicular phase on ART treatment outcomes...............279 Figure 6 Scree plot for studies repor ting the effects of chronic stress measured during the follicular phase on ART treatment outcomes...............280 Figure 7 Scree plot for studies repor ting the effects of acute stress measured during oo cyte retrieval on ART tr eatment outc omes.....................280 Figure 8 Scree plot for studies repor ting the effects of chronic stress measured during oo cyte retrieval on ART tr eatment outc omes.....................281 Figure 9 Scree plot for studies repor ting the effects of acute stress measured at embryo transfe r on ART treatment outcomes...........................281 Figure 10 Scree plot for studies repor ting the effects of acute stress measured during the luteal ph ase on ART treatment outcomes....................282

PAGE 13

x Figure 11 Funnel plot for studies repor ting the effects of stress on ART treatment outcomes .......................................................................................187 Figure 12 Dot plot for studies reporting the effects of psychoeducational interventions on stress duri ng ART treatment regimens ................................196 Figure 13 Scree plot for studies repor ting the effects of psychoeducational interventions on acute stress experienced during AR T treatment regi mens........................................................................................................ 296 Figure 14 Scree plot for studies reporting the effects of psychoeducational interventions on chronic stress experienced during ART treatment regim ens................................................................................ 296 Figure 15 Scree plot for studies repor ting the effects of psychoeducational interventions on depression experienc ed during ART treatment regimens...297

PAGE 14

xi The Stress Response, Psychoeducational Interventions and Assisted Reproduction Technology Treatment Ou tcomes: A Meta-Analytic Review Karen Rose Mumford ABSTRACT The psychological impacts of infert ility have been well documented in the literature, providing evidence to support t hat at least some women who confront infertility are at risk for heightened di stress and depressive symptoms. In response to this accumulated evidence, it has been argued that psychoeducational interventions may pr ovide an important component to the treatment of infertility. Wh ile several theoretical model s postulate the effects of stress on infertility treatment outcomes, resu lts of these investigations have led to conflicting conclusions. However, a synthesis of the accumulated data examining the effects of st ress on ART treatment outc omes was nonexistent until the conduct of this study. Therefore, t he purpose of this study was to investigate the impact of stress on ART treatm ent outcomes and to determine whether psychoeducational interventions mitigate the impact of stress experienced by women during an ART treatment program. Two hypotheses were tested: 1. Increased levels of stress will be associat ed with a lower likelih ood of Assisted Reproductive Technology (ART) treatment success, and 2. Psychoeducational interventions will mitigate the effects of stress experienced during Assisted

PAGE 15

xii Reproductive Technology (ART) treatm ent. A meta-analysis analyzing the results for each hypothesis was tested th rough a hierarchical linear regression model. A total of 13 studies, representing 43 effect sizes, were included in the analysis investigating the relationsh ip between stress and ART treatment outcomes. Results of the HLM regression model suggest that stress has a small negative association with ART treatment outcomes (d=0.2012, p< .05). The analysis investigating the relationship between psychoe ducational interventions and stress included a total of 4 studies, repr esenting 12 effect sizes. Empirical evidence gathered through this analysis revealed that the effect of psychoeducational interventions on the stress experienced by women participating in an ART treat ment program were not statistically significant (d=0.3071, p>.05). However, because this analysis was based on such a small sample of studies, generalizations regar ding the efficacy of psychoeducational interventions cannot be made. Therefore, research ai med at investigating the impacts of a variety of pr ograms should continue in an effort to provide more conclusive information.

PAGE 16

1 CHAPTER ONE: INTRODUCTION Stress is a common part of life and has come to describe a myriad of events or circumstances in a variety of contexts. For many couples, the problem of infertility diagnosis and treatment is one of life’s stressful circumstances that they face. The role of stress in infertilit y research has been a topic of interest for many decades. The field of research in the study of stress and infertility advanced along with technological and medi cal options in the treatment of infertility. While originally, research fo cused on whether stress was the cause of infertility, current research focuses on t he effects of stress on infertility treatment outcomes. Although medical ad vancements in the treatment of infertility provides hope for conceiving a child to infertile couples, the plethora of treatment interventions available to couples may also be the source of considerable stress. Recognizing this paradox, many physicians and medical facilities offering Assisted Reproduction Technology services to infertile couples have begun to offer psychoeducational interventions to pati ents. Only recently has interest in the efficacy of thes e programs grown. Stress Stress is a complex, dynamic in teraction between a person and the various conditions, changes, and the demand s of his or her life. Stress can be categorized as either acute or chronic. Acute stress is a short-lived response to

PAGE 17

2 a one-time incident or event that usually comes and goes quickly. Its effects can last from minutes or hours to days or w eeks. Chronic stress is caused by a continuing string of stressful incidences or an ongoing situation. Stress is experienced in various degrees. Low levels of stress may not even be noticeable while slightly higher levels of stress c an be positive by challenging one to act in creative and resourceful ways. However, high levels of stress can be harmful leading to a variety of illnesses and chronic diseases (Millon et al., 1982). Over the past several decades, the interest in “stress” has increased leading to a number of articles and studi es published from medical, physiological, biochemical, sociological, psychologi cal, educational, and even spiritual perspectives (Breznitz & Goldberger, 1993; Selye, 1983). Most theoretical models of stress conceptualize stress as a response elicited by an individual to an event or situation. M odels based on theories of stress as a response explain stress as a physiological adjustment pr ocess and postulate that stress is an internal response. Selye’s (1936) psycho-physiological model, “General Adaptation Syndrome” (GAS), presents stress as a non-specific, physiological response that consists of three sequent ial stages of alarm, resistance and exhaustion. He later broadened his theory to recognize stress as an internal condition of an individual when faced wit h environmental stressors. Selye also highlighted that any emotion or activity, whether it produced joy or sadness, causes stress. In addition, he noted that stressful life events often result in disease and unhappiness w hen individuals are unprepared to cope with these events (Selye, 1993).

PAGE 18

3 Infertility Fertility is the ability of a m an and a woman to reproduce. The epidemiological term, fecundability, refe rs to the monthly probability of conception without the use of any contraception (Jansen, 1993; Tsaltas, 1997). A couple is said to be subfecund, or infe rtile, when there is an involuntarily long interval until their first conception or between births. A common definition of infertility is the state in which a couple desiring to have a child cannot conceive after 12 months of unprotected intercourse (Tsaltas, 1997). Failure to conceive after 12 months of unprotected intercourse is taken to be abnormal as 90% of couples will have conceived within that time (Tietze, 1956, 1968; Tsaltas, 1997). Infertility is either classified as prim ary or secondary. Primary infertility refers to women who have never ac hieved pregnancy, whereas, secondary infertility refers to those women w ho have achieved pregnancy at least once before, regardless of the outcome, and who cannot achieve a subsequent pregnancy (Thonneau et al., 1991; Seibel, 1993; Tsaltas, 1997). Sterility, on the other hand, is the absolute inability to r eproduce. When the fecundability of a couple is zero, for whatever reason, the couple is defined as infecund, or sterile (Tsaltas, 1997; Jansen, 1993). Although t he true incident rate of sterility is unknown, it is believed that 3-5% of the population is sterile (Spira, 1986; Jansen, 1993, Tsaltas, 1997). Estimates regarding the actual in cidence of infertility, primary and secondary, vary. The sole reliable sources of demographic information about infertility and the use of infertility servic es in the United St ates are national

PAGE 19

4 surveys conducted by the National Center fo r Health Statistics, the last of which was carried out in 1995 (Seibel, 1993; CDC, 1997). At that time, an estimated 7.1% of married couples, or 2.1 million, in which wi ves aged 15 to 44 years were infertile. Among all women of reproducti ve age, 24% of women of reproductive age were surgically sterile. An additi onal 10.2% (6.1 million) had impaired fecundity (CDC, 1997). The proportion of patients seeking treatment presenting with primary and secondary infertility hav e remained remarkably constant with 67% to 71% of patients categorized as presenting primary infertility and 29% to 33% presenting with secondary infertility (H ull et al., 1985; Templeton et al., 1991; Thonneau et al., 1991; Tsaltas, 1997). Social Context of Infertility For many American couples, raising a fam ily is a major life event for which they have planned as meticulously as they did their education, career, and finances, carefully weighing all factors and waiting for the most opportune time to start a family. Most couples assume that they are in control of their reproduction and that when they are ready to begin their family, they will conceive with ease. Unfortunately, many couples desiring to conceive a child face the problem of infertility. For many year s, the incident rate of infe rtility was expected to range around 10% (Tsaltas, 1997). However, a num ber of recent studies give a clear indication that the actual incident rate of infertility ranges fr om 13.5% to 18.4%, which translates to one in seven women. These studies demonstrate that infertility is a common and important health problem (Tsaltas, 1997).

PAGE 20

5 Currently, more than 5 million coupl es in the United States can be classified as having a substantial infert ility-related problem. At least 8% of married couples will be initially unsuccessful in their attempts to have a biological child. The inability to conceive can be the source of considerable trauma and often prolonged stress, which may lead to a series of steps involving medical intervention for those who can afford t he expense (Schneider 2000; Whiteford & Gonzalez, 1995; Menning, 1980; 1982; Pearson, 1992). While most Americans view parenti ng as a central life role, American society emphasizes controlling fertility th rough contraception and on choosing when to have children and raise a family (Pearson, 1992). American society and many of our idealizations when gro wing up focus on becoming parents and raising children as an expected experience in adult life. According to Cahill and Suchy (1981), the family forms an import ant part of American social structures and is an integral part of our own identities as indi viduals. Veevers (1980) and Miall (1985) identify two pr edominate procreative soci al norms in American society: 1. All married couples should reproduce, and 2. All married couples should want to reproduce. These social norms are also support ed through American governmental policies that encourage reproduction and reward the image of parenthood through policies such as income tax deductions. Furthermore, Greil (1991) states “…the heart of the experience of infertility appears to lie in the inability to proceed with one’s life according to life course norms that are both rein forced by others and

PAGE 21

6 accepted as valid by the affected individual.” In addition to t he social affirmation, parenting often involves a confirmation of our sexual identity and feelings of selfworth (Greil, 1991; Shepherd, 1992; Whiteford & Gonzales, 1995). The Stress of Infertility While infertility can be a painful and devastating experience for both men and women, their responses are influenced by differential role expectations and socializations. The strong desire to hav e children and the subsequent responses to infertility have been shaped by our cult ure through a complex system defined by personal, familial, social, and medical expectations that transcends sex, age, religion, ethnicity, and socio-economic cl ass (Whiteford & Gonzalez, 1995). In American society, much of women’s per sonal and social identity is linked to motherhood. The experiences of pr egnancy, childbirth, and motherhood are seen as an intrinsic part of a woman’s adu lt life (Pearson, 1992). In a culture where womanhood is sometimes thought of as synonymous with motherhood, infertility, for many women, carries a stigma borne of shame and secrecy. Despite increasing awareness of the di agnosis and treatment options, infertility remains an “invisible” health issue. Not vi sible, life threatening, or disfiguring, infertility is often a secret life crisis ex perienced by couples in isolation (Menning, 1982; Sandelowski & Pollock, 1986; P earson, 1992; Whiteford & Gonzalez, 1995; Shoener & Krysa, 1996; Greil, 1997). Shaped by cultural, social, and personal expectations, infertility can create overwhelming stress, which tests a coupl e’s normal coping mechanisms because it is usually unexpected, may be unexp lained, and lasts for an indeterminate

PAGE 22

7 length of time. The diagnosis and attempte d treatment of infertility is often an acute and unanticipated life crisis of consi derable proportion for couples. A state of crisis can be defined as a disrupti on in the steady state, or a period of disequilibria. Elements common to a state of crisis include: A stressful event occurs that poses a threat that is insoluble in the immediate future, The problem overtaxes the exis ting resources of the person(s) involved because the remedy or solution is beyond traditional problem-solving methods, The problem is perceived as a thr eat to important life goals of the person(s) involved, and The crisis situation may reawak en unsolved key problems for both the near and distant past (Menning, 1980). In addition, infertility can create a chroni c or prolonged state of crisis with no identifiable solution (Menning, 1980; Boivin et al., 1995; Whiteford & Gonzalez, 1995; Schoener & Krysa, 1996). Medical developments in the treatment of infertility have led to a “merrygo-round” of interventions available to thos e couples that can af ford them. Prior to these developments in medical science, an infertile couple would decide either to remain childless or to adopt. T oday, however, couples have an excessive number of interventions available to them increasing hope that they will be successful in achieving and completing a pregnancy. In the current venue of infertility treatment, couples may c hoose to undergo years of treatment,

PAGE 23

8 postponing resolution to the life crisis they are experiencing. These advancements in medical treatment, while offering hope to infertile couples, may create unrealistic expectations in which in fertile couples define themselves not as childless, but as ‘not yet pregnant’ (Boi vin et al., 1995; Whiteford & Gonzalez, 1995; Schoener & Krysa, 1996; Menning, 1980). In addition to the emotional stress and responses shaped by societal expectations, couples have described the medical tests and treatments for infertility as stressful. When a couple decides on the necessity of seeking specialized medical attenti on because of their inability to conceive, the initial medical interview investigates extremel y personal subjects: the couple’s sexual performance, sexual history including the frequency of sexual intercourse, premarital and extramarit al relationships, previous pregnancies including abortions and miscarriages, attitudes about sex, and usual sexual practices. These very personal questions can be th reatening, embarrassing, intrusive or demeaning for the couple (Abbey et al., 1992). The medical treatment and interv entions for infertility have been universally described by patients as painful, embarrassing, and physically, psychologically and financially draini ng (Berg & Wilson, 1991; Boivin & Takefman, 1995; Schneider, 2000 ). Side effects from m edication, recovery from surgery, and time loss at work due to frequent medical appointments have all been identified as stressful events by c ouples. While these experiences are more devastating for some individuals than for others, the experiences can result in disrupted relationships, increased so cial isolation, depression, and increased

PAGE 24

9 hopelessness (Abbey et al., 1992; Domar, et al., 1992; Pearson, 1992; Whiteford & Gonzalez, 1995; Greil, 1997). The Role of Psychoeducational Interventions in Infertility Health Care Cognitive therapy is in creasingly being accepted as an effective treatment option for depression (Domar et al., 2000). Several studies with cancer patients combined support with cognitive-behavioral techniques. The benefits of this educational model included decreased psychol ogical distress, longer life span, and decreased mortality (Helgeson & Cohen, 1996; Domar et al., 2000). Another study including patients with multiple scler osis emphasized coping-skills training. This study demonstrated gr eater advantages in wellbeing and coping for the patients receiving the copi ng skills training than those who participated in only a peer telephone support group. Cognitive-be havioral approaches have also been shown through a multitude of studies and investigations to be effective in reducing symptoms of depression and decr easing health costs in patients with a wide variety of conditions including cardiac, abdominal, orthopedic, dental surgery, and invasive medical procedur es (Mandle et al., 1996; Domar et al., 2000). The most common intervention to help couples cope with the stress of infertility in the United States is pati ent education and social networking through support groups. While the programs desi gned for couples with infertility vary from program to program, some elem ents common among all of them include educational programs and emot ional supports designed to alleviate symptoms of stress for individuals or coupl es experiencing infertility. One such program is the

PAGE 25

10 Mind/Body Program for Infe rtility. Based on the elic itation of the relaxation response exercise, this program follows a 10-week protocol designed to treat individuals experiencing any medical sympt oms caused or aggravated by stress. In addition to relaxation-response training, The Mind/Body Program for Infertility and other similar programs include se ssions focusing on stress management training and strategies, exercise, di scussions and education on nutrition, and group support. Couples parti cipating in cognitive-behavio ral therapy are typically introduced to a wide variety of techniques including relaxation-response training, cognitive restructuring, emotional expression, and nutrition and exercise education relevant to infertility (Domar et al., 1990; 2000; Connolly et al., 1992). Stress and Infertility Research Given the developmental salience of reproduction, the intimacy of the situation, and the uncertainty of the outcome, the role stress plays in infertility and infertility treatment has been a common t opic of research in the area of reproductive health and medicine. Contem porary studies dominating this area of research encompass four major themes: 1. Does stress cause infertility? 2. Does infertility cause stress? 3. Does infertility treatment cause stress? 4. Does stress affect the success of infertility treatment? The first area of research, which has a long-standing focus in medical research, explores the possibility that infertility may have psychological causes. These studies are based on the psychogenic hypothe sis. The foundation of this model

PAGE 26

11 rests on the assumption that many case s of infertility, especially non-organic infertility, are caused by an unconscious resistance to motherhood on the part of the infertile woman. These earlier studi es attempted to show that unconscious conflicts and/or psychological maladjustm ent caused infertility. While several recent studies provide some support for the psychogenic hypothesis, most researchers now reject the psychogenic hypothesis (Greil, 1997). However, citing the limitations and flaws to prev ious studies, some researchers have attempted to revive the psychogenic hypothesi s in recent years. In addition, at least one team of researchers has argued for an interactive model, as opposed to the current approach, which assumes that infertility must either cause psychological distress or be caused by it (Greil, 1997; Csemiczky et al., 2000). In addition to these earlier research initiatives, t he literature has documented the psychological impact of in fertility and infertility treatment. Most contemporary studies examining the rela tionship between infertility and stress assume that infertility is the source, ra ther than the cause, of psychological distress known as the psychological cons equences hypothesis. The majority of the literature focusing on the experience of infertility is primarily, but not exclusively, qualitative in nature and provides a more complete picture of the experiences of couples facing infertility. The anecdotal evidence about the correlation between infertility and stress as well as the destructive impacts that infertility-related stress can have on the marriage and the couple’s quality of life is compelling. The research findings in this area of interest can be divided into studies on the psychological impact on women, men, and couples experiencing

PAGE 27

12 infertility and treatment. This area of research has advanced the field in that much of the qualitative research analyzes the experiences of infertility within its social context, paying special attenti on to gender roles, family structure and relationships, the effects of the medica l establishments, and the importance of assisted reproductive technology (Andrews et al., 1992; Greil, 1997; Hart, 2002). Growing out of this field of research are studies examining the relationship between stress and outcomes of assist ed reproduction technology (ART). Success of ART, defined as a live birth delivery, depends on many factors such as maternal age, type of infertility, and t he experience of the medical clinic and staff. Psychological stress and coping styl es may also play a role in the success of ART (Klonoff-Cohen et al., 2000). Ho wever, results from these studies are varied and the role of psychological fact ors and the impact of stress on treatment outcomes are yet to be established (New ton et al., 1990; Smeenk et al., 2001; Wilson & Kopitzke, 2002). As evidence about the stress of infert ility and of infertility treatment has mounted, physicians who treat couples wi th fertility problems have found they need to be concerned not only with the medi cal and physiological aspects of the problem but also with its psychosocial ramifications. To this end, several interventions have been implemented with pat ients including, but not limited to, traditional individual psychotherapy, which often emphasize the similarities in psychological responses between infertility and grief work, as well as support groups that allow the couple to share their experiences with others who have faced similar situations (Honea-Fleming, 1986; March, 1986; Andrews et al.,

PAGE 28

13 1992; McNaughton-Cassill et al., 2000). As an extension of the research examining the effect of stress on ART treatment outcomes, researchers have begun to examine the efficacy of psychoedu cational interventions as well as benefits to patients in terms of success ra tes with ART treatment (Domar et al., 2000; McNaughton-Cassill et al., 2000). As with much of the research that has been conducted to date relating to stress and infertility, these recent st udies have been criticized for various limitations. One limitation cited is t he use of convenience samples and/or small sample sizes. In addition, much of the research on this topic is based on nonexperimental designs. Several statis tical models using combinations of biomedical factors in relation with in vitro fertilization outcomes demonstrate limited external validity because they are based only on stable variables. Additionally, in the area of infertility, even well designed, randomized, controlled trials rarely have sufficient statistical power to demonstrate small but clinically significant differences between contro l and treatment groups (Hughes, 1992; Smeenk et al., 2001; W ilson & Kopitzke, 2002). Stress and infertility can be viewed as dynamic and circular. Studies investigating the role of stress on infe rtility treatment outcomes have led to conflicting results. Several methodologic al issues may have contributed to these varying results including small sample si ze, insufficient amount of power to detect small but significant differences, and homogenous groups of women in terms of ethnicity, age, and socio-economic status in cluded in the samples. This study will use the accumulated evidence regarding the role of stress and the efficacy of

PAGE 29

14 patient education programs through psychoe ducational interventions on infertility treatment outcomes through a meta-analysis. Meta-Analysis in Medical Research Gene Glass first used the term “meta -analysis” in 1976 to refer to a philosophy, not a statistical technique. Glass argued that a review of the literature should be as systematic as prim ary research and should interpret the results of individual studies in the cont ext of the findings. Over the past three decades, meta-analysis has grown from an unheralded preoccupation of a very small group of statistici ans working on problems of research integration in education and psychotherapy to an ac ademic industry encompassing an assortment of procedures used in a variet y of disciplines. Its popularity in the social sciences and education pales in comparison to its influence in medical research. Evidence-based medicine has been given increasingly more emphasis in recent years. Evidence-based m edicine focuses on the examination of empirical evidence from clinical resear ch for sound medical decision-making. Meta-analysis provides a quantitative approa ch to the review of the medical literature and is especially useful in prov iding information regarding the strength and quality of the evidence either supporting or refuting a medical practice as well as providing empirical evidence fo r developing practice guidelines. In medical research, meta-analysis uses the accumulated evidence about a treatment or procedure to provide guida nce to clinicians and to suggest directions for future study. The use of meta-analysis has generated considerable interest in the medical lit erature and has proven to be a powerful tool in the field

PAGE 30

15 of perinatology. Likewise, obstetrics is one field leading other medical specialties in the attempt to systematically review all randomized trials conducted in its discipline (Hughes, 1992; Pei pert & Bracken, 1997). There are several advantages to conduc ting a meta-analysis. First, it allows the researcher to identify gaps, problems and limitations in the primary research base. Many individual trials and observational studies lack sufficient statistical power to detect small but c linically significant differences. One advantage of a meta-analysis is its ability to increase statistical power to detect overall differences between groups and within subgroups. In addition, this technique allows researchers to resolv e some uncertainty and controversy when individual research studies provide oppos ing conclusions. Through increasing sample size, meta-analysis can prov ide stronger evidence for or against a treatment effect than one can derive from any of the individua l studies because a more precise estimate of t he effect size or measure of association is generated. In addition, meta-analysis allows researc hers to investigate research questions not posed at the start of an individual treatment tria l. Finally, another major advantage of this type of research review is the opportunity for others to judge the quality of its conclusions (Hughes 1992; Peipert & Bracken, 1997). Problem Statement The psychological impacts of infert ility have been well documented in the literature. While resear ch findings regarding the prevalence of distress and depressive symptoms in infertile women ar e inconsistent, there is evidence to support that at least some women who confront infertility are at risk for

PAGE 31

16 heightened distress and depressive symptom s. Psychoeducational interventions may provide an important com ponent to the treatment of in fertility. In addition, these programs may prove to be an effe ctive intervention in preventing the anticipated increase in psychological di stress as the duration of infertility increases. While several theoretical m odels postulate the effects of stress on infertility and infertility treatment outc omes as well as the efficacy of psychoeducational interventions, a synt hesis of the accumulated data incorporating a qualitative assessment of the methodology of reviewed studies as well as a quantitative method of combin ing and analyzing the data examining the effects of stress on ART treatm ent outcomes is nonexistent. Research Purpose and Questions Although sometimes viewed as a so cial condition of childlessness, infertility is a significant health problem The role of stress on infertility and treatment outcomes is complex, oft en leading researchers to conflicting conclusions. The purpose of this study is to investigate the impact of stress on the success of Assisted Reproductive Te chnology (ART) treatments and to determine whether psychoeducational interv entions mitigate the impact of stress during Assisted Reproductive Technology (A RT) treatments. The study will test two hypotheses. 1. Increased levels of stress will reduce the likelihood of Assisted Reproductive Technology (ART) treatment success, and

PAGE 32

17 2. Psychoeducational interventions provided to patients receiving infertility treatment will mitigate the effects of stress during Assisted Reproductive Technology (ART) treatment. Limitations of the Study Limitations to any meta-analysis in clude methodological variability and bias. With respect to methodological variability, the extraction and pooling of data brings about statistical concerns of heterogeneity relating to the problem of variability among studies and the appropria teness of combining them into one meta-analysis (Hughes, 1992; Peipert & Bracken, 1997). The possibility for bias in a meta-anal ysis exists in several forms. First, the potential for publication bias, that is a bias toward published studies that have demonstrated “positive” results, re mains a concern with all meta-analytic studies. The omission of negative data, which may arise more frequently with unpublished trial studies, could lead to erroneously enhanced treatment effects and outcomes. Conversely, the inclusi on of unpublished studies may lead to the inclusion of studies of poor er quality since they have not passed a peer review process. In addition to publication bias is the potential for selection bias, which may occur when the researcher is select ing studies for inclusion in the metaanalysis. A researcher conducting a meta-a nalysis may unwittingly be partial to the selection of certain studi es over others for inclusion in the analysis. Once the studies have been selected, the potential fo r bias in the quality assessment of these studies also exists (Hughes, 1992; Peipert & Bracken, 1997).

PAGE 33

18 Professional Signific ance of the Study The relationship between stress and in fertility has long been an issue of interest and debate among researchers and ph ysicians. Several aspects of the relationship between stress and infertility or infertility treatment outcomes have been examined by researchers through i ndependent trial studies. Recognizing the emotional aspects of infertility, many physicians treating couples for infertility now also offer their patients psychoeducationa l interventions. Over recent years, investigation into the effects of stress on ART treatment outcomes has expanded. An extension of this res earch includes studies focusing on the mitigating effects that psychoeducational interventions have on elevated stress levels experienced by couples during ART. However, results of these studies are conflicting, leading to a m uddled picture of the relati onship between stress and treatment outcomes and the mitigating impact of these psychoeducational interventions. This study will combine t he results of many studies, providing a comprehensive synthesis of existing resear ch on the topic of the effects of stress on ART treatment outcomes and the impac t of psychoeducational interventions on ART treatment outcomes. To date, this is the only synthesis of current studies delineating the role of stre ss on ART treatment outcomes and the efficacy of psychoeducational interventi ons provided to infertility patients experiencing stress. This study has the potential to resolve uncertainty and controversy that exists with respect to th is topic in the literature and to provide physicians and others with evidence-based research to guide their treatment practices with infertile couples.

PAGE 34

19 Definitions of Terms Anxiety. An ambiguous emotional reaction of a culmination of fear resulting from a threat to an individual’s well-being in which the indivi dual does not know what is going to happen, when it is going to happen, and therefore, does not know what is to be done about it. Artificial Insemination (AI) The injection of semen in to the vagina by means of a syringe rather than by coitus. Assisted Reproduction Technology (ART). Any fertility procedure in which both oocytes and sperm are handled outside the body. Asthenospermia. Poor motility or movement of the sperm in a man’s ejaculate. Azoospermia. The complete absence of sperm in a man’s ejaculate. Congenital. A condition present at birth. Contraceptively Sterile. Represents women who underw ent tubal operations or women married to men who underwent vasectomies and other sterilizing operations in order to prevent pregnancy. Distress. Thought of as “bad stress” which is an inherently unpleasant emotional experience such as frustration and re sentment associated with unpleasant outcomes and destructive to health. Effect Size Displays the magnitude of effect s in terms of standard deviation units and allows comparison across different clients or studies and is calculated as the difference between the means of 2 groups divided by the standard deviation of the control group. Embryo. Fertilized ovum and sperm that have begun cellular division.

PAGE 35

20 Endocrinology. The branch of biology dea ling with the endocrine glands and their secretions especially in their re lation to their processes or functions. Epidemic. A rapid spread or increase in the occurrence and prevalence of a condition or disease. Etiology. The cause or origin of a disease or condition. Fertility. The ability for a man and a woman to reproduce. Fecund. Producing or capable of producing children. Fecundability. The monthly probability of conception without the use of contraception. Follicle. One of the small ovarian sacs containing an immature ovum. Follicular Phase. A stage in the menstrual cycle that begins with the onset of menstruation and ends with ovulation. Gamete. Raw ovum and sperm. Gamete Intrafallopian Transfer (GIFT). A specialized technique by which a woman’s mature oocyte and her partner’s washed sperm are mixed together in a syringe and inserted via laparoscopy into the woman’s fallopian tube. Idiopathic Infertility. Infertility of unexplained etiology. Impaired Fecundity. Any woman of childbearing age who reported whether it is difficult, impossible, or dangerous to become pregnant or carry a pregnancy to term. Intracytoplasmic Sperm Injection (ICSI) A form of micromanipulation involving the injection of a single sperm directly into the cytoplasm of a mature oocyte

PAGE 36

21 using a glass needle (pipette), with the re sulting embryo later implanted in the uterus for gestation. Intrauterine Insemination (IUI). The injection of semen into the uterine by means of a syringe and catheter rather than by coitus. In Vitro Fertilization (IVF). A specialized technique by which an ovum is fertilized by sperm outside of the body, with the re sulting embryo later implanted in the uterus for gestation. Infecund. Sterility; when the fecundabi lity of a couple is zero. Infertility. The state in which a couple desiri ng to have a child cannot conceive after 12 months of unprotected intercourse. Luteal Phase. Usually referred to as “days past ovulation” (DPO) and refers to the part of a woman’s menstrual cycle that begins at ovulation and ends the day before menstruation. Meta-Analysis. A statistical process of combining the results of many quantitative studies for an overall synthesis. Neuroendocrinology. The study of the anatomical and physiological interactions between the nervous and endocrine system. Noncontraceptively Sterile. Represents those women who had surgery to correct medical problems with their reproductive organs, such as hysterectomies for fibroid tumors or endometriosis. Oligospermia. Low concentrations of sperm present in a man’s ejaculate. Oocyte. An immature egg cell contained in the follicle. Parity. The condition or fact of having born offspring.

PAGE 37

22 Primary Infertility. Women who have never achieved pregnancy. Psychoeducational Interventions. Any therapeutic techni que, which includes educational training, aimed at the reducti on of distress experienced by patients participating in m edical treatment. Secondary Infertility. Women who have achiev ed pregnancy at least once before, regardless of the outcome, and who cannot achieve a subsequent pregnancy. Sterility. The absolute inability to reproduce. Stress A response to any noxious agent in an individual’s environment. Three conceptualizations of stress include the response-oriented approach, the stimulus-oriented approach, and the interaction approach. Stressors. External events or cond itions affecting an organism. Subfecund. Infertile. Teratospermia. An increased percentage of abnormally shaped sperm in a man’s ejaculate. Zygote. Fertilized ovum and sperm prior to cellular division. Zygote Intrafallopian Transfer (ZIFT). A specialized technique by which a woman’s mature oocyte is fertilized by her partner’s washed sperm outside of the body, with the resulting zygote being insert ed via laparoscopy into the woman’s fallopian tube.

PAGE 38

24 CHAPTER TWO: LITERATURE REVIEW Introduction Much public attention has been given to infertility and infertility treatment, leading many to believe that infertility has become an epidemic. While estimates of infertility and trends demons trating an increase in number of couples seeking treatment for infertility have risen over the years, the evidence does not support the claims that infertility has become an epi demic. However, the incidence rate of infertility does support the conclusion that infertility is an important societal health issue. Therefore, infertility tr eatment and research in this field are important to the further advancement of this specialized field. One longstanding popular topic in the field of infertility has been the relationship between stress and infertility. Shaped by cultural, social, and personal expectations, infertility can cr eate overwhelming stress. Research literature provides a plethora of info rmation documenting the psychological impacts of infertility. While early investi gations focused on attributing idiopathic infertility to psychological maladjustment, current research focuses largely on the impact of stress on ART treatment outcomes. In addition, investigations into the efficacy of psychoeducational interventi ons and their impact on the effects of stress on ART treatment outcomes have begun. However, investigations into the relationship between stress and infertility vary considerably, leading many to conflicting conclusions. One source of the controversie s may lie in the ambiguity

PAGE 39

25 of the definition of stress. Since the begi nnings of stress research, a variety of theoretical conceptualizations of stress have evolved leading to a multitude of definitions and measures of stress. In addition, many researchers use the terms “stress” and “anxiety” interchangeably, leadi ng to an even more unclear definition of stress. Three classical conceptua lizations of stress include the responseoriented approach, the stimul us-oriented approach, and the interaction approach. While stress is conceptualized in broad, ambiguous terms to include any noxious agent in an individual’s environment, anxie ty is regarded by many researchers and theorists as an emotional reaction to a perceived threat to an individual’s well being and may represent only one of many emotional responses to a stressful situation or event. Although investigati ons into the relationship between stress and infertility span a multitude of topi cs, a common theme examined is the distress experienced by individuals and coup les seeking infertility treatment. To understand the complex and o ften ambiguous relationships between stress and infertility, one needs to grasp the scope of the issues surrounding infertility as well as the prevalence of infertility, the conceptualizations and theoretical approaches to stress, and evidence of the relationships between stress and infertility provided through research. The Infertility Problem Definition, Prevalence and Trends of Infertility in the United States Fertility is the ability for a man and a woman to reproduce. The epidemiological term, fecundability, refe rs to the monthly probability of conception without the use of any contraception (Jansen, 1993; Speroff, et al,

PAGE 40

26 1994; Tsaltas, 1997). A couple is said to be subfecund, or infe rtile, when there is an involuntarily long interval until their first conception or between births. A common definition of infertility is the st ate in which a couple desiring to have a child cannot conceive after 12 months of unprotected intercourse (Tsaltas, 1997). Failure to conceive after 12 months of unprotected intercourse is taken to be abnormal as 90% of couples will have conc eived within that time (Tietze, 1956, 1968; Tsaltas, 1997). Broadly defined, infertility is a dis ease of the reproduc tive system that impairs conception and the ability to carry a pregnancy to full term leading to live birth (ASRM, 2000-2003; National Women’s Health Information Center [NWHIC], 2003). Infertility is either classified as primary or secondary. Primary infertility refers to women who have never ac hieved pregnancy, whereas, secondary infertility refers to those women w ho have achieved pregnancy at least once before, regardless of the outcome, and who cannot achieve a subsequent pregnancy (Thonneau et al., 1991; Seibel, 1993; Tsaltas, 1997). Sterility, on the other hand, is the absolute inability to r eproduce. When the fecundability of a couple is zero, for whatever reason, the couple is defined as infecund, or sterile (Jansen, 1993; Seibel 1993; Tsaltas, 1997) Although the true incident rate of sterility is unknown, it is believed that 3-5% of the population is sterile (Spira, 1986; Jansen, 1993, Tsaltas, 1997). The incidence rate of infertility in less industrialized nations is markedly higher and reflects a greater proportion of infertility probl ems attributed to infectious diseases than those in t he United States and other industrialized

PAGE 41

27 nations. Over recent year s, a number of factors hav e contributed to the public’s perception that there is an infertility epidemic in t he United States. This perception could be tied to the “medica lization” of infert ility and treatment. Medical advancements in the underst anding of human endocrinology and medical technology have significantly impacted the development of this specialized field. New medication s and technologies have been noteworthy advancements offering childless couples hope to conceive. Also, the supply of physicians trained to provide specialized infertility services to couples has dramatically increased over the past 20 years. In addition to these medical and technological advancements, several social factors have influenced the development of the medical industry in in fertility. The first factor includes the sexual revolution and the increased incidenc e of sexually transmitted diseases. Secondly, there is an increase in wom en entering the job market and delaying childbearing until ages when it is considerably more difficult to conceive. In addition, the aging of baby boomers has increased the absolute numbers of couples trying to have children. Finally, in addition to the increased number of women trying to conceive and the increase in the number of women delaying trying to conceive until they are older, the number of healthy infants, especially white infants, available for adoption has decreased. The development of infertility diagnosis and treatment as a m edical industry along with intense media attention given to couples seeking help from physicians for infertility has provoked the perception that infertility has become an epidemic (Mosher & Pratt, 1991; Whitford & Gonzalez, 1995; Chandra, 2003).

PAGE 42

28 Estimates regarding the actual in cidence of infertility, primary and secondary, vary greatly, ranging from 8% to 33%, depending on th e criteria used to define infertility and the population included (Seibel, 1993; Schneider, 2000). The National Survey of Family Growth (NSFG) and its predecessor surveys, the Growth of America Study in 1955 and 1960 and the National Fertility Survey in 1965 and 1970, have been the only source pr oviding reliable national estimates regarding the actual incidence of infertilit y, both primary and se condary, in the United States. To date, five NSFGs have been conducted by the National Center for Health Statistics – 1973, 1976, 1982, 1988, and 1995. The primary purpose of the 1973-1995 surveys was to provi de reliable national data on marriage, divorce, contraception, infertility, and the health of women and infants in the United States as well as to provide info rmation on factors that affect the nation’s birth rate. The last NSFG survey, c onducted in 1995 by the NCHS, contained an enhanced set of infertility questions t hat covered the respondent’s pregnancy history, past and current use of contracept ion, the ability to bear children, the use of medical services for family planning, in fertility, prenatal care, marital history, and associated cohabiting unions. The Survey Research Center of the University of Michigan is presently working on the 2002 NSFG survey and the findings are projected to be released in early 2004. The 2002 NSFG includes interviews and free-response items onli ne with both men and women. Survey participants are being asked questions regar ding their schooling, work, marriage and divorce, contraceptive use, infertilit y, parenting, and related medical care. The 2002 NSFG is distinguished from previous studies in that it captures, for the

PAGE 43

29 first time, reliable data for men, providing essential in formation regarding national infertility prevalence ( NCHS, 2001; Chandra, 2003). The NSFG produces data that classi fies women or couples into three major groups: surgically sterile, impaired fecundity, or fecund. In addition to these broad categories, measures of infe rtility are included. The category for surgically sterile is reported as either “contraceptively sterile” or “noncontraceptively sterile”. The subcategory of contraceptively sterile represents women wh o underwent tubal operations or women married to men who underwent vasectomies or other ster ilizing operations in order to prevent pregnancy. Noncontraceptively sterile represents those women who had surgery to correct medical problems with t heir reproductive organs, such as hysterectomies for fibroid tumors or endometriosis. Impaired fecundity included any woman of childbearing age who reported t hat it is difficult, impossible, or dangerous to become pregnant or carry a pr egnancy to term. Impaired fecundity includes all women regardless of marital st atus. Infertility st atus in the NSFG refers to married couples of childbear ing age that have not been surgically sterilized, have not used contraception, and have not become pregnant for at least 12 months. For unmarried women, impaired fecundity refers only to the woman herself whereas, for married wo men, the questions capture information about female and male infertility or impai red fecundity (Chandra, 2003; Mosher & Pratt, 1991). The most recent report of NSFG findings from the 1995 administration estimated that 7.1% of married couples, or 2.1 million, in which wives were aged

PAGE 44

30 15 to 44 years were infertile. Among all women of reproductive age, 24% were surgically sterile. An additional 10.2% (6.1 million) had impaired fecundity (CDC, 1997). The ratio of patients presenti ng with primary and secondary infertility represents 67% to 71% of patients cat egorized as presenting primary infertility and 29% to 33% presenting with secondary in fertility (Hull et al., 1985; Thonneau et al., 1991; Tsaltas, 1997; Templeton et al., 1999). Table 1 below presents historical fi ndings from the NSFG regarding the fertility of women of childb earing age. As shown in Table 1, in 1995, 24 percent of all women ages 15-44 (or their hus bands, if they were married) were contraceptively sterile, including the th ree percent of wom en with no children and 40 percent of women with one birth or more, reflecting an increase when compared to 1982. Another seven per cent of women had been surgically sterilized for noncontraceptive reasons, in cluding one percent of women with no children and 11 percent of women with one bi rth or more, reflecting a decrease in the percentage of women as compared to 1982. In addition, approximately 6.1 million women (or 10.2 perc ent) had impaired fecundity reflecting an increase when compared to 1982. Of the 6.1 million women with impaired fecundity, 2.8 million had no children and 3.3 million had on e or more children, reflecting an increase since 1982 with 1.9 million and 2.6 m illion, respectively (Mosher & Pratt, 1990; CDC, 1997; Chandra, 2003).

PAGE 45

31 Table 1 Percent distribution of wo men 15-44 years of age by fecundity status, according to parity and age: United States, 1982 to 1995. Surgically Sterile Contraceptive Noncontraceptive Impaired Fecundity Fecund Age in Years 1982 1988 1995 1982 1988 1995 1982 1988 1995 1982 1988 1995 Parity = 0 15-44 1.7 2.8 2.8 1.4 1.5 1.5 8.4 8. 8 11.0 88.5 86.9 84.7 15-24 0.1 0.2 0.2 0.0 0.0 0.1 4.1 4. 1 5.5 95.8 95.7 94.3 25-34 3.3 3.1 2.9 1.8 1.6 0.7 14.7 13. 4 13.9 80.2 82.0 82.5 35-44 10.3 15.8 11.9 12.7 9.2 8.1 25.7 21. 4 25.7 51.3 53.6 54.3 Parity = 1 or More 15-44 31.2 39.0 39.7 10.5 7.1 4.2 8.5 8. 1 9.6 49.9 45.8 46.5 15-24 9.0 9.8 6.7 0.6 0.7 0.3 5.2 7. 7 8.4 85.2 81.8 84.6 25-34 28.1 32.8 32.1 6.1 3.3 1.5 8.1 7. 8 9.8 57.8 56.1 56.7 35-44 42.7 52.3 52.9 19.0 12.5 7.2 10.1 8. 5 9.8 28.1 26.7 30.1 Parity = All 15-44 18.6 23.3 24.2 6.6 4.7 3.1 8.4 8. 4 10.2 66.3 63.6 62.5 15-24 2.1 2.0 1.6 0.2 0.2 0.1 4.3 4. 8 6.1 93.4 93.0 92.2 25-34 21.0 22.9 22.0 4.9 2.7 1.2 10.0 9. 6 11.2 64.2 64.7 65.6 35-44 38.7 46.3 45.3 18.3 12.0 7.4 12.1 10. 6 12.8 31.0 31.0 34.6 Note. Parity is defined as the number of children born. Adapted from “Fertility, Family Planning, and Women’s Health: New Data from the 1995 National Survey of Family Growth”, by Centers for Disease Control and Pr evention, 1997, p. 59; “Fecundicity and Infertility in the United States, 1965 – 88”, by Willia m D. Mosher & William F. Pratt, Advanced Data, 192, December 4, 1990, p. 4. Although the percent of women wit h impaired fecundity increased 1.8 percentage points, from 8. 4 percent in 1982 and 1995 to 10.2 percent in 1995, no evidence supports the perception that in fertility has become an “epidemic”. Further analyses investigating a variety of potential reasons explaining these findings are needed. One plausible expl anation is that the changes observed since 1982 are merely an artifact of the aging baby boom generation. Other

PAGE 46

32 plausible explanations of th is trend include: women ar e waiting longer to begin trying to start their families; the 1. 6 percentage point drop in the percent categorized as surgically sterile fo r noncontraceptive reasons; the many new medications and treatments now available to couples; as well as an increase in the number of physicians specializing in this field and couples seeking advice and treatment (Mosher & Pratt, 1991; CDC 1997; Chandra, 2003). While earlier studies have projected t he number of women in the United States with impaired fecundity to range from 5.1 million in 1995 to 4.8 – 5.9 million in 2020, a more recent study has provided projection data while considering several trends evidenc ed in the 1995 NSFG. Given population projections provided by the U.S. Bur eau of the Census for the years 2000 to 2025, Stephen and Chandra (1998) estimate that nearly 6.5 million women can be expected to be infertile in the year 2025. While this estimate cannot possibly anticipate factors such as the emergenc e of highly prevalent new disease processes related to infectious agents, environmental chemicals, or other unpredictable events, it highl ights the importance of continuing research in this field (Stephen, 1996; Phipps, 1996; Stephen & Chandra, 1998; Schneider, 2000). Physical Causes of Infertility The diagnosis and treatment of infert ility contrasts with almost all other medical conditions in that it involves a couple. When a couple presents concerns of infertility to their phy sician, thorough assessment of both individuals is essential. Approximately 40 percent of the causes of explained infertility can be established as male factors such as abnormal spermatogenesis, abnormal

PAGE 47

33 motility of the spermatozoa, anatomic di sorders, endocrine disorders, and sexual dysfunction. Female factors such as cervical, uterine or tubal, ovulatory, peritoneal and pelvic disorders account fo r approximately another 40 percent of explained infertility cases. In approximat ely 20 percent of the cases, explained infertility problems can be attributed to a combination of both male and female factors. In addition to a ttributing infertility to female, male, or a combination of both female and male factors, the etiology of infertilit y is sometimes defined as unexplained infertility. Estimates regarding the prevalence of idiopathic infertility vary considerably due to a number of r easons. The proportion of cases of infertility that are unexplained range from 0 to 31 percent. Most studies between 1950 and 1995 found an average rate between 15 and 25 percent. Approximately 2 to 15 percent of infert ility cases are categorized as unexplained or idiopathic infertility where no di agnosis can be made following a thorough investigation (Seibel, 1 993; Schneider, 2000). Conception is an intricate and comp lex neuroendocrinologic process. Any one of a number of physical factors may in terfere in this process and contribute to infertility. Approximately 30 to 40 percent of infertilit y problems among women are due to peritoneal factor s. The two most comm on peritoneal factors are endometriosis and tubal disease. Failure to ovulate is the major problem in approximately 25% (Seibel, 1993) of in fertility diagnoses. Although ovulation may be affected by several sources, the three most common causes are extreme emotional distress, excessi ve weight loss or gain, and excessive exercise. Cervical factors are identified in no more than 5 to 10% (Seibel, 1993) of infertility

PAGE 48

34 cases; these disorders may be either congen ital or acquired. Regardless of their etiology, these disorders can signific antly impact the proc ess of conception through both the receptivity of the cervical mucus and the ability of sperm to reach and survive in the mucus. Approx imately 5% (Seibel, 1993) of infertility cases are caused by uterine factors. Uterine factors are commonly structural abnormalities and are associated with feta l wastage and increased frequency of obstetric problems (Seibel, 1993, 1997; Pernoll, 2001; Schneider, 2000). Factors contributing to infertility am ong men can be attributed to several types of disorders. Abnormal sperma togenesis, including low semen volume, high sperm viscosity, and low sperm motilit y may occur as the result of mumps orchitis during childhood, chromosomal abnormalities, cryptorchidism, chemical or radiation exposure, variocele, test icular failure, obstruction, or other anatomical disorders. In addition, endocri ne disorders may also contribute to male-factor infertility (Seibel, 1993; Schneider, 2000; Pernoll, 2001). Before 1900, virtually all cases of infertility were classified as unexplained due to the lack of clinical tests availa ble to diagnose the etiology. Although innovations between 1900 and 1940 have led to considerable improvement in the diagnosis of tubal, seminal, and ovulatory pr oblems, the etiology of all cases of infertility cannot be diagnosed. Infert ility is categorized as unexplained or idiopathic when an extremely long delay in conception occurs by chance in otherwise healthy couples or because under lying defects cannot be detected with current clinical diagnostic tests or tools. Also contributing to idiopathic infertility is the declining fertility of older fe male partners. It is impo rtant to recognize that in

PAGE 49

35 any normal population, otherwise healthy couples may appear to be infertile by chance alone because a small proportion of healthy couples have low fecundity, and therefore may not conceive within a year (Seibel, 1993; Collins, 1997; Schneider, 2000). A recent study conduct ed by David Dunson suggests that among outwardly healthy couples with no known conditions associated with infertility who failed to conceive naturally within the first year, most will conceive naturally in the second year. Dunson r eported that 97% of women ages 19 to 26, 94% of women ages 27 to 34 years, and 91% of women ages 35 to 39 years conceived naturally within the second y ear, provided that the male partner was under the age of 40. If the male partner was over the age of 40, only 84% of women ages 35 to 39 years conceived naturally within the second year. Therefore, Dunson asserts t hat couples who might othe rwise be categorized with idiopathic infertility should be patient and that physicians should not intervene too quickly with assisted reproductive techni ques unless there are known reasons for a couple not conceiving naturally within a year (Hawkins, 2002). In addition to identified physical causes of infertility, other demographic factors contributing to impaired fecundi ty include advancing age and smoking. The decline of fecundity among married couples with advancing age has been documented throughout the research literat ure. Approximat ely one-third of women who defer pregnancy until their mi d to late 30s will have an infertility problem. At least half of the women wait ing to conceive until over the age of 40 will have an infertility problem. Although the risk of infertility related to increasing age has primarily been focused on women, the changes in male fertility with

PAGE 50

36 aging, although modest, are significant. Ad vancing age in men is significant for two major reasons. First, the quality of sperm decreases with age and is evidenced by the absolute frequency of autosomal dominant disease representing 0.3% to 0.5% among offspri ng in fathers 40 y ears of age and older (Seibel, 1993). In addition, men with advancing age also demonstrate diminished fecundity, evidenced through the rates of conception. The rate at which men over the age of 40 impregnate t heir partner within 6 months is onethird (Seibel, 1993) of that of m en under the age of 25 which may be due to factors such as involution of testicular function, decreased sperm production, and maturation arrest of sp ermatogenesis, all of which have been associated with advancing age (Collins & Rowe, 1989; Seibel, 1993; Speroff et al., 1994). Another contributing factor to decreas ed fecundity is smoking. While the hazardous effects of smoking on repr oduction have been widely discussed, epidemiological studies have documented that fecundity decreases directly with the number of cigarettes smoked. Fec undity decreases 25% (Seibel, 1993) in women who smoke up to 20 cigarettes per day in comparison to nonsmokers. Among those women who smoke more than 20 cigarettes a day, women demonstrate a 57% (Seibel, 1993) decreas e in fecundity in comparison to nonsmokers. In addition, ectopic ges tations are far more frequent among smokers than among nonsmokers. Among men, cigarette smoking reduces sperm density by 22% on average. While re sults of studies examining the effects of smoking on human sperm morphology and motility are inconsistent, adverse byproducts of smoking are evidenced in men through testicular atrophy, blocking

PAGE 51

37 spermatogenesis, and in experimental animals through altering sperm morphology. In addition to these effect s, male smokers who have a testicular varicocele are 10 times more likely to have oligospermia than nonsmokers who do not have a varicocele. Based on this evidence, couples with unexplained infertility or in whom fertility is marg inal, cessation of smoking may play a substantive role in increasing fecundity (B aird & Wilcox, 1985; Howe et al., 1985;; Phipps et al., 1987; Seibel, 1993). Physical Infertility Treatment Options Since the 1950s, the treat ment of infertility has evolved dramatically. During the 1960s, safe and effective ovulation-inducing medications were developed and introduced. Re finements in surgical techniques and technology during the 1970s paved the way for the dev elopment of microsurgery for tubal disease. Further developments in cell cu lture and embryology during the 1980s produced in vitro fertilization and related assisted reproductive technologies. Finally, during the 1990s, the tr eatment of infertility was fu rther refined to include micromanipulation of cells through the development of intracytoplasmic sperm injection (ICSI) and other procedures (Seibel, 1993; ASRM, 2001). It appears that with advancing technology and an incr eased number of specialists in the field, the diagnosis and treatment of infe rtility will progress to a sub-cellular level and molecular biology will move to the forefront. Hormonal Treatment Two primary fertility hormones contro l follicle development and ovulation in women and sperm development in m en. If the body does not produce these

PAGE 52

38 hormones in exactly the right way, a hormone imbalance or deficiency may cause infertility. In women, treat ment with gonadotropins can increase the likelihood of conception by st imulating inactive ovaries to ovulate or to produce more than one egg at a time. In m en, gonadotropins can stimulate the production of sperm. However, fewer t han 5 percent of infertile men have a hormonal disorder that can be treat ed with hormonal therapy (Seibel, 1993, 1997; Speroff et al., 1994; Lunenfe ld et al., 1997; Pernoll, 2001). By far, the most prevalent medication for ovulation induction is clomiphene citrate. Clomiphene citrat e stimulates ovulation for women with irregular menstrual cycles or stimulates the dev elopment of multip le eggs during a woman’s menstrual cycle. Clomiphene citrat e is an anti-estrogen that acts on the brain to stimulate the ovaries. It is an oral medication taken during the early part of a woman’s menstrual cycle which blocks estrogen receptors in the hypothalamus, causing the hypothalamus to signal the pituitary gland to release more follicle stimulating hormone (FSH) and luteinizing hormone (LH) into the bloodstream. These increased levels of FSH lead to the development of the follicle and egg, which, in turn, secret es more estrogen into the bloodstream. Approximately one week after ingestion of the last clomiphene citrate tablet, the hypothalamus receptors are no longer bl ocked, triggering an LH surge in response to the artificially elevated leve ls of estrogen in the bloodstream. At the proper dosage level, ovulation usually occurs 7 to 10 days after the last tablet is ingested. Serious side effects associated with this medication are rare. Multiple births, the most of which are twins, occu r in less than 10% of the cases. Other

PAGE 53

39 side effects, including but not limited to hot flashes, breast tenderness, mood swings, visual and gastrointestinal sym ptoms, are not uncommon and completely reversible upon cessation of the medicati on. A rather large percentage of couples report that the use of clomiphene ci trate results in some emotional side effects. The emotional in stability often experienced by patients using clomiphene citrate is described by one patient: “Hormone hell is probably the best wa y to describe it. By third or fourth day of taking it, I would become hysterical for absolutely no reason. I felt as if I had no c ontrol over my body. I began to wonder if this was worth it. But, it only lasted a few days, and we all got through it. But it wasn’t f un” (Seibel, 1993, 1997; Speroff et al., 1994; Chamoun et al., 1997; Pernoll, 2001). In addition to clomiphene citrate, two ma in types of medications are used to induce ovulation: Gonadotropi ns containing follicle st imulating hormones (FSH) and Human Menopausal Gonadotr opins (hMG). When given to premenopausal women, these medications stimulate the ovaries to form follicles that mature and produce eggs. In clinical use for 30 years, Human Menopausal Gonadotropin (hMG) is distributed as a lyophilized pow der containing a luteinizing hormone (LH) and follicle-stimulating hormone (FSH ). These hormones stimulate the woman’s ovaries to produce more follicles, thus increasing the number of eggs available for fertilization. The FSH is primarily responsible for follicular recruitment, selection, growth, and ripeni ng of a woman’s eggs. The LH part is

PAGE 54

40 responsible for the final maturation of t he FSH-stimulated follicles, ovulation, and transformation of the follicular remnants into functional corpora lutea. Unlike clomiphene citrate treatm ent, gonadotropins act directly on the ovaries and are often prescribed to stimul ate the development of mu ltiple eggs. Gonadotropin treatment requires a series of injecti ons and careful monitoring of follicular development through trans-vaginal ultr asound and serum estradiol levels throughout the cycle. If satisfactory blood estradiol levels and follicle development occur, ovulation is then induc ed with an intramuscular (IM) injection of human chorionic gonadotropi n (hCG). In less than 2% of the cases, severe overstimulation occurs and the cycle must be cancelled. Multiple pregnancies occur in approximately 26 to 40% of cycles 75% of which are twins while 25% is triplets or more. The most common r eason for using hMG is anovulation. However, it is also used to treat cases of infertility caused by oligoovulation, luteal phase deficiency, idiopathic infert ility, and in the harve sting of multiple follicles for in vitro fertilizat ion, gamete intrafallopian transfer, zygote intrafallopian transfer, and intracyplasmic sperm injecti on (Seibel, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Follicle Stimulating Hormone (FSH) bec ame available for clinical use in the United States in 1986. FSH is a furt her purification of hM G and is primarily indicated to treat clomiphene-resistant patients with polycystic ovary disease (POCS). As with the hMG, FSH stimulates the ovaries to produce more follicles, thus increasing the number of eggs. T hese medications are initiated between days 4 and 6 of a woman’s menstrual cycle and are administered with a

PAGE 55

41 subcutaneous (SQ) injecti on. Ovulation may occur spontaneously but is much more consistent if human chorionic gonadot ropin (hCG) is administered. As with hMG, ovulation induction with FSH must be carefully monitored through measurements of serum estradiol leve ls and follicular development through pelvic ultrasound to reduce the potential for ovarian overstimulat ion and multiple births (Seibel, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Gonadotropin-Releasing Ho rmone (GnRH) is responsible for the release of LH and FSH from the pituitary gland. GnRH regulates the reproductive cycle in both sexes. Pulsatile secretion of GnRH from the hypothalamus is key in establishing normal gonadal function. Failure of this release results in isolated GnRH deficiency that can be distinguished by partial or complete lack of GnRHinduced LH pulse, normalization with Gn RH replacement, and otherwise normal hypothalamic-pituitary neuroanatomy and n europhysiology. This medication can be administered subcutaneously or intravenously. Side effects of this medication include ovarian overstimulat ion, multiple births, infection at the indwelling catheter site, allergic reactions, and rarely anaphylaxis with the development of anti-GnRH antibodies (Seibe l, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). GnRH agonists are prescribed for nearly all patients during gonadotropin therapy for IVF or GIFT to prevent prem ature ovulation. Administration of GnRH agonists can be started either in the lu teal phase around day 21 of a woman’s menstrual cycle or in the early follicular phase just after the menstrual period has begun. GnRH agonists such as Lupron and Sy narel are synthetic imitators of GnRH. However, when treated with GnRH, the pituitary initially increases its

PAGE 56

42 production of FSH and LH, but then stops FSH and LH production due to “down regulation”. Therefore, GnRH agonist s serve to suppress the ovaries and provide the physician with greater cont rol over controlled ovarian hyperstimulation and in preventing premature ovulation (Seibe l, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Prolactin is a hormone produced by the pituitary gland and is usually elevated in women during pr egnancy and breast-feeding to promote lactation. When prolactin is elevated in non-pregnant and non-lactating women, the result can be irregular menstrual cycles or an inadequate luteal phase. Bromocriptine is a medication designed to lower the le vels of prolactin in the bloodstream. Bromocriptine is prescribed in cases of el evated prolactin levels and results in 85 to 90% ovulatory success rates when other infertility factors are not present (Seibel, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Another hormonal imbalance that is present in a few women is characterized by excess of amounts of androgens, or male type hormones. These increased levels of androgens, such as testosterone and androstenedione, may interfere with pr ocesses such as normal follicular development and ovulation. In these ca ses, low doses of corticosteroids are used to lower the androgen levels to within normal range (Seibel, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Surgical Treatment Advancement in surgical techniq ue is one the many developments in reproductive endocrinology over the past several decades. Magnification and

PAGE 57

43 microsurgery have significantly cont ributed to the advancements made in infertility surgery. Adv ancements in magnification include loupes and later operating microscopes while micr osurgery has been advanced through developing technologies such as intricat e surgical tools and techniques (Seibel, 1993). For women, surgical options are available for diagnostic as well as therapeutic treatment. For diagnostic purposes, the hyst erosalpingogram (HSG), laparoscopy, and hysteroscopy are commo n procedures. These procedures are useful diagnostic procedures for identif ying tubal disease and obstruction, and conditions such as endometriosis and fibr oid tumors that may interfere with a woman’s ability to achieve a successful pregnancy. Any pelvic conditions that may inhibit a woman’s ability to conceive may also be corrected at the time the laparoscopy and hysteroscopy are performe d (Seibel, 1993, 1997; Speroff et al., 1994; Pernoll, 2001). Among men, surgical techniques ar e performed for conditions such as varicocele or to remove any obstruction in the sperms’ path. Blockages can be found in the vas deferens, epididymis, or in the ejaculatory duc t. In men who are diagnosed with non-obstructive azoospermi a, a testis biopsy may be performed to identify the predominant pattern of the testicular histology. Surgical procedures such as Testicular Fine Needle Aspiration (TFNA), Percutaneous Epididymal Sperm Aspiration (PESA), Microsurgical Epididymal Sperm Aspiration (MESA), Testicular Sperm Extraction (TESE), and Microdisection Testicular Sperm Extraction (TESA) offer infertile men who in previous years were unable to produce a biological child the possibility of fathering biological

PAGE 58

44 children. Among men with a blockage, in nearly 100% of cases, surgical procedures for sperm extraction are succe ssful. In approximately 40% of nonobstructive azoospermia cases, a few sperm cells are identified during the Testicular Sperm Extraction (TESE). In some cases, the testicle may produce only a few sperm. However, even if only a few sperm are found, they can be used in ART procedures such as ICSI. In male infertility patients with tubular sclerosis, maturation arrest, or Sertoli ce ll-only syndrome, mature spermatids or testicular spermatozoa can be recovered in approximately 50% of the cases. However, new innovations and research in micromanipulation such as with Round Spermatid Nuclei Isolated (ROSNI) are providing these men with prospects to father a biological child in the future (Speroff et al., 1994; Seibel, 1993, 1997; Sofikitis et al., 1995; Schlegel & Girardi, 1997; Johnson et al., 1999; Pernoll, 2001; Goldstein et al., 2003;). Artificial Insemination John Hunter performed the first artifi cial insemination with the husband’s sperm intravaginally in E ngland during the late 18th century. His nephew reported a normal pregnancy and de livery as a result of this procedure. As early as the beginning of the 1950s, reports of t he use of AI with the husband’s sperm were published. Artificial Insemina tion is a rapidly advancing science and a relatively simple procedure in which specially treated sperm from the male partner are injected into the female’s repr oductive tract. The indications for AI have been classified into five categories:

PAGE 59

45 1. Mechanical problems in the male such as impotence, hypospadias, premature or retr ograde ejaculation, 2. Mechanical problems in the female such as vaginismus or prolapse, 3. Cervical “hostility”, 4. Poor semen quality in terms of either volume, concentration, motility, morphology, or the presence of antibodies, and 5. Idiopathic subfertility (Nuoj ua-Huttunen et al., 1995, 1997, 1999; Kovacs & Vollenhoven, 1997). While AI is generally performed with the husband’s sperm for most couples, using donor sperm is considered a tr eatment option for couples when the husband’s ejaculate contains few or no live sperm or when he is genetically or anatomically unable to produce any sperm. Indication for the use of therapeutic donor insemination (TDI) is the absence of sperm sufficient in quantity or quality to be likely to produce a pregnancy. While TDI is an option for these couples, new assisted reproductive techniques c an now provide alternative treatment options for many men who pr eviously had been considered irreversibly subfertile. TDI may also be a treatment option for those men who are genetically and anatomically capable of reproduction but also have a genetic or psychological reason compelling him and his partner to request donor insemination. Other cases seeking donor insemination include the woman who has no sexual partner yet desires to have a child (Helsa, 1995; Kovacs & Vollenhoven, 1997).

PAGE 60

46 Artificial insemination (AI) is a general term. Several diffe rent types of AI, which are named for the location of s perm insemination into the female, are available. One type is intracervical inse mination (ICI); in this procedure, the sperm is injected into the female’s cervical canal. ICI is an appropriate therapy to enhance fecundity in circumstances such as when intravaginal intercourse with ejaculation in their pericervical area is not possible or when the use of donor sperm is planned. Pregnancy rates for ICI treatment average 8 – 12% per treatment cycle. At least 60% of coupl es should conceive within six treatment cycles (Helsa, 1995; Nuojua-Huttunen et al., 1995, 1997, 1999; Kovacs & Vollenhoven, 1997). Intrauterine insemination (IUI), the most common form of AI used, is a procedure where the sperm are injected into the female’s uterine cavity. This method is particularly useful if the caus e of infertility has been determined as an insufficient or hostile cervical mucus or low sperm count or motility. In circumstances where cervical mucus is insufficient or hostile, IUI is advantageous because it allows sperm to by pass the cervix completely. In cases where infertility is due to low sperm count or motility, IUI is advantageous because it places the healthiest sperm in to the female tract to increase the likelihood that one of those sperm will fertilize an egg. Although IUI is relatively uncomplicated and less invasive than ART procedures, one disadvantage associated with IUI is that it does not a llow the physician to evaluate whether or not fertilization is capable of taking place. Although IUI is a more natural method of conception than ART procedures, whether or not the sperm actually fertilizes

PAGE 61

47 the egg to make an embryo is unk nown unless the woman becomes pregnant after the first IUI cycle. The pregnan cy rates for IUI treatment cycles range on average between 15 – 20% per cycle. If a ll other conditions affecting fertility are thought to be normal or adequately treat ed, then a reasonable length of treatment time with IUI is about three to six treatment cycles (Helsa, 1995; Nuojua-Huttunen et al., 1995, 1997, 1999; Kovacs & Vollenhoven, 1997). Other methods of AI include Direct Intraperitoneal Insemination (DIPI), Transuterine Intratubal or Semen Intrafallopian Insemi nation (SIFT), and Direct Intrafollicular Insemination (DIFI). M anhes and Hermabessiere initially described DIPI in 1985. This AI te chnique is a process in which sperm are introduced into the body cavity between the uterus and the re ctum around the time of ovulation. This procedure has been described as useful for certain categories of infertility such as cervical factor or male subfer tility. Transuterine Intratubal or SIFT has been made possible by the development of catheters capable of being inserted through the uterus and placed in the ampu lloisthmic region in an atraumatic fashion. This technique is a sonogr aphically guided procedure in which sperm washed free of seminal fluid are injected in to the fallopian tubes. This procedure has been suggested as a means to reduce the total number of motile sperm required for the insemination or as a means to decrease the need for superovulation in subfertile couples. In Intrafallopian insemination the sperm are injected directly into the female’s fallopian tubes. Finally, DIFI is a technique in which the sperm are injected into the fema le’s ovarian follicle under ultrasound guidance transvaginally. Note that any of the AI procedures may also be

PAGE 62

48 combined with other treatm ent options such as ovarian stimulation through hormone therapy or following surgical pr ocedures (Helsa, 1995; Nuojua-Huttunen et al., 1995, 1997, 1999; Kovacs & Vollenhoven, 1997). Assisted Reproductive Technology ART procedures include any techni que in which both the oocytes and sperm are handled or manipul ated outside of the body. ART procedures are generally the last treatment option for those couples for whom other less invasive and expensive treatments have been uns uccessful. The most commonly performed and recognized ART procedure fo r treating infertility is in-vitro fertilization (IVF). In 1976, Edwards and Steptoe first described the technique for in vitro fertilization (IVF) and embryo transfe r (ET). In 1978, t heir work resulted in the birth of two normal babies and repr esented a major milestone in infertility treatment. Their success dramatically c hanged the treatment options for infertile couples. IVF is a procedure in which fertilization occurs in vitro outside of the woman’s body in a laboratory. Th e man's sperm and the woman's egg are combined in a laboratory dish. The result ing embryo is then transferred to the woman's uterus. The five basic steps in an IVF treatment cycle are ovarian hyperstimulation, egg retrieval, fertilization, embryo cult ure, and embryo transfer. A treatment option for couples with various types of infertility, since IVF allows the doctor to perform in the laboratory w hat is not happening naturally. Initially, IVF was only used for female tubal factor infertility such as blocked, damaged, or absent fallopian tubes. Today, IVF is used to circumvent infertility caused by practically any problem, including endom etriosis; immunological problems;

PAGE 63

49 unexplained infertility; and male factor in fertility (Damewood, 1995; Laufer et al., 1997; Talbot & Lawrence, 1997). In 1984, gamete intrafallopian transfe r (GIFT), designed for women with idiopathic infertility, was first used in humans. At that time, GIFT provided a much better pregnancy rate and is cons idered a more natural method of conception than other ART procedures. GIFT is a procedure in which the female patient undergoes a controlled ovarian hyperstimulation. The oocytes are retrieved transvaginally using ultrasound gu idance. Three to 4 oocytes are then placed via laparoscopy into one of the fa llopian tubes along with the sperm. One of the disadvantages of the GIFT procedure has been t he transfer of sperm and oocytes into the fallopian tube via lapar oscopy necessitating abdominal incisions and anesthesia. Consequently, some authors have suggested the use of hysteroscopic techniques in which the gam etes are transferred into the fallopian tubes via a transcervical tubal catheteriz ation (Balmaceda et al., 1995; Dlugi et al., 1997; Wood, 1997). Modifications of the GIFT procedure in clude Zygote intrafallopian transfer (ZIFT) and tubal embryo transfer (TET). ZIFT is used primarily for couples with severe male factor infertility. In this procedure, the fema le patient undergoes a controlled ovarian hyperstim ulation and the oocytes are retrieved in a similar fashion as in GIFT. However, unlike GIFT, the oocytes are allowed to fertilize in vitro in the laboratory. At the 2-pronuclear stage, wh ich usually occurs 24 hours later, 3 to 4 embryos are then transfe rred via laparoscopy into one of the female’s fallopian tubes. If the embryos are allowed to develop beyond the 2-cell

PAGE 64

50 stage, the procedure is then termed a tubal embryo transfer (TET). The only benefit to performing ZIFT or TET versus the tradition IVF procedure is for women who are thought to have compro mised embryo quality due to embryo culture in vitro. The prevai ling belief is that placement of the zygotes or embryos back into their own natural incubator, the fallopian tube, will enhance subsequent development, will be enhanced leading to improved pregnancy rates (Balmaceda et al., 1995; Dlugi et al., 1997; Wood, 1997). In 1992, the first pregnancies and births resulting from Intracytoplasmic sperm injection (ICSI) were reported. ICSI is the preferred ART procedure for couples in which the male partner has azoospermia, severe oligospermia, or severe defects of sperm shape known as te ratospermia. This treatment option is also recommended for those men with si gnificant antisperm antibodies, low sperm motility, or significant sperm mor phology. The final indication for the use of ICSI is when poor fertilization occurs with regular insemination techniques in the laboratory. In this pr ocedure, sperm are obtained eith er from the ejaculate or directly from the epididymis or testic le through surgical procedures such as TFNA, PESA, MESA, TESE, or TESA. As with IVF, the female patient undergoes controlled ovarian hy perstimulation and egg retrie val is performed. A single spermatozoon is injected through the zona pellucida directly into the oocyte. Once fertilization has taken plac e, the embryos are then transferred into the woman’s uterus (Steir teghem, 1995; MacLachlan, 1997). Finally, preimplantation genetic diagnosis (PGD) is currently being offered to couples receiving IVF and ICSI treatment s. PGD is a technique that combines

PAGE 65

51 recent advances in genetic research and reproductive medicine and examines the early embryo after ART procedures such as IVF for inherited diseases or to determine the sex of the embryo for sex -related genetic disorders. In 1968, Edwards and Gardner successfully performe d the first known embryo biopsy on rabbit embryos (Edwards & Gardner, 1968) Subsequent to this research, preimplantation genetic diagnosis for humans was developed in the United Kingdom during the mid 1980s. In 1989, Handyside and his colleagues refined the procedure and reported the first unaffected child born following PGD performed for an X-linked disorder. As of May 2001, more than 3,000 PGD clinical cycles have been documented and nearly 700 children have been born, demonstrating the reliability and safety of this procedure. The process is performed with ART procedures such as IVF where an embryo develops in a laboratory. When the embryo is at the 6 to 8-ce ll stage of development, 1 to 2 cells are removed. These cells are then sent to a genetics laboratory for diagnosis using either polymerase chain reaction (PCR) or fluorescence in situ hybridization (FISH) techniques. The embryos unaffected by genetic disorders or disease are then transferred into the wom an’s uterus. By transferring only the unaffected embryos, adverse outcomes such as miscarriages, pregnancy termination following prenatal diagnosis, or birth defects can be prevented. PGD is useful for identifying three major groups of disease. Chromosomal disorders include a variety of chromosomal rea rrangements, including translocations and inversions and deletions. PGD is also used to identify single gene defects including cystic fibrosis, Tay-Sachs di sease, sickle cell anemia, and Huntington

PAGE 66

52 disease. Finally, PGD is useful for dete rmining the sex of an embryo in which the specific genetic defect at a molecular level is unknown, highly variable, or unsuitable for testing on single cells. Disorders such as hemophilia, fragile X syndrome, neuromuscular dystrophies, and hundreds of other diseases can be identified through PGD. Sex-linked dominant disorders include Rett syndrome, pseudohyperparathyroidism, incontinentia pigmenti, and vitamin D-resistant rickets (Handyside et al., 1989; F linter, 2001; Harper, 2001). ART Success Rates Since the first infant was conceived from in vitro fertilization (IVF) in 1983 in the United States, the use of IVF and assisted reproductive technology (ART) has increased substantially. The 1992 Fertility Clinic Success Rate and Certification Act direct all U.S. clinics performing ART procedures to report data annually to the CDC for every ART procedure initiated. ART is defined as any fertility procedur e in which both oocytes and sperm are handled outside the body. Therefore, information regarding the number of clinics, cycles performed, live-birth deliv eries, and total number of live babies born as a result of in-vitro fertilization (I VF), gamete intrafallopian transfer (GIFT), and zygote intrafallopian transfer (ZIFT) procedures is collected by the CDC through the Society for Assisted Reproducti ve Technology reporting system. To date, six reports have been published under t he Fertility Clinic Success Rate ad Certification Act, the latest of which was in 2000. In the 2000 report, 98 percent of ART procedures include IVF stimul ated cycles, while the remaining two percent include GIFT and ZIFT treatm ent approaches and shows that the number of ART clinics, cycles performed, liv e-birth deliveries, and total number of

PAGE 67

53 live babies born as a result of ART hav e steadily increased since 1995. The number of infertility clinics in the United States has grown from five in 1982 to 383 in 2000. The number of ART cycl es performed in the United States increased 54 percentage points overall fr om 64,724 ART cycles 1996 to 99,639 ART cycles in 2000. In addition, the num ber of live birth deliveries increased from 14,573 in 1996 to 25,228 in 2000, repr esenting an increase of 73 percent. The number of live babies born as the resu lt of ART increased 67% overall, from 20,921 born in 1996 to 35,025 born in 2000 (Abma et al., 1997; Wright et al., 2003; CDC 2002). Although the National IVF Registry records information on all treatment approaches in which both oocytes and sperm are handled outside the body including IVF, GIFT, and ZIFT, the number of couples seeking other treatment options such as IUI is not as readily avai lable. However, the NSFG does collect information regarding the number of women who seek treatment for infertility as well as the type of treatment received. Findings of th e NSFG show that the use of infertility services and treatment incr eased steadily from 1982 to 1995. Of the 60.2 million women of reproductive age in 1995, 9.3 million reported having used some kind of infertility service at some ti me as compared to 6.8 million in 1988. Among childless women aged 35-44, 21 perc ent reported having used infertility services. The most common infertility services received by these women included medical advice, diagnostic testi ng and evaluation for either the woman or man, and ovulation inducing medica tions (Abma, 1997; Wright et al., 2003; CDC 2002).

PAGE 68

54 Associated Costs and Insurance for ART. By the end of 2000, more than 212,000 babies have been born in the United States as a result of reported ART procedures. IVF currently accounts for approximately 98% of the ART procedures with GIFT, ZIFT, and a combi nation representing the remaining 2% of procedures being conducted. Howeve r, ART procedures are the most expensive infertility treatment procedures performed. Reports of the average cost of one IVF cycle in the United States range between $8,000 and $12,400. While IVF and other ART procedures are not inexpensive, they account for 0.03% of U.S. health care costs. In 1988, the U.S. Office of Technology Assessment (OTA) published a study on th e medical and social aspects of infertility. In this study, the estimated total infertility expenditures nationally for 1987 were $1.0 billion. OTA provided a categorical breakdown of infertility treatment into the follo wing four typical stages: 1. Diagnosis and fertility drug treatment, 2. Complete evaluation of both partners, 3. Tubal surgery, and 4. In vitro fertilization. The average cost of the first stage of treatment was repor tedly $3,668; the second stage was $2,055; the third stage was $7,118; and the fourth and final stage was $9,376 (Ryan, 2001; RESO LVE, 1998-2003; ASR M, 2002-2003). ART in the United States is largely a fee-for-service, private market business. Because most insurance companies offer only partial coverage or none at all, an average of approximately 85% is the patient’s share of costs for

PAGE 69

55 IVF and IVF-related therapies in the United States, while a patient share of costs in France is approximately 7% and 15% to 28% in Canada. One consequence of this privatization is that access to AR T procedures is determined principally by financial status. Although some ART prov iders limit access to IVF to married couples using their own gametes or to women under forty and others exclude single women or gay or lesbian couples in general, the type and extent of treatment are primarily governed by how much the patient is able and willing to spend. As a result, there is a marked di sparity between the popul ation of infertile couples in the United States and those who are receiving services (Ryan, 2001). Although no federal law requires in surance coverage for infertility treatment, to date, 15 stat es have enacted some type of infertility insurance coverage law. While each law is unique, these laws require insurers to either cover or offer to cover so me form of infertility diagn osis and treatment. For states mandating coverage, health insur ance companies are required to provide coverage of infertility treatment as a benefit included in every policy. In states mandating the offer for coverage, health insurance companies are required to make available for purchase a policy that offers coverage of infertility treatment. However, the law does not require employ ers to pay for the infertility treatment coverage; instead employees may be offe red coverage as a rider to the insurance policy. In addition, coverage for infertility services varies from state to state. While some states require only that in vitr o fertilization treatments be provided by insurance, others specifical ly exclude coverage for this treatment. State law mandating employers to cover in fertility and related services currently

PAGE 70

56 include Arkansas, Hawaii, Illinois, Maryland, Massachusetts, Montana, New Jersey, Ohio, Rhode Island, and West Virgin ia. States mandati ng that employers offer coverage include California, Connecticut, and Texas (RESOLVE, 19982003; ASRM, 2002-2003). Currently, RESOLVE, a national infertility patient advocacy and information organization, is working with members of Congress on legislation requiring insurance coverage of infertility tr eatments. Legislat ion currently being introduced includes: The Family Building Act of 2003 (HR 3014) introduced by Representative Anthony Weibner and would require insurance coverage of infertility treatments, incl uding up to 4 IVF attempts, by all group health plans that also cover obstetrical benefits. In addition, coverage in self-insured health pl ans would also be required. HR 3026 is legislation sponsored by Representative Marty Meehan to require health plans available to federal employees, military personnel and their families to cover infertility treatments. HR 969, sponsored by Representat ive Rob Andrews would require Medicare coverage of infertility treatm ent services for those entitled to health insurance benefits under that pr ogram by reason of a disability. The Equity in Fertility Coverage Act of 2003 (HR 1852) sponsored by Representative Rob Andrews would re quire all health plans that cover Viagra, and similar medications, to also cover infertility treatment (RESOLVE, 1998-2003; ASRM, 2002-2003).

PAGE 71

57 Mandating coverage for infe rtility treatments is a c ontroversial issue. Infertility treatment, when it is covered at all, is generally limited to the diagnosis or diagnosis and treatment of correctabl e medical conditions. Insurers have justified exclusions of pr oviding coverage of IVF on several grounds. Advocates for mandating infertility treatment cover age argue that infertility is a physical problem that could be corrected medicall y and therefore that patients presenting with this condition should be afforded the same health care rights as those who suffer from illnesses such as diabetes; opponents contend that infertility is not an illness. Insurers typically employ a broad understanding of t he term “illness”. This term includes “diseases” or condi tions such as chemical dependency, congenital defects, alcoholism, hernias, headaches, senility, exogenous obesity, etc. In addition, since the majority of insurance carriers cover at least some infertility services such as diagnosti c testing and surgical correction for endometriosis, the argument that infertility should not be defined as an illness may be viewed as arbitrary and inconsist ent. Secondly, opponents of mandating insurance coverage that in cludes treatments such as IVF argue that IVF is not medically indicated because the proc edure does not correct the underlying medical problem. However, in Ralston v. Connecticut General Life Insurance Company, the court defined the standard for inclusion of a treatment for coverage as evidence demonstrating that the treatment wor ks as well as or better than presently availabl e methods. In addition, it was noted that consistent application of this principal asserted by insurance companies would require the denial of coverage for any treatment that merely compensates for or replaces

PAGE 72

58 any impaired or lost bodily function such as kidney dialysis, coronary bi-pass surgery, limb prosthesis, etc. Finally, opponents argue that IVF is an experimental therapy. In the Reilly v. Blue Cross and Blue Shield United case in 1988, the courts agreed with the insure r’s argument that a treatment with success rates less than 50 percent could be considered an experimental treatment. However, advocates for m andated insurance coverage laws contend that consistent application of this principa l would lead to exclusions of treatments typically provided to terminally ill pati ents since such cases have a zero success rate. In addition, advoc ates for mandatory insuranc e coverage of infertility treatment contend that since infertile c ouples pay premiums for health insurance benefits such as maternity services that t hey are unable to use, infertility services should also be covered by their health plan. However, opponents argue that the addition of this benefit increases the prem iums for a larger number of people as compared to the number of people who wi ll take advantage of this benefit and, therefore, that the increased costs for covering these procedures are unjust. In addition, employers and insurers contend t hat requiring coverage for infertility services, treatment or pr ocedure, increases the over all cost of insurance, resulting in an increased number of uninsured. Rebutting this argument, however, is a study conducted by Griffi n and Panak who examined the actual costs of providing coverage for infert ility, including IVF, and found that under Massachusetts’ mandate for group coverage plans, the increase in annual premium per person was $1.71 per month (Ryan, 2001; RESOLVE, 1998-2003; ASRM, 2002-2003).

PAGE 73

59 Legal and Ethical Considerations for Infertile Couples and ART. In American society, there is a strong legal and moral basis for the protection of autonomy in reproductive decisions. Deci sions made by individuals and couples about whether or when to produce children are thought to be a matter of personal private concern, not a subj ect for governmental legislat ion. These decisions are protected by the U.S. Constitution as t he right to privacy, has evolved through case law and has been supported by judici al decisions as early as 1942. In Skinner vs. Oklahoma, the U.S. Suprem e Court overthrew an Oklahoma statute authorizing the sterilization of habitual criminals convicted of crimes of moral turpitude. In this decision, the court stated: “[W]e are dealing with legislation which involves one of the basic civil rights of man. Marriage and procreation are fundamental to the very existence and survival of the race…” (Skinner v Oklahoma, 1942). This decision was followed by a series of cases involving contraception and abortion that further delineated how an i ndividual’s decision whether or not to have children was constitutionally protect ed from governmental interpretation. Legislative rulings defined the right to conceive and produce children as “far more precious than property rights.” T he constitutional right to privacy as recognized by the American judicial syst em, protects decisions to reproduce coitally because of the biologic and so cial importance placed on parenting. Furthermore, reproductive autonomy is ex tended to decisions to reproduce using alternative methods. For governmental r egulation interfering with reproductive

PAGE 74

60 decisions to be upheld as constitutional, su ch regulation must be necessary to further a compelling state interest and mu st regulate in the least restrictive manner possible. This right to privacy serves as the setting in which governmental actions and legislation must be measured. This right constitutes the basis on which health care professiona ls and infertility patients can challenge legislation that prohibits or restricts res earch and clinical practi ce in the area of assisted reproduction (Andrews & Hendricks, 1987). Although ART procedures are often the last hope for some couples to conceive a child, many ethical and mora l issues must be considered before embarking on such an invasive treatment pl an. Ethical consid erations regarding infertility treatment include issues such as the use of donor sperm, donor oocytes, and donor pre-embryo s. Issues regarding tr eatment options can be particularly difficult for infertile couples to resolve within their moral and personal belief systems. Other issues that c onfront infertile couples include the cryopreservation of oocytes and pre-embryo s and more specifically, what should be done with any unused oocytes and pre-em bryos. The latest procedure, preimplantation genetic diagnosis (PGD), may also provide ethical dilemmas to infertile couples. Because PGD provides physicians with the ability to select only the healthiest pre-embryos for implantatio n, this may conflict with a couple's ethical and moral beliefs about the sanctity of all life. In addition, issues regarding the use of PGD fo r the selection of a spec ific gender have evolved. Not only do couples have to weigh these infertility treatment options against their personal ethical belief systems, but they must also grapple these decisions with

PAGE 75

61 their religious teachings and convicti ons (Andrews & Jaeger, 1997; Schenker, 1997; Seibel et al., 1994; Quinn et al., 1997; Zilberstein & Seibel, 1997). Conceptualization and Operationa l Classifications of Stress While couples often describe infertility as the most stressful experience of their lives, the stress associated with infert ility is only one topic within a huge field of stress research. Public and scientific interest in the relevance of stress to health and disease developed shortly befor e World War II. One measure of the perceived importance of this issue is the amount of money and attention given to combating the effects of stress. Executive managemen t courses and many other self-help programs as well as a wide range of books aimed at preventing or alleviating stress-related problems have bec ome a thriving industry. In addition, stress-related medical complaints have hel ped to make antianx iety medications some of the most widely prescribed medi cations in the United States (Elliott & Eisdorfer, 1982). Along with public interes t, the proliferation of stress and stressrelated literature provides evidence of heightened scientific interest in the concept of stress. Resear ch on stress and the effects of stress has reached an all-time peak during the past two decades (Goldberger & Breznitz, 1993). Although stress has become a common topic of research, no one has formulated a definition of stre ss that satisfies even a ma jority of researchers. Controversies regarding the conceptua lization of stress and stress related research plague the field. While some scientists believe that the conceptualization of stress has become ove r-generalized, other scientists believe that the broad conceptualization of st ress provides an invaluable unifying

PAGE 76

62 terminology for a particular type of important research (Elliott & Eisdorfer, 1982). However, Pearlin, Lie berman, Menaghan, and Mullan (1981) suggest that the core meaning of the concept of stress is not confusing. They assert that there is general agreement that the term stress refers to a response of the organism to a noxious or threatening condition. T he confusion and disagreement arise with regard to where and how to identify this re sponse. Is stress to be identified by the functioning of an or gan or a system of organs, by biochemical or physiological response patterns, by changes in emotional states, or by the presence of illness or disease entities? Other debatable dimensions of stress arise with respect to the duration, indivi dual perception, and si tuational context. For example, is stress mani fested in short-term reacti ons of the organism or in long-term dysfunctions? Are individuals aw are of the stress they harbor to the extent that they can report it or mu st the presence of stress be determined by independent measures? Is stress a global, encompassing state, or is it confined to specific situations or contexts in wh ich it is aroused? According to Pearlin (1993), one problem in defining and underst anding stress is that the nature of stress is a diffuse and multidimensi onal phenomenon and can mean so many different things. The many different c onceptualizations of st ress that exist in research appear to meet specific needs within a given context of research. However, stress tends to be characterized in three broad categories: systemic or physiological, psychological, or social. Systemic or physiological stress is concerned primarily with t he disturbances of tissue systems. Psychological stress focuses on cognitive factors leading to the evaluation of a threat. Social

PAGE 77

63 stress is defined by the disruption of a so cial unit or system. While many agree that the three types of stre ss are related and overlap in various ways, the nature of their relationships is far from clear (Monat & Lazar us, 1977; Pearlin, 1993). Within these three categories of st ress, basic themes emerge in the research literature. Co mmon among all research related to stress is the identification of stressors. Stressors ar e defined as external events or conditions that affect an organism. In human research the investigation of the impact of a given stressor on an individual is widely studied within the context of the cognitive appraisal of stressors. In accordance with Lazarus’s formulation, cognitive appraisal plays a vital role in the transaction between the person and a potentially stressful environment. Resear chers are largely interested in the effects of stress. Research investigat ing the effects of stress includes a wide range of impacts from minor changes in behav ior to dramatic clinical symptoms. Another common theme in stress research relates to coping. After an individual appraises a situation, the i ndividual will use one or more coping strategies in an attempt to adjust the environment or sit uation. Within the stress literature, studies investigating the various coping strategies encompass a large body of research (Goldberger & Breznitz, 1993; Selye, 1993). In July of 1979, the Office of Science and Technology Policy (OSTP), Executive Office of the President, request ed a “definition of research issues, delineation of desirable and adverse aspects of stress in its various forms, and biomedical, behavioral, and sociologica l approaches to the description and alleviation of excessive stresses”. Several other agencies supported this

PAGE 78

64 research including the National Scienc e Foundation (NSF), the Office of Prevention of the National In stitute of Mental Health (NIMH), and the National Institute on Aging (NIA). The National Academy of Sciences’ Institute of Medicine accepted this daunting project. Their study titled “Research on Stress in Health Disease” indicated a strong bias in stress research. Their review of the stress literature revealed that many st ress researchers have emphasized mainly adverse consequences of stress, confirmi ng many people’s conceptualization of stressors as being inherently “bad”. T hat stressors and reactions to stressors produce a wide range of consequences, only so me of which may be undesirable, was highlighted through this study. Focusing more attention in the literature and future research on the positive cons equences associated with stressors was advocated (Elliott & Eisdorfer, 1982). Theories of Stress The foundation for modern theories of stress can be traced back to ancient Greece. Hippocrates, often cons idered the “father of medicine”, clearly recognized the existence of the healing power of nature comprised of inherent mechanisms of the body for restoring health after exposur e to pathogens. However, early investigations conc eptualized stress as being inherently unpleasant rather than encompassing experiences with both positive and negative outcomes. In 1879, the French physiologist Bernard advanced this subject by pointing out that the internal environment of a living organism must remain fairly constant despite changes in the external environment. Bernard’s pioneering studies on the particular adapt ive changes by which the steady state

PAGE 79

65 is maintained provided the foundation for later research. During the early 1900s, Cannon, an American physiologist, coi ned the term “homeostasis” from the Greek homoios meaning similar, and stasis meaning position. This term referred to “the coordinated ph ysiologic processes, whic h maintain most of the steady states in the organism” (Cannon, 193 9). In his work on blood hormones, Cannon frequently studied the effects of physi cal or emotional “stress” defined as stimuli that disrupted an i ndividual’s normal internal en vironment. In totality, Cannon’s studies established the existence of many highly specific physiological mechanisms for protection against a variety of threats to a body’s constant state. In particular, he emphasized the stimul ation of the sympathetic nervous system and the resulting hormonal discharge from the adrenal glands. The stimulation of the sympathetic nervous system occurs during emergencies such as pain or rage. In turn, this autonomic proce ss induces the cardiovascular changes that prepare the body for flight or fight. This research provided the foundation for stress research and the resulting theorie s of stress (Selye, 1977, 1993; Elliott & Eisdorfer, 1982). Most current stress models are based on the conceptualization of stress in three distinct ways. The first approac h conceptualizes stress as a physiological adjustment process and views stress as an internal response. In the second approach, stress is characterized as an external or situational stimulus. Proponents of this model attempt to descri be the characteristics of a stressful environment as well as the particular st imuli that produce stress. The third conceptualization of stress is a synthesis of the first two approaches. While

PAGE 80

66 circumstances may be intrinsically stre ssful, psychological process such as perceptions of abilities, needs, persona lity, and resources interact with the external events and stimuli to produce a vari ety of responses. In this interactive approach, the pivotal concept is that individuals are active agents who continuously cognitively appraise themse lves and their environment and evaluate the fit or misfit between these vari ables. When perceived environmental demands exceed the individual’s perce ived response capabi lity or when the environment is not able to meet the indi vidual’s internal needs and values, an individual’s stress increases ( Lazarus, 1966; McGrath, 1970). The first conceptualization of st ress as a response arose from physiological research. O ne of the most popular m odels conceptualizing stress as a response is the General Adaptation Syndrome (GAS). The popularity of this line of stress research can arguably be attr ibuted to the work conducted by Hans Selye. While attempting to discover a new sex hormone, Selye discovered that rats receiving multiple doses of a cr ude ovarian extract developed many physical maladies including enlargement and hyperacti vity of adrenal glands, involution of the thymus and lymph glands, and gastric ulce rs. It was later discovered that all toxic substances, irrespective of thei r source, elicited the same pattern of physical responses. In addition, he doc umented identical organ changes evoked by stimuli such as cold, heat, infection, trauma, hemorrhage, nervous irritation, and many other stimuli. These changes we re identified as objective indices of stress and furnished the basis for the dev elopment of his stress concept. He suggested that individuals exposed to a noxious stimulus responded with what

PAGE 81

67 he referred to as the general adaptation syndrome (GAS) or the biologic stress syndrome. Three stages are identified in the GAS. T he first stage is the alarm reaction, in which adaptation has not yet o ccurred. Selye asserts that this stage probably represents the somatic expression of a generalized call to arms of the body’s defense forces. During the alarm re action, the cells of the adrenal cortex discharge their secretory granules in to the bloodstream and thereby become depleted of corticoid-containi ng lipid storage material. The second stage is the stage of resistance, in which adaptation to the stressor is optimal. During this stage, the cortex becomes particularly ri ch in secretory granules. Evidence of the body’s adaptation is demonstrated th rough hemodilution, hyperchloremia, and anabolism, with a return to normal body we ight. The final stage in this model is the stage of exhaustion, in which the ac quired adaptation is lost again. This stage follows the stage of resistance as long as the demand is severe enough and applied for a sufficient length of time. This stage exemplifies that the body’s ability to adapt is finite, since, under constant stress, exhaustion eventually ensues (Selye, 1966, 1977, 1993; Dohrenwend & Dohrenwend, 1980; Elliot & Eisdorfer, 1982). Selye later revised t he GAS stating that organismic stress is basically the same regardless of the type of stressor and described two distinct types of stress: eustress and distress. Eustress is defined by Selye as “good” stress such as commitment to accomplis hment while distress was viewed as “bad” stress such as frustration and resent ment. According to Selye, distress is destructive to health while eustress is not (Lazarus et al., 1980; Selye, 1977, 1993).

PAGE 82

68 While stress research expanded thr ough Selye’s work on physiological responses to stress, research on the p sychological effects of major life events received increasing attention with the pub lication of the Holm es-Rahe scale of life-change events in 1967. Life events ar e defined as objective experiences that disrupt or threaten to disr upt an individual’s usual acti vities, causing substantial readjustment in their behavior. The foundation of life-events research is established in the work completed by Cannon. Through his research, Cannon demonstrated that emotion-pr ovoking stimuli could produce the physiological alterations necessary for “fight or flight”. He further proposed that physical illness would result with long or persistent stim uli producing such physical reactions. Adolf Meyer modified this argument during the 1930s by asserting that ordinary, normative changes in patients’ lives such as births, deaths, and job changes may play a part in the etiology of disease. The work conducted by Meyer and Selye gave legitimacy and impetus to studies ex amining not only reactions to physical stimuli, but also to studies of psycholog ical stimuli as potential stressors (Thoits, 1983). Theories of psychological stress c enter on negative emotions such as anger, fright, anxiety, shame, guilt, s adness, envy, jealousy, and disgust. However, it is recognized that positive em otions including happiness, pride, relief, and love can present some mediating effe cts (Lazarus, 1993). One of the main distinctions between psychological levels of stress as compared to stress at the physiological levels is the presumption that cognitive activities such as evaluative perceptions, thoughts, and inferences are used to interpret and guide every

PAGE 83

69 adaptational interchange with the environment. In this conceptualization of stress, an individual is said to apprai se each ongoing and changing transaction with the environment. This appraisal includes judgments about environmental demands and constraints as well as resources and options available for managing them (Lazarus et al., 1980). T hese conceptualizations of stress have developed into a cognitive-motivational -relational theory that highlights the importance of emotions in stress research (Appley & Trumbull, 1977; Lazarus, 1993; Strelau, 1995). Within this theory lies the assertion that emotions are organized psychophysiological reacti ons to information and knowledge. Although Duffy (1941) argued t hat emotion is an unnecessary concept because it refers to activities that ar e not different from life it self, concerned with adapting to the demands, constraints, and opportunities of living, Lazarus (1993) contends that emotions are different from many other adaptationa l activities because they are characterized by active psychobiologi cal involvement in what is happening. Furthermore, Lazarus asserts that wit h emotion, an individual has a vested interest in the outcome (Appley & Trumbull, 1977; Lazarus, 1993). In the cognitive-motivational-relational theory, relational refers to the metatheoretical assumption that emoti ons are always about the relationship between the person and environment as opposed to environmental demands or individual needs and processes. This t heory defines emotion as an interaction through an event creating personal harm, threat, or benefit on which the emotions are predicated. The principle underlying motivation in emotion is that emotions are reactions to the status of an indivi dual’s goals in everyday

PAGE 84

70 encounters and in the individual’s life overa ll. The term “moti vational” refers to the hierarchies of importanc e for goals that an individual brings to any event. The transactions that take place in a parti cular situational context activate these goals as stakes in the outcome of t he event and generate new goals. The final component of this theory refers to the cognitive knowledge and appraisal of what is happening during an event. While kno wledge consists of a set of beliefs, which are either situational or generalized across sit uations about how the world works, appraisal is an evaluation of t he significance of what is happening in terms of one’s well being. Appraisal is essential in the gener ation of emotions because it concerns an individual’s personal stake in an encounter or event. The quality and intensity of emotions depend on s ubjective evaluations or cognitive appraisals of how an individual is doing with respect to their goal commitments both short-term and long-term. In addition, the tendencies of an individual to act are generated by their cognitive apprai sals, which, in turn, influence an individual’s emotional reaction (Appley & Trumbull, 1977; Lazar us, 1993; Strelau, 1995). According to Lazarus (1993), emotions are salient cognitive-motivationalrelational configurations within the person-environm ent relationship and are shaped by the understanding and ev aluation of this relationship by the individual. In addition to the theoretical framework including emotions in stress research, Lazarus defined distinctive patterns of appraisal for individual emotions. As defined by Lazarus (1993), the central theme of anger is a demeaning offense against an individual, which de pends on the individual’s appraisal that their self-

PAGE 85

71 esteem is at stake in an encounter. A ccording to Lazarus (1993), anger arises when an individual is treated as less than he or she would wish. Anger is elicited whether or not the intent is malevolent, but especially when it is. Those individuals with a vulnerable self-esteem become angry more quickly than others because it is difficult for them to wave off the attack as unimportant. Although Lazarus (1993) described the action tenden cy of anger is to attack through retaliation or vengeance, Averill (1983) revealed that the episodes of anger reported by college students rarely involv ed an actual attack. Lazarus concedes that anger is regarded ambivalently in our society: on the one hand, anger is self-preservative; on the other hand, anger may be socially destructive (Lazarus, 1993). Many researchers and theoris ts regard anxiety as an emotional result of a threat to an individual’s well being and to the essential meanings that comprise that sense of well-being. When a threat is present ed to an individual and the threat is ambiguous, in other words, the individual does not know what will happen, when it will happen, and therefore, does not know what is to be done about it; then the emotional response is sa id to be anxiety. In addition, even when the threat is concretized and externalized, the concrete condition represents the more existent ial questions of who we are and what life meanings we hold. With anxiety, when one such th reat has been dealt with, another threat always comes in its wake. Distinguished by this pattern, anxiety is an emotion different from all others. Becaus e of this unique pattern, theories of psychopathology often center on anxiety and more specifically on inappropriate

PAGE 86

72 ways of coping with anxiety as the basic cause of psychopathology. Anxiety is produced when meaning and ideas are at stak e. A threat to these meanings and ideas creates goal incongruence. If so mething or someone is held accountable for the threat, then the emot ion is not anxiety. While anger assesses blame to something or someone else, anxiety is uniqu e in that there is no blame to place elsewhere. When blame is assessed to ones elf and the individual believes that a moral or imperative has been transgress ed, the emotion provoked is guilt. Similarly, shame assesses blame to onesel f. However, shame is the resulting emotion when an individual believes he or sh e has failed to live up to an ideal or expectation. Unlike anger, shame and guil t, sadness is an emotion similar to anxiety in that no blame is assessed. As with anxiety, sadness is ambiguous. Ambiguity about sadness revolves around w hether it should be classified as a mood or an acute emotion. However, s adness is described by Lazarus (1993) as a unique negative emotion pr oducing a reaction to an irrevocable loss which creates goal incongruence. Furthermo re, sadness comes at the end of the grieving process. When an individual has accepted that he or she is helpless to change the situation, accepting the loss as irrevocable, then the individual experiences sadness. Before the acc eptance of a loss as irrevocable, the individual will experience a variety of ot her emotions, which focus on trying to restore or ameliorate the threat or loss including numbness, denial, anger, anxiety, guilt, or shame (Lazarus, 1993). The conceptualization of stress in so cial contexts identifies modern man as the stress system while dominant st ressors are derived from the social

PAGE 87

73 environment. Responses to these stresso rs are principally alterations in perceptual, intellectual, emotional, and consciousness activities, while physiological changes are secondary. Soci al stress is defined by Brown (1980) as an unfavorable perception of the soci al environment and its dynamics. In addition, Brown asserts that nearly all psychosocial stress-related disturbances develop from an individual’s perception of their social situation. Psychological and physiological disturbances believed to be caused by, related to, or aggravated by social stress include em otional, psychosomatic, organic, psychological adjustment, and sociological problems. The existence of such diverse and enumerated reactions to soci al stress has made the application of a specific term or descriptive definition la beling the major dete rminants difficult. The conceptualization of social stress, how ever, is similar to conceptualizations in psychological stress. Complex inte llectual functions are assumed to be involved in reactions to social stress. The sequence of mental activities delineated by conceptualizati ons of social stress includes an individual’s expectations, perceptions of the social environment, interpretation of disparity between expectations and perceptions, rumi nation, perceptual distortion, and cortical inhibition. Expectations are mult i-dimensional, subjective activities. An individual’s history and experiences s hape a person’s environmental and social expectations. In additi on, expectations are determined by, modified by, and related to personal aspirations, and mo tives. Perception of the social environment is a global activity and is in large part determined by the individual’s expectations. A number of events must be observed, associated, analyzed, and

PAGE 88

74 judged. An individual’s perception of social situations is an interpretation and a mental construction of social events and their significance not only in present tense, but also in the past and future (Brown, 1980). When a significant disparity between the expectations of an individual and their perception of the environment occurs, in tense cognitive activity takes place. This activity has a direct effect upon bot h subjective sensations and physiological activities. The disparity represents a conflict between the emotions and appraisals of expectancy as well as the em otions and appraisals of denial or of a lack of satisfaction. Following the rec ognition of a disparity between expectations and perceptions of reality, mental activity begins both at the conscious and unconscious level in an attempt to re solve or understand the reason for the disparity. Problem solving activities aimed at developing coping devices or resistance to the perceived stress are ac tivated. Rumination generates anxiety or apprehension which directly activates the physiological defense mechanisms resulting in muscle, visceral, and subjec tive tension. At the same time, rumination involves the almost constant creation and re-creation of the social context and problem as mental images. Along with these re-creations of the context and problem are projections of va rious alternative solutions into both past and future imagined situations. These images directly induce physiological activation. For example, the thought or imag e of eating a sour lemon leads may individuals to the production of saliv a and other physiological responses. Likewise, conscious or unconscious images of anxiety-producing situations can excite physiological responses that mimic the original reactions. In addition, all

PAGE 89

75 organs of the body respond to mental im ages involving those organs. With certain predispositions, the images may produce any or all of the physiological arousal responses accompanying anxiety (Brown, 1980). Following rumination, perc eptual distortion occurs when an individual’s attention becomes directed toward thos e elements or stimu li in the social environment that are related to their ment al construction of the problem. As a result, a skewed perception develops in that the individual sees and hears predominantly those things of the social dynamics that fit their preconceived mental images of the sit uation and problem. This distortion intensifies the significance and breadth of the problem while strengthening inappropriate solutions, eventually leading to height ened levels of distress (Brown, 1980). The final stage in this conceptualization of the stress process is referred to as “cortical inhibition”, a term coined by Hefferline. This stage occurs when an individual’s cognitive activity becomes mo re narrowly focused. At this stage, normal homeostatic mechanisms regulatin g neural conduction of muscles and viscera become impaired. These physiological responses reduce the subjective appreciation of the cause and effect relati onship of the tensions experienced by the individual while at the same time increasing the sensation of tension from unknown origin and physiologica l adaptation. The cortical inhibitory effect is often inferred to indicate that a distortion in the individual’s perceptions of social reality takes place (Brown, 1980). The sequence of mental activities fr om an individual’s expectations to perceptual distortion structure the cognitive activities t hat process social data in

PAGE 90

76 such a way that elements of the social environment are interpreted as posing threats to an individual’s well-being. Specifically, rumination generates both threats to well being along with emotional sensations and mental images which excite defense postures. These emotional sensations and mental images appear to be the primary psychological and physiol ogical manifestations of the social stress response (Brown, 1980). The conceptualization of psychosocial stress is a synthesis of psychological and social stress theories. Research into the psychosocial basis of psychological distress has a long histor y in the behavioral sciences. However, the focus of this research is limited to a num ber of specific areas. The majority of this research is primarily found in either ongoing interpersonal relationships or in discrete life events. According to Kapl an (1983), “psychosocial stress refers to socially derived, conditioned, and situated psychological processes that stimulate any or all of the many manife stations of dysphoric affect falling under the rubric of subjective distress” (p. 196). Likewise, in the conceptualization of psychosocial stress, psychological distress reflects an individual’s inability to prevent or diminish perception, recall, anticip ation, or imagination of devalued circumstances. The emphasis, however, is placed on the individual’s cognitive and affective-evaluative interpretation of circumstances rather than on the circumstances themselves in the elicitati on of psychological distress. There are three mutually influential psychologica l process components of psychosocial stress. The affective process refers to the individual’s need – value structure that influences the individual’s perception of reality and motivates the individual to

PAGE 91

77 behave according to valued states. Cogniti ve processes include the individual’s perception, recall, anticipat ion, and imagination of the situation or environment. Finally, the attributes and responses of the individual and others constitute the behavioral processes. These behavioral processes stimulate an individual’s subjective awareness which in turn stimulates the i ndividual’s need-value system. The conceptual framework of psychol ogical distress directly reflects an individual’s subjective perc eption of devalued circumstan ces. Subsequent to this perception is the outcome of the infl uences on the individual’s need-value system, the occurrence of t he circumstances that, from the perspective of the individual’s need-value system, are di svalued, and the perception of the occurrence of the disvalued circumstances (Kaplan, 1983). Measurements of Stress In scientific investigations, the nat ure of the theorie s proposed frequently dictate operational definitions of the constructs and boundaries of the conceptual domain from which theories are derived. These operational definitions specify or structure the nature of t he measurement used to conduct investigations. While stress research is no exception in th is regard, the complex and sometimes contradictory array of stre ss theories has resulted in a multitude of operational formats for stress measurem ent. However, one of t he primary modalities for stress measurement is through self-reporting. Self-report measures of stress are a predominant choice for researchers bec ause so many of the popular theories of stress emphasize intrapsychic cogniti ve processes such as the use of appraisal and application of coping skills or emotional states such as anxiety or

PAGE 92

78 depression as central themes to the definition of stress. Self-report measures of stress have been largely influenced by thr ee classical approache s to theories of stress: stimulus-oriented theories, respons e-oriented theories, and interactional or transactional theories (Derogatis & Coons, 1993). Because stimulus-oriented stress research focuses on the intrinsic potential for stress in the environment, m easurements of stress derived from this approach address the significant characteri stics of the environm ent that impinge upon the individual and incl ude methods that different ially assign weights or quantify the stress value of the environm ental stimuli. Unfortunately, few approaches have given rise to a consistent psychological measurement strategy. However, life events research is an except ion to this trend. Modern research on life events can be traced back to the publication of the Schedule of Recent Experiences (SRE) developed in 1957. Research on the psychological effects of major life events has proliferated over t he past few decades. Since World War II, the relationship between life events and psychological disturbance has been studied in three major ways: 1. Psychiatric effects of particular events such as marriage, the birth of a child, divorce, and death, 2. Psychological effects of multip le events in the lives of random samples of adults and children, and 3. Comparisons of the number and ty pes of life events experienced by psychiatric patients prior to hospita lization to those experienced by nonpatient control groups.

PAGE 93

79 The SRE originally contained 42 it ems and was regarded as a life events incidence measure. The earliest, best known, and most widely used approach to measuring life events is the Holmes-R ahe scale of life events (1967). Rahe (1974) modified this original set of items on the SRE and added 13 questions designed for prospective research on life change and illness. This instrument is known as the Recent Life Changes Q uestionnaire (RLCQ) (Derogatis & Coons, 1993). Since the development of this scale a host of life events checklists have been developed, including the Psychiatric Epidemiology Research Interview (PERI) Life Events Scale. An alternativ e to these self-repor t instruments is the structured event probe and narrative rati ng method. This approach requires semi-structured interviews designed to e licit a detailed descr iption of each life event and are then evaluated by independen t raters (Dohrenwend et al., 1993.) Numerous other scales have since been developed to measure life stress. In addition to the SRE and RLCQ, the Life Experiences Survey (LES) (Sarason, et al., 1979) and the Global Assessment of Recent Stress (GARS) (Linn, 1985) and many other innovative life stre ss measures have subsequently been developed. One criticism of these scales is that the events that comprise the scales are irrelevant for many subgroups of society. Current research is beginning to address this issue (Horo witz et al., 1977; Sarason et al., 1979; Blake et al., 1984; Linn, 1985; Dise-Le wis, 1988; Greenberg, 1990; Derogatis & Coons, 1993). The work of Cannon (1932) and Selye ( 1994) has led to many responseoriented theories of stress. The majori ty of instruments developed from this

PAGE 94

80 approach have arisen from clinical resear ch in psychopathology. A multitude of self-report measures have been developed to address t he various domains of psychopathology, mood and affect, psychological adjustment, and social competence. In a review by Piotrowski and Lubin (1990), seven out of ten of the most frequently used scales in health psychology were reported to be psychological symptom inventories and sca les that reflect mood and affect. These instruments are self-r eport instruments t hat are used most prominently as presumptive measures of stress. Most of these instruments are multidimensional, reflecti ng the multitude of symptom complexes and myriad dysphoric emotions typically in voked to define stress. The Minnesota Multiphasic Personality Inventory (MMPI) (Hathaway & McKinl ey, 1940) is one of the bestknown multidimensional psychological te sts and has been regarded as pivotal in the development of personality research over the past 50 years. A second popular multidimensional self -report instrument is the SCL-90-R (Derogatis et al., 1973) and is designed to assess symptomati c psychological distress. The SCL90-R reflects psychological distress in nine primary symptom dimensions: Somatization (SOM), Obsessive-compulsive (OBS), Interpersonal sensitivity (INT), Depression (DEP, Anxiety (ANX), Hostility (HOS), Phobic anxiety (PHOB),

PAGE 95

81 Paranoid ideation (PAR), and Psychoticism (PSY). Although most self-report instruments fr om the response-oriented perspective of stress are multidimensional, specific syndromes, particularly those that have become synonymous with definitions of st ress such as anxiety, have fostered dedicated unidimensional inst ruments. Examples of popular unidimensional scales include the Beck Depression Inventory (BDI), the State-Trait Anxiety Inventory (STAI), and the Center for Epidemiological Studies Depression Scale (CES-D). The BDI (Beck et al., 1961) is a unidimensional symptom inventory focused on the measuring the behavioral mani festations that define the construct of depression. The STAI (Spielberger et al., 1970) is a self-report symptommood inventory comprised of two unidim ensional scales intended to provide an operational distinction between anxiety as a transient emotional experience and anxiety as an enduring personality characte ristic. The CES-D (Radloff, 1977) is a brief scale measuring depression by assessing mood and level of overall functioning during the most recent week (Derogatis & Coons, 1993). Along with the responseoriented instruments based on psychological symptoms, affect and mood scales have been employed to operationalize stress through this approach. These scales are typically collections of adjectives depicting various mood states that are often selected on the basis of factoranalytic studies. Three popular affect and mood scales are:

PAGE 96

82 1. Profile of Mood States (POMS) (McNair et al ., 1971) – reflects six primary mood states including tension-anxiety, depressiondejection, confusion, angerhostility, vigor, and fatigue. 2. Affects Balance Scale (ABS) (Derogatis, 1975) – incorporates positive and negative affect measur ing current emotional status, general well-being, or treatment -induced change (Derogatis & Coons, 1993). 3. Positive and Negative Affect Schedule (PANAS) (Watson et al., 1988) – designed to express t he idea that both positive and negative effects must be measured to achieve a valid estimate of general well-being. The interactional theories of stress posit that cognitive, perceptual, personality, and other characteristics inherent in the individual mediate the response to stress. Instruments devel oped from these theoretical approaches have been based on individual and family in teractionist models and represent newer measures of stress. The Jenkins Activity Survey (JAS) (Jenkins et al., 1967) is a self-report screening instrument developed to meas ure a specific pattern of behavior thought to have a high association with proneness to coronary disease known as the type A behavior pattern. The Derogatis Stress Profile (DPS) (Derogatis, 1987), an interact ional stress measure, reflects stimulus, response, and interactional el ements derived directly from stress theory. This instrument was conceived to be a truly interactional measure of stress that provides information on t he level of environmental stress the

PAGE 97

83 individual is subjected to, the impact of stressors, and the level of conscious emotional distress the individual is exper iencing as a result of the stressormediator interaction. Al though interactional theories of stress have guided the development of relatively few self-report instruments, other in struments include: Ways of Coping Checklist (WCCL) (Folkman & Lazarus, 1980, 1985) – designed to identify strategies indi viduals use to deal with the demands of a stressful event, Family Inventory of Life Events and Change (FILE) (McCubbin et al., 1981) – assessing family reactions to the accumulation of demands associated with stressf ul events and changes, Family Crisis Oriented Personal Evaluation Scales (F-COPES) (McCubbin et al., 1981) – evaluati ng the cognitive and behavioral coping strategies families use when faced with stressful events, and Coping Health Inventory for Parents (CHIP) (McCubbin et al., 1983) – designed to examine the coping stra tegies used by parents who have a chronically ill child (Derogatis & Coons, 1993). In addition to these instruments and methods used to capture information about stress with any given theoretical approach, one reviewing the research literature on stress is likely to encounter references to autonomic responses. These physiological indices of stress, in fact, are rooted in the work conducted by Selye and other researchers that culminated in such a vast and varied field of research. Physiological measures of stre ss are just as varied as the theoretical conceptualizations and definitions of stress. Psychophysiology represents a

PAGE 98

84 common interest of psychologists, physi cians, and biomedical engineers in the analysis of mind-body interactions th rough the measurement of bioelectric signals. Therefore, this field of res earch represents a heterogeneity of interests largely defined by its methods. Alt hough the sympathetic and parasympathetic divisions of the autonomic nervous system consists of all the nerves that innervate the smooth muscl es of the viscera, the endocrine glands, the heart, and the blood vessels, psychophysiologic al attention primarily focuses on cardiovascular measures and on the m easurement of palmar sweat gland secretions. Electrodermal responses refe r to skin conductance. Measurement of these responses has become a common practice in psychophysiological assessment. Although the exact relati onship between psyc hological phenomena and electrodermal responses is still not ent irely understood, its popularity as an index of stress lies in the fact that the innervation of the sweat glands is exclusively known to evoke anxiety and stress reactions. Because many technical considerations are necessa ry for the proper measurement of electrodermal activity, recommended stand ards have been established by the Society for Psychophysiological Resear ch. Other popular psychophysiological measures evaluate cardiovascular activi ty and include the following measures: Electrocardiography – meas uring the electrical activity associated with the contraction of the cardiac muscle, Impedance Cardiography – used to measure the impedance of the thorax to a high frequency low level alternating current going from the

PAGE 99

85 neck to the abdomen through a band el ectrodes wrapped around the abdomen, and Blood Pressure – measuring the pr essure within the arteries during each cardiac cycle. In addition to these psychophysiological measures, fundamental biological markers are also used as indices of stress These include, but are not limited to, adrenaline excretion, plasma triglycerides, serum iron levels, neurotransmitters, corticosteroids, and neuropeptides. Wh ile these indices are arguably more objective measures of stress, criticism of these biological makers as indices is that they lack sufficient specificity (Fo wles et al., 1981; Derogatis & Coons, 1993; Katkin et al., 1993; Thoits, 1983). Stress Research Since the publication of the study “Research on Stre ss in Health Disease” in 1979, while stress research has still remained mainly concerned with maladjustment or the negative consequences associated with stressors, interest in successful coping strategies has bec ome a field of interest and study among researchers. The domain of stress research now puts heavy emphasis on coping and coping mechanisms. Interest in coping strategies and individual predispositions, as well as the efficacy of teaching coping skills, has grown substantially. Goldberger and Breznitz ( 1993) assert that this new emphasis in stress research demonstrates an optimistic bias. They maintain that practices related to coping skills rest on the assu mption that given the right tools, an individual can cope effectively with most so urces of stress. A nother indication of

PAGE 100

86 an optimistic bias in the stress research and literature is the at tention given to the importance of the idea of control. Researchers asse rt that an internal locus of control is preferable to an external one and argue that self-control can be used effectively to combat the potentially ha rmful effects of stress (Goldberger & Breznitz, 1993). Based on this optimistic bias now found in the stress research and literature, Breznitz and Goldberger ( 1993) predict that st ress research and theory are about to undergo a major change in emphasis. As opposed to investigating the negative im pacts of stressors, Breznitz and Goldberger contend that illness related impacts of stress will give way to consideration of stress as a force conducive to health. Furthermore, they assert that this upsurge of interest in the positive effects of stress will signi ficantly increase the relevance of the research field of stress (Goldberger & Breznitz, 1993). While early investigations conc eptualized stress as a nonspecific response of the body to any demand, it now encompasses experiences as well as the body’s response to these experienc es. In current re search, the term stress generally refers to challenges, real or implied, to the homeostatic regulatory processes of the organism. T herefore, stimuli such as heat or cold and physical trauma are direct assaults on the homeostasis of an organism and emotions represent internal states that threaten the inte rnal stability of a body. Current research has moved past invest igations focusing on stressors to investigations focusing on the stress re sponse. Investigations of the stress response consist of a cascade of neur al and hormonal events that may have short term or long term consequences on the brain and the body. Within this

PAGE 101

87 view, a stressor is defined as an environmental event that is likely to cause a negative outcome such as disease and psychological stressors such as fear or anxiety involve perceived threats to a body’s homeostasis and are likely to evoke psychosomatic reactions such as gastr ic ulcers or immunosuppression and involve changes in the neural and horm onal output. A primary research focus between stress and pathophysiological responses has been on the adrenal steroids as agents that m ediate adaptation and damage as a result of stress. Other research has focused on the psyc hological and biological effects of stress on the immune system through neuroendocrine pr ocesses, while other research is demonstrating that in the stress res ponse, a dissociation between the different endocrine systems often occurs (McEwen & Mendelson, 1993; Stein & Miller, 1993). In addition, psychoendocrinological stress responses are being examined in infertility research based on the evidenc e that prolactin (PRL) and cortisol concentrations appear to be important in fe rtility, (Demyttenaer e et al., 1992). As this research continues to grow in the fields of neurochemistry, neurobiology, and neuroendocrinology, and the neurosciences in general, a new scientific paradigm that redefines the stress construct to include the complex relationships between developmental, psychosocial, and biolog ical science may emerge and provide a more precise construct with more tangi ble, quantifiable entities in which to investigate the stress respons e (Derogatis & Coons, 1993). The widespread interest in stress and the stress response can be seen through the establishment of a variety of professional organizations dedicated to this field of research. One such organi zation, The American Institute of Stress,

PAGE 102

88 continues to report on the latest advanc ements in stress research and health related issues. This organization also sponsors the International Congress on Stress which was initiated in order to a ssemble leading authorities from all over the world to present cutting edge research advances and state of the art reviews (American Institute of Stress [AIS], 2003; Rosch, 2003). Other established professional organizations include the Stress and Anxiety Research Society (STAR) and the Center for the Study of Stress and Adaptation that are both multidisciplinary research centers. In addi tion to the medical interest in stress and the stress response, social scientists c ontinue research in the field. Current popular research focuses on stress ecol ogy, job or work related stress, and research on family stress and coping mechani sms. In addition, with national high stakes testing requirements for students nationally, educators have a continued interest in test and performance anxiety as well as how to mitigate these responses in order to provide accurate estimates of a student’s achievement on these high stakes tests (Ball, 1995; Stress Research Center [SRC], 2003; Spielberger & Vagg, 1995). Stress and Infertility Generally, infertile couples have carefully examined their reasons for desiring a child. They identify many of t he philosophical values that most take for granted such as links to the future, sour ces of pleasure, pr ide, and challenge, and meaning in life. For many couples se eking medical treatment for infertility, parenthood is an important life goal that satisfies the need to develop as an adult and to demonstrate independence and creat ivity (Anderson & Alesi, 1997).

PAGE 103

89 According to Savage (1989), the psycholog ical importance of parenthood is that it biologically preserves the continuation of the spec ies and the child, in part completes the parents. The inability to c onceive a biological child is not only an interference in the progression of adult development, but may also be a frustration of a deep need to reproduce one’s self and create the next generation. While medical treatment of infertili ty can lead to resolution through the conception and birth of a biological child for many couples, infertility treatment will not be successful. Therefore, the couples’ task becomes one of resolving this interference in this important life goal on a psychological and emotional level (Mahlstedt & Wood, 1995; Anderson & Alesi, 1997; Domar & Seibel, 1997). Infertile couples experience a variet y of emotions and reactions. These emotions and reactions can vary greatly across infert ile individuals and couples and may be greatly effected by a number of factors including, but not limited to, the individual’s or couple’s support system, religious belie f, and financial stability. Ambiguity revolving around the diagnosis and not knowing what is going to happen next along with being forced to change daily routines an d life plans are central themes to the emot ional experiences of many infertile couples who are trying to define and resolve the dilemmas cr eated by their infertility. Infertile couples have to develop ways in which to cope with the uncertainty that infertility creates as well as make decisions about tr eatment that will be in the best interest for their relationship and their hoped-for children. While infertility patients struggle with the uncertainty surrounding their infertility and eventual treatment outcomes, physicians specia lizing in reproductive endocrinology must also deal

PAGE 104

90 with the frustrations of not knowing w hat will happen next or how they should respond to their patients’ needs for answers to the psychosocial questions that they pose. It is fair to conclude t hat all participants in assisted reproductive technology have hope that medical goals will be achieved and desire a happy ending for all involved even though there is uncertainty regarding the duration or ramifications of the trea tment decisions (Menning, 1980; Mahlstedt & Wood, 1995; Anderson & Alesi, 1997; Domar & Seibel, 1997). Many studies investigating the em otional experiences of couples experiencing infertility are anecdotal, repor ting the comments and descriptions of the experience of couples interviewed. Perhaps one of the most compelling comments illustrating the poignancy of the ambiguity and emotional distress surrounding the experience of infertility follows: “When you absolutely cannot have child ren, it’s called sterility. When it seems to be taking an awfully long time but you still have hope, it’s called infertility. Infertility is worse” (OTA, 1988) Creating a “happy ending” may be viewed as the resolution of the emotional dilemmas created by infertility, as well as making good decisions about infertility treatment options. The impact of infertility and the resulting stress emanating from infertility diagnosis and treatment have been well documented in the research literature (Domar & Seibel 1997; Greil, 1997; Schneider, 2000). The stress of infertility on indivi duals and couples can have negative effects on physical and mental health, the couples’ relationship, career,

PAGE 105

91 finances, as well as social and family networks (Domar, 1997; Domar & Seibel, 1997; Schneider, 2000). I ndividuals and couples experiencing infertility treatment can f eel emotionally depleted, is olated from others, and vulnerable to experiencing a se ries of losses and chronic stress (Mahlstedt, 1985; Schneider, 2000). Emotional Aspects of Infertility The diagnosis and treatment of infe rtility may have a profound impact on the lives of those couples. The proce sses of diagnosis and treatment may lead to conflict in the most stable of relati onships or exacerba te existing problems between spouses. While this conflict has been shown to be greater when the treatment process is prolonged or unsucce ssful, the emotional distress begins to develop when a couple realizes that their plans to conceive and to begin their family are not coming to fr uition. Immediate conseque nces of infertility include worrying, having doubts about the realization of their family goals, and frustration. Daily routines change significantly for in fertile couples after medical treatment begins. These changes can take an exacti ng toll on the couples’ quality of life, affecting the emotional, social, physica l, occupational, in tellectual, and even spiritual well being of t hose involved (Seibel & Ta ymor, 1982; Cooper, 1993; Mahlstedt & Wood, 1995; A nderson & Alesi, 1997). According to Menning (1980), infertility is a process that consists of several different emotional phases beginnin g with a phase of initial shock leading to denial, depression, and eventually to resolution and acceptance of the infertility. The first phase that couple s typically experience when diagnosed with

PAGE 106

92 an infertility problem is one of shock and surprise. Because many couples carefully plan when to begin their family, the inability to conceive is generally unexpected. Most assume that they will have no di fficulty in conceiving and reproducing. In addition, society tends to support controlling fertility until children are desired. Therefore, little preparation ex ists for the possibilit y of infertility. Following the initial shock of an infert ility diagnosis, couples typically proceed through a phase of denial. Denial is espec ially characteristic for couples whose diagnostic evaluations reveal an absol ute and untreatable problem (Menning, 1980; Domar & Seibel, 1997). While most adults feel that they ha ve some control over the decisions affecting their lifestyle, fo r those couples seeking treat ment for infertility, the treatments superimpose an agenda that severely comp romises their ability to either enjoy the present or plan for the fu ture. Couples struggling with infertility tend to be highly motivated towards parenting and readily forego many other positive aspects of life in pursuit of c onceiving and having a child. In addition, couples seeking infertility tr eatment are usually self-direc ted, in control of their lives, and have committed to the ethic that hard work leads to success. For these individuals, infertility can be a devas tating experience. Regardless of the financial burden felt by infertility treatm ent, the lack of children makes everything that they work for seem meaningless to some infertile couples. These feelings lead to the next emotional phase of anger that couples generally encounter through the infertility process. Even t hough an infertile couple may attempt to control their infertility by seeking di agnostic testing and therapeutic procedures,

PAGE 107

93 powerlessness is a major feature of t he infertility experience because only a small percentage of those who undergo treatment for infertility achieve pregnancy. The reality is t hat the outcome of infert ility treatment is beyond the couple’s and physician’s control. T he unanswerable question “Why?” plagues infertile couples while their anger is f ueled by the conflicting demands of work responsibilities and infertility treatment as well as any perceived insensitivity of friends, family, and even their physicians. Independence and flexib ility that were once in the control of the couple are lost as their lives begin to revolve around their physician’s plans for conception including medications, injections, ultrasound examinations, and surgeries. The couples’ attention becomes focused on conceiving, and other goals, pr iorities, and needs are neglected. The focus on medical treatments and any subs equent failures can have a negative impact on their self-esteem, health, rela tionships, security, and even their ambitions. Each month of treatment c ontains hope for two weeks and despair for two weeks with each failure to conceive, cr eating a roller coas ter of emotion with hope on the upside and depression, anger, and guilt on the downside. In addition to the uncertainty of infertilit y and treatment outcomes, given the fact that couples must surrender much of thei r control over their bodies as well as their plans for a family, anger is a predi ctable response. According to Menning (1980), even the best relationship betw een the physician and the patient does not alleviate all of the fr ustration, helplessness, and embarrassment that may be experienced by the couple. The anger experienced by couples may be rational, focusing on real and correctly perceived insu lts, such as societal pressure to

PAGE 108

94 reproduce, or the pain and inconvenience of infertility test s and treatments. However, the anger experienced by couples may also be irrational where anger is projected against others such as abor tion-rights advocates, people who “breed like rabbits”, or those who neglect, abuse, or mistreat children. In addition, infertility patients may projec t their anger onto those who they feel have control over them such as their physician or nurses providing the in fertility treatment. According to Menning (1980), this irrational anger is usually a front for intense pain and grief that cannot yet be ack nowledged by the person (Menning, 1980; Seibel & Taymor, 1982; Imeson & Mc Murray, 1996; Domar & Seibel, 1997; Anderson & Alesi, 1997). Feelings of guilt contribute to the secr ecy that so often surrounds infertility. Infertility patients often feel guilty about t heir inability to conceive or impregnate, for letting down their spouse, families, and other loved ones, as well as for the emotional responses they have to the infertility diagnosis and treatment. Infertility patients often feel responsible fo r their infertility. Many couples view infertility as a “punishment” and search for a reason on which to blame their infertility. Couples have identified a variet y of reasons for whic h they attribute the punishment of infertility including premarital sex, the use of birth control, a history of abortion or impregnation, venereal disease, extramarital affa irs, masturbation, homosexual thoughts or acts, and even se xual pleasure. Those patients who attribute their infertility to a particular deed or action in their history will typically go to great lengths to atone for these “sin s”. Atoning may take many forms from religious acts to personal denial or to wo rking in painful areas such as counseling

PAGE 109

95 unwed mothers. Most often, these feelings of guilt are generally resolved by the acceptance that people cannot control every aspect of their lives and that there is no relationship between fertility and wo rthiness (Menning, 1980; Mahlstedt & Wood, 1995; Anderson & Alesi, 1997; Domar & Seibel, 1997). Depression is the emotional state t hat permeates all other experiences and is common among couples during infert ility evaluation and treatment. Even when the infertile couple has reason to be optimistic about the success of treatment leading to pregnancy, this hope is usually coupled with feelings of depression that generally intensifies wit h each failure. Infertility involves experiencing a series of losses whic h have been found to be of greatest importance in the etiology of depression. Some of the losses felt by the infertile couple are resolved by medical technology or through other alternatives such as adoption or the use of donor gametes. One important contributing factor to depression is the loss of a relationship with an emotionally important person due to death, divorce, the waning of affection, or separati on. Because infertility is often a difficult subject for most to discuss due to the very personal and inherently sexual nature of it, couples may tend to keep their infertility secret from others. This secrecy cuts couples of f from potential sources of support and comfort, which leads to feelings of isolat ion. Without children, infertile couples are inherently excluded from important social networks, social events, and parenting experiences. The social isolati on experienced by infertile couples can have a major impact on their lives. In extreme situations, couples faced with infertility may be so sensitive to issues of pregnancy or little children that they

PAGE 110

96 withdraw from all social or work relat ed situations that might lead to such a contact. While the isolation of couples experiencing infertility from others in family, work, or social environments may be difficult, even mo re devastating is isolation from their spouse. This isolat ion generally results from a breakdown in communication and erodes the family dynam ic leading to increased marital stress. The partners themselves may experience a loss of closeness simply because the infertility diagnosis and treatment affects them differently or because they cope differently. For wo men, the potential loss of fertility is generally more threatening than for men. Typically, women are more emotionally expressive, willing to discuss their anxiety and depre ssion with their spouse, who may often feel powerless to help or unsure of how to respond. Men, however, typically contain their emotions, not only to sustain the stoicism that th ey feel is expected of them, but also out of a sense of responsibility to be the stable and calming force in the relationship (Sherrod, 1995; Schneider, 2000). Unfortunately, women may misunderstand this basic difference, often leaving women to feel unloved or abandoned when their husbands do not experience the same kinds of feelings of loss or emotional pain (Menning, 1980; Cooper 1993; Mahlstedt & Wood, 1995; Imeson & McMurray, 1996; Domar & Seibel, 1997). Another important etiological factor contributing to depression in adulthood is the loss of health, important body func tions, or acceptable self-image or body image. Most couples seeking medical tr eatment for infertility are in good health and have positive body and self-images. Diagnostic procedures and infertility treatments can assault pos itive images of oneself. Patients relinquish their

PAGE 111

97 bodies over to medications that may have severe or debilitati ng side effects. Their sexual spontaneity all but disapp ears and is replaced instead by sexual schedules with the frequency and timing of intercourse prescribed by their physician. Additionally, the amount of time spent in physician appointments or in decisions to proceed with surgical treatment options may threaten the individual’s sense of health. Individuals accustomed to feeling in control of their health, bodies, and lives may suddenly feel that they are defective because they are unable to perform the most fundamental task of life, that of reproduction (Mahlstedt & Wood, 1995; A nderson & Alesi, 1997). In addition to the loss of a relationship and sense of health, the loss of status or prestige, self-esteem, self-c onfidence, and security all contribute to feelings of depression. The experiences of infertility can invoke feelings of loss of these types in varying degrees among couples. Because society places great value on parenthood, infertility can lead to the perception of lo ss of status or prestige in the eyes of others. The expecta tion that infertile couples feel to have children can leave infertile couples feeli ng different, abnormal or less acceptable. Because they are unable to meet societal expectations to conceive and raise children, they suffer a social stigma a ttached to the infertility (Cooper, 1993; Mahlstedt & Wood, 1995, Domar & Seibel, 1997). Some infertile individuals feel that their sexual identities are inextricably linked to reproduction. These feelings are reinforced socially through notions that masculinity, virility, fertility, and potency are equat ed. Infertile men often report feelings of emasculation which deals a considerable blow to their sexual

PAGE 112

98 self-image. Some infertile men with no history of sexual dysfunction or inadequacy become impotent for a period of time after the discovery of male factor infertility. Among women, the idea of motherhood is usually linked to their adult identity. Infertility may cause fee lings of inadequacy and many women feel that they are unlovable, unfeminine, damaged, or defective. For both men and women, sexual functions may become sy nonymous with failure, highlighting the failure to conceive and, therefore, the failure to be an adequate man or woman. These negative emotions deepen as infertility continues and are often generalized so that their gl obal self-esteem is affected. The loss of self-esteem or pride generally leads to a sense of failure Infertility is the inability to conceive after one year of unprotected intercourse or the inability to carry a pregnancy to a live birth. By definition, infertility is a description of failure. Although the emotional responses that accompany infertility can vary greatly among individuals, they are generally very st rong and usually come as a surprise to many infertile couples. When an indivi dual or couple experiences a strong and unexpected emotional response, it is common for self-esteem to erode as they become so poignantly aware of the extent to which emoti ons appear to rule their lives as never before. In addition, th is damage to self-esteem is not easily repaired and usually affects the individual’s deepest sense of self, including their feelings about their own masculinity, femininity, and sexuality (Cooper, 1993; Mahlstedt & Wood, 1995; Anderson & Alesi, 1997; Domar & Seibel, 1997). The time and expense involved in the treatment of infertility can result in the loss of occupational, fi nancial, social, and cultural security. Occupational and

PAGE 113

99 financial security is especially vulnerabl e when infertility is prolonged, requiring frequent visits to the physician, expensiv e medications, and costly diagnostic or surgical procedures that often interf ere with work schedules. Because most insurance companies cover only part, if any at all, of infertility diagnosis and treatment procedures, most couples ar e burdened with the financial strains that infertility creates. Job performance as we ll as job security and relationships with coworkers can be jeopardized. Social and cu ltural security is endangered if the infertile couple experiences feelings of is olation from others whose lives center to a large degree on their children (Mahl stedt & Wood, 1995; Domar & Seibel, 1997). Once all hope for achieving pregnancy and a live birth is abandoned, the most compelling emotional reaction to infe rtility experienced by many couples is grief. However, the grief experienced by c ouples as a result of infertility is unlike any other grief experience. While grief usually follows the loss of something or someone important to an indi vidual, the grief experie nced by an individual resulting from infertility is based on some thing intangible. It revolves around the loss of a potential, not an act ual life. For many couples, infertility is the loss of a dream or the hope of fulfilling an important fantasy or life goal. These infertile couples yearn for the child who may nev er be and mourn for t he child who never was. To these infertile couples, this loss is as a real as if the child had been born, lived, and died. Infertility that is c onclusive with no potential of conceiving or having biological children represents losse s of many types. Fi rst, it the loss of one’s fertility and all that it means in rela tion to an individual’s sexuality. In

PAGE 114

100 addition, it is a loss of the pregnancy exper iences itself. Menning (1980) reports the loss experienced by one infertility patient: “Death. Death of a lot of things. It is the end of the Bowes family and the Bowes family name. It dies with us because of me. My husband is the last of t he male children in his family. Death before life… before we even knew our ch ild, because he never existed. The hardest part of this kind of deat h is that it is the death of a dream. There are no solid memories, no pictures, no things to remember. You can’t even remember your child’s blonde hair, or brown eyes, or his favorite toys or the way he laughed, or the way it felt to be pregnant with him. He never existed.” Although society has elaborate rituals to comfort the bereaved in the death of a loved one, none of these rituals are availa ble for those experie ncing the grief of infertility. In addition, because there is no tangible loss for all to see, family and friends may never even know of the grie f experienced by couples, leaving these couples to grieve for their loss alone. Generally, infertile couples feel wounded and those wounds of hurt and mourning ar e often reopened. Children remind them that they are infertile. Pregnant women remind them of the pregnancy that they were never able to experience. Te levision commercials featuring toddlers, birth announcements from friends, and fam ilies with children in Sunday worship services can all contribute to the insult and grief felt by infertile couples (Menning, 1980; Mahlstedt & Wood, 1995; A nderson & Alesi, 1997).

PAGE 115

101 The final emotional stage that infertile couples experience is the stage of resolution. When an infertile couple final ly reaches the stage of resolution and acceptance, plans for the future can be made, building a way around the obstacle of infertility. Couples are ready to select alternative life plans including decisions about adoption with confidence. With the ambiguity of infertility past, couples are ready to once again exert contro l over their life plans and to proceed with their lives. Unfortunately, some in fertile couples or individuals may never reach acceptance or resolution. For so me the loss and the pain of infertility will be felt throughout their lives (Menning, 1980; Mahlstedt & Wood, 1995). Conceptualization of Stress in Infertility Research The emotional experiences of in fertility have been conceptualized in a variety of frameworks (Schneider, 2000). During the 1940s and 1950s, both medical and psychiatric res earch literature contained several case reports of infertile men and women who sought psych iatric care for depression, obsessivecompulsive behavior, or neurosis. As a result, a theory of the psychological etiology of infertility known as the p sychogenic morbidity model or psychogenic hypothesis evolved. The psychogenic m odel was clearly the dominant theory up through the mid-1980s (Bernstein, 1993; Gr eil, 1997). This theory postulated that female infertility, particularly t hose cases with unidentified organic causes, are caused by an unconscious resist ance to motherhood on the part of the female. The theory asserts that males with idiopathic infertility exhibit extreme animosity and aggressiveness toward their wives, thereby preventing conception. As described by Rutherford et al. in 1966, the psychogenically infertile male is

PAGE 116

102 typically from an above-average educatio nal background, with a domineering mother who is sexually unsatisfied hersel f, the experience of a childhood threat to have love withdrawn for being “naughty” such as exhibiting any sexual behavior, lack of sexual activity befor e marriage, and conflict within marriage between sexual desire and fear of offending their partner (Bernstein, 1993; Greil, 1997). One major contributing factor in the development of the psychogenic hypothesis is that during this time, reproduc tive endocrinology was in its infancy. While approximately 50% of infertility pati ents received a diagnosis identifying the etiology of their infertility, the rema ining 50% of patients were diagnosed with idiopathic infertility. Theref ore, psychiatrists incorrectly concluded that in patients diagnosed with infertility of unknown origin, the etiology of the infertility must be psychogenic. In addition to the lack of understanding of reproduction as well as the etiology of infertility, the foundation for the develop ment of the psychogenic theory can be found in the work conducted by Helena Deutsch during 1945. In her writing, The Psychology of Women, Deutsch identified five types of women who she believed caused their own infert ility: (1) the infantile, dependent woman looking for attention, (2) the woman whose motherliness is “consumed in a fire or erotic love…or devotion to an ideology ”, (3) the woman who exhausts her motherliness on her husband and know s that he would not welcome the competition of a child, (4) the mascu line, aggressive woman who refuses to accept her properly passive role in so ciety, and (5) the emotionally disturbed woman who fears the additional burdens a child woul d create. At the conclusion

PAGE 117

103 of the 1950s, these five types of women had become t he standard for the examination of emotional issues in infe rtility (Deutsch, 1945; Rutherford et al., 1966; Bernstein, 1993). During the 1960s and 1970s, a new formulation and insight into the theories of the role of st ress on infertility began to develop. Psychologists, social workers, nurses, infertility specialists, and psychiatrists began to research the relationship between emotional distress and in fertility. This effort was greatly advanced with the application of several neuropsychologi cal tests including the Minnesota Multiphasic Personality Invent ory (MMPI), Rorschach, Neuroticism Scale, and the Thematic Apper ception Test (TAT). Results of this research indicated that there were no differences in test performance between the infertile couple and the fertile population. Resear ch also demonstrated that the distress experienced by couples related to infe rtility was not sufficient to manifest psychiatric symptomatology. In addition, the treatment of psychiatric symptoms among infertility patients experiencing di stress and anxiety did not change their infertility status. However, psychiatric treatment was shown to improve patients’ sense of well being. These studies led to mounting evidence refuting the psychogenic hypothesis (Bernstein, 1993). Some researchers have begun to modify the psychogenic model with another theory based on the idea that stress causes infertility. The importance of the relationship between stress and health as well as the suspicion that stress may be causally linked to infertility is not unreasonable and supports the modifications of the psychogenic model. However, this modified view has not

PAGE 118

104 become popular with researchers, counselor s, or infertile couples themselves because the psychogenic hypothesis is gener ally regarded as a mechanism for minimizing the reality of the distress associated with infertility and also as a means for casting blame on infertility patient s for their own suffering. To infertility patients, as well as to those who sympathize with them, the psychogenic hypothesis seems like a reiteration in sc holarly guise of well-intentioned advice such as “Relax” or “Take a vacation and you’ll get pregnant in no time” (Greil, 1997). While most researchers today re ject the psychogenic hypothesis, some attempts have been made to revive this theory as conceptual framework for investigating the relationship between stre ss and infertility. According to Astor and Pawson (1986), personality m easures do not adequately portray unconscious motivation. Therefore, they conclude that these measures do not contest the psychogenic hypothesis (Greil, 1997). Barbara Eck Menning is recognized as in strumental in the development of a cognitive construct for describing the emot ional aspects of infertility for couples, how couples process the diagnosis, and how these couples proceed toward resolution of the psychologic al distress that couples experience as a result of infertility. The crisis model, which wa s first described by Kubler-Ross in 1980, was subsequently refined through the work conducted by Menning and others in order to be applied appropriately to infertility. Loss is a central theme in the crisis model, incorporating the st ages of grief. The losses described by Mahlstedt and Wood (1995) include: the loss of a relations hip or potential relationship; the loss of one’s health, the loss of status or pr estige, the loss of self-esteem, self-

PAGE 119

105 confidence, or security; the loss of a f antasy or hope of fu lfilling an important fantasy; and the loss of something or so meone of great symbolic value. This model recognizes infertility as a disrupti on in the normal equilibrium, which taxes couples’ existing resources and threatens lif e goals, and recognizes that infertility may also awaken unresolved psychological problems. This conceptualization of infertility acknowledges the emotional roller coaster couples experience throughout the process. In addition, this model acknowledges that couples may experience a variety of emotions and behav iors at different times and in a different sequence from other infertilit y patients. These emotions include surprise, guilt, denial, anger, a feeling of is olation, grief, and finally, a sense of resolution or acceptance. In addition, it is recognized that infertility, unlike other life crises, is not bound within a given time frame. Infertility is viewed as processional, encompassing a series of rela ted crisis events before resolution is reached. Therefore, infertility is concept ualized as both a major life event as well as a prolonged crisis with a series of daily hassles that may potentially impact all aspects of the infertile couple’s life (Blenner, 1990; Mahlstedt & Wood, 1995; Imeson & McMurray, 1996; Schn eider, 2000). Similar to any other life crisis, resolution could either be maladaptive change or an increase in maturity and emotional strength. The devel opment of this cognitive construct of infertility led physicians, for the first time to consider infertility patients’ emotional needs simultaneously with their physical and medical needs. It was at this time that the concept of infertility physicians join ing with appropriate mental health professionals began to evolve (Bernstein, 1993).

PAGE 120

106 In addition to the work conducted by Menning, several other researchers have been instrumental in the expansion of the descr iption of the emotional experiences of couples dealing with infe rtility and the refinemen t of the crisis model as it applies to infertility. Resear ch has expanded the literature to include the intrapsychic conflicts that infertility ar ouses, the effect of infertility on marital relationships, gender differences in the response to infertility, common coping strategies employed by infertile women, and the aspects of loss of control as well as the negative effects infertility has on self-image and generational identity (Bernstein, 1993). The Reproduction Filtering Model (som etimes referred to as the adaptive reproductive failure model) has also been deve loped by evolutionary biologists to explain why the rates of reproductive failu re are so high in mammals. Adherents of this theoretical conceptualizatio n of the relationship between stress and infertility argue that the high cost of r eproduction in terms of time, energy, and risk has naturally selected for physi ological mechanisms that terminate reproductive attempts when the likelihood of reproducing viable offspring is low. Therefore, this response conserves time and energy for reproduction that is most likely to succeed. This model implies t hat the reproductive system has evolved a high physiological responsiveness to env ironmental change. The application of this model to humans is based on the ar gument for the likely effectiveness of acute and long-term environmental therapy such as diet, stress reduction, or psychosocial therapies as treatment fo r some forms of r eproductive failure (Wasser et al., 1993).

PAGE 121

107 The Stress and Infertility Relationship Research has established that many endocrine systems respond to stress. As originally pr oposed by Selye, stress is a nonspecific response of the body to any demand. This paradigm assert s that any kind of stressor provokes the same kind of response in every indivi dual. This response was believed to be dependent only upon the time a stressor coul d influence an organism. However, research over the past several dec ades has demonstrated that different stressors, whether they be physical or p sychological, can result in different responses by a system and the psychol ogical mechanism of anticipation determines the effect of the stressor. Furthermore, evidence suggests that individuals may respond differently to the same stressor. It has also been demonstrated that in the stress respons e, there can be dissociation between emotional and endocrine states. In addition, research has found that in the stress response, a dissociation between t he different endocrine systems often occurs suggesting that different parts of the endocrinological st ress response are linked to specific psychological stressors or to the effectiveness of psychological functioning such as cortisol to ineffect iveness of defenses (Selye, 1956; Mason, 1975; Ellertsen et al., 1978; Weinberger et al., 1979; Vaer nes et al., 1982; Allen et al., 1985; Demyttenaere et al., 1989). The importance of coping mechanisms is also established by the scope of the research dedicated to the role c oping mechanisms play in psychological functioning. Current stress and coping rese arch focuses primarily on the stressreducing effects of coping skills. Althoug h coping may lead to a reduction in the

PAGE 122

108 stress experienced by an individual, the ex ecution of coping ski ll or strategy may serve as a source of stress and therefore may activate some dimensions of the stress response (Demyttenaere et al., 1991). The Role of Psychoeducational Interventions in Infertility Cognitive therapy is in creasingly being accepted as an effective treatment option for depression (Domar et al., 2000). Several studies with cancer patients combined support with cognitive-behavioral techniques. The benefits of this educational model included decreased psychol ogical distress, longer life span, and decreased mortality (Helgeson & Cohen, 1996; Domar et al., 2000). Cognitive-behavioral appr oaches have been shown through a multitude of studies and investigations to be effect ive in reducing sympt oms of depression and decreasing health costs for patient s with a wide variety of conditions including cardiac, abdominal, orthopedic dental surgery, and invasive medical procedures (Mandle et al., 1996; Domar et al., 2000). Another study of patients with multiple sclerosis emphasized c oping-skills training. This study demonstrated greater advant ages in well-being and co ping for the patients receiving the coping skills training than fo r those who participated in only a peer telephone support group. Several principals form the foundati on of cognitive behavioral treatment: Unconditional regard for se lf and others involves the acknowledgement and acceptance of an individual’s strengths and limitations. While all humans are predisposed to self-rating and rating of others, the objective in cognitive behavioral therapy is to

PAGE 123

109 minimize the rating and judgment of self and others and to control the manner in which it occurs (Ellis, 1977; Granvold, 1994). The continuity assumption is the presumption that covert behaviors are subject to the same rules of learning as overt behaviors. Therefore, cognitions can be t aught like overt behavior is taught through techniques such as modelin g, coaching, rehearsing, and a variety of other methods. Comp lex cognitive transactions can be broken down into smaller “chun ks” and learned incrementally. Cognitions can variably function as a stimulus, response, or consequence and can be strengthened, weakened, or extinguished through reinforcement (Mahoney, 1974; Granvold, 1994). Conditioning is a term referring to the individual l earning history and associated attitudes that influenc e an individual’s process of anticipation, selection and decisio n-making in a given situation (Schindler & Vollmer, 1984; Granvold, 1994). Self-responsibility for one’ s own emotional disturbance and maladaptive behavior. Questioning and the Socratic Me thod to promote desired changes (Beck & Emery, 1985; Granvold, 1994) Couples participating in cognitive-behaviora l therapy are typically introduced to a wide variety of techniques, including relaxation-response training, cognitive restructuring, emotional expression, and nutrition and exercise education relevant to infertility (Domar et al ., 1990, 2000; Connolly et all, 1993).

PAGE 124

110 Based on mounting evidence, repr oductive endocrinologists have begun to acknowledge the import ance of addressing the emot ional needs of their patients simultaneously with their physica l and medical needs. However, several factors have limited the acceptance and application of pr ograms aimed at educating patients about stress-reduction t herapies as compared to the rapidly expansion of the biomedical technology: 1. Few studies have empirically cha llenged the belief of physicians that stress is not causally linked with infertility to a significant degree; 2. Psychosocial problems tend to be more difficult for patients to acknowledge than biological problems, 3. Psychosocial therapy, patient educa tion programs, and social support networks are relatively more time consuming and require considerably more effort on the part of the patient as comp ared to the biomedical treatments, and 4. Reproductive endocrinologists are not trained to diagnose psychological problems or to prov ide the therapies beneficial to mediate the effects of psychosocial stress. The combination of these factors has l ed reproductive endocrinologists to rely primarily on biomedical treatment s for infertility in this highly competitive field of practice. Unfortunately, a purely biotec hnological approach may be successful in treating only the physiologic symptoms of infertility patients, without eliminating the stressors that triggered or accompany the symptoms. However, biomedical

PAGE 125

111 treatment for infertility along with treat ment for psychosocial stress could, arguably, markedly improve ov erall reproduction outcomes (Wasser et al., 1999). Based on the mounting evidence about t he stress of infertility and infertility treatment, some investigators have suggest ed that infertility treatment programs should incorporate a psychological tr eatment component. As early as 1959, acute psychological supports for infert ile couples as an adjunct to medical treatment can be found. Recently, national organizations such as Resolve, Inc. in the United States and the National A ssociation for the Childless in the UK offering referral support groups and infertility counseling have been established. These programs can be provided to pat ients in a variety of ways including traditional psychotherapy, group therapy, and social support networks (Anderson & Alesi, 1997). Traditional individual psyc hotherapy interventions typically focus on the similarities in psychological response s between infertility and grief counseling. Counseling is particularly beneficial to infe rtile couples at three times during the treatment cycle: 1) t he beginning of an infertility evaluation, 2) when a psychiatric disorder is present, and 3) t he termination of unsuccessful treatment. These interventions generally address the long-term adjustment to infertility as well as the acute reactions to failed tr eatment attempts. Two main approaches in traditional interventions are: brief s upport counseling and therapeutic counseling, each of which may be offered to the indi vidual or to the couple. The main components of a brief suppor t program include empat hetic listening, patient education, and problem solving. Becaus e infertility is a significant loss to

PAGE 126

112 couples, infertility patients often ha ve a great need to talk about their experiences. The role of the counselor is to facilitate the expression of fears, hopes, disappointments, anger, and sadness. This free expression is designed to facilitate further communication betw een the couple and to increase their understanding and tolerance of each ot her. In addition to facilitating communication, infertility patients c an derive benefit from education about common emotional reactions to infert ility. Finally, traditional counseling approaches can help patient s to conquer a range of problems. Everyday problems can become compounded by infertility. Patients often become overwhelmed with questions such as w hat to tell their employer about their frequent absences, how to respond to genuine questions about whether they have children, and how to preserve a relati onship with a sister or friend who is pregnant. Systematically working out i ndividual patient concerns and problems can help to reinstate a sense of control ov er their lives (Anders on & Alesi, 1997). While brief support counseling pr ograms are beneficia l to patients experiencing acute distress related to t heir infertility diagnosis or treatment, the aim of most therapeutic counseling is to help patients reach resolution in terms of their infertility. The concept of reso lution within the context of therapeutic counseling implies that it is possible for an infertile couple to reach a point where infertility is not associated with emotional di stress. The goal for infertile couples in this approach is to accept and incorporat e the pain of infertility into their life so that it loses the acute qual ity experienced in the earlie r stages of diagnosis and

PAGE 127

113 treatment. The durat ion of therapeutic counseling varies and can be spread over many years (Anderson & Alesi, 1997). Couples’ approaches typically address the negative impact of infertility on marital adjustment and sexual satisfaction. Couples’ approac hes use therapy as a means of improving communication and interactions between the spouses. Group therapy has been described as a valuable venue for couples experiencing infertility because so many of them report feelings of isolation. While reading material can be beneficial to infertility patients, one of the most powerful experiences is through support groups. It is generally believed that patient education through contact with others who ar e also infertile can restore a sense of normality, and begin to break down feel ings of isolation and helplessness. Groups demonstrate that infertile couples are not alone. In addition, they can share thoughts, problems, experiences, and concerns with others in similar situations. Participants can also lend support to other mem bers of the group, which is often beneficial and rewarding. Group therapy can lead to improved communication with physicians as pati ents become educated about special needs and learn to express their frustrati ons about medical interventions and it is a cost effective way to address infert ility patient’s emoti onal needs (Anderson & Alesi, 1997; Domar & Seibel, 1997; McNaughton-Cassill, 2000). Unfortunately, few studies address the pr ovision of therapy for couples in the midst of their infertility procedures or the use of group therapy approaches for the treatment of such couples (Dom ar & Seibel, 1997; McNaughton-Cassill, 2000). However, the value of psychoeducat ional interventions for individuals

PAGE 128

114 undergoing treatment fo r other devastating illnesses, such as cancer, has long been documented to minimize both adverse psychological and physiological side effects. Behavioral techniques incl uding the relaxation response, stress management, nutrition, and exercise c ounseling have been widely successful when applied to a variety of medical and psychiatric symptoms. The first application of these programs to infertile women in a clinical setting was in 1987. In this program developed and implement ed by Domar, Seibel, and Benson, infertile women were educated in a variet y of ways to elicit the relaxation response, “mini” relaxation techniques, cognitive restructuring, nutrition and exercise counseling, and methods for dea ling with negative em otions (Domar & Seibel, 1997). The most common interventions to help couples cope with the stress of infertility in the United States include patient education and social networking through support groups. While the pr ograms vary, some common elements among all of them include educationa l programs and emotional supports designed to alleviate symptom s of stress for i ndividuals or couples experiencing infertility. One such program is the Mind/Body Program fo r Infertility. Based on research regarding the emotional aspects of infertility, the Mind/Body Program for Infertility was developed and implemented in September of 1987. The foundation of this program lies in the assumption that women who regularly elicit the relaxation response will demonstrate decr eased levels of tension and anxiety leading to an increased rate of concepti on. The relaxation response has been described as an innate physiologic respons e that is the c ounterpart of the

PAGE 129

115 physiological response to stress referred to as the fight-or-flight response. The relaxation response is associated with a se t of coordinated physiological changes including a decrease in oxygen consumpt ion, carbon dioxide elimination, heart rate, respiratory rate, blood pressure, muscle tonus, and arterial blood lactate. In addition, an integrated hypothalamic response appears to lead to a series of physiologic changes that are consistent in the sympathetic nervous system activity. The relaxation response can be induced through a wide variety of activities including progressive muscle relaxation, diaphr agmatic breathing, meditation, repetitive prayer, mental imagery, exercise, or absorption in a pleasant task. Furthermore, the several benefit s of the relaxation response have been documented including decrease in chr onic pain, hypertens ion, preoperative anxiety, ventricular arrhythmias, and anxiety. The Mind/Body Program for Infertility, established at the Deaconess Ho spital in collaborat ion with the Division of Reproductive Endocrinology at Beth Is rael Hospital, is a ten-week educational program including relaxation-response training, stress management training, exercise, nutrition, group support, coupl es’ cognitive-behavioral training, and developing self-empathy and compassion. While this training focuses primarily on women participating in ART procedures, regardless of the etiology of the infertility, husbands are invited to attend tw o of the ten sessions. The main topics for the 10 sessions are as follow: 1. An introduction to the physiology of stress, the relaxation response, and the relationship between stre ss and the reproductive system;

PAGE 130

116 2. The use of diaphragmatic br eathing and mini-relaxation-response exercises; 3. Cognitive restructuri ng and affirmation training; 4. Developing self-empathy and compassion; 5. Exercise and nutrition; 6. Mindfulness, specifically an incr eased awareness of sensations and perceptions; 7. Emotions; 8. Anger and forgiveness; 9. Yoga, exercise, and couples’ cogn itive-behavioral exercises; and 10. Review and follow-up information (D omar et al., 1990; Domar & Seibel, 1997). This program follows a 10-week protocol designed to treat individuals experiencing any medical symptoms c aused or aggravated by stress and is based on the elicitation of t he relaxation response exercise In addition to this program, cognitive behavioral therapy is also accepted as an effective treatment option for depression among infertility i ndividuals and couples (Domar et al., 1990, 2000; Connolly et al., 1993). Summary of Research Examining the Re lationship Between Stress and Infertility The psychological effect of infert ility experienced both individually and as a marital couple has now been widel y documented through the research literature. Because ART procedures such as IVF and ICSI are the most complex, expensive, and invasive forms of infertility treatment and are generally

PAGE 131

117 the final treatment option for many patient s, much of the research focuses on infertility patients participating in thes e treatment program s. Depressive symptoms are common in infe rtility patients undergoing IVF. Using the Schedule for Affective Disorders and Schizophrenia, Bell (1981) found that 40% of infertile women had mild to moderate symptoms of depression while 7% presented severe symptoms of depression. Fr eeman et al. (1983) found that 16% of women preparing to participate in an IV F treatment cycle had scores of 70 or higher on one or more subscales of the MMPI. Likewise, Ga rner et al. (1984) found that depressive symptoms were present in 34% of infertile women prior to an IVF treatment cycle and in 64% of t he participants after an unsuccessful IVF cycle. In a prospective, longitudinal study of 59 infertile women seeking treatment, 9% of the infert ile participants met the crit eria for a major depressive episode as compared to 3% of the parti cipants in a control group (Downey et al., 1989). In addition, one-half of the infertile participants reported changes in their sexual functioning while 75% reported c hanges in their mood. Others studies have documented similar findings. Wright et al. (1989) found that infertile women were significantly more distressed than control participants on the majority of psychological parameters investigated. In one investigation conducted by Freeman et al. (1987), half of their sample of infertile couples described infertility as the most upsetting experience of thei r lives. In another study conducted by Mahlstedt et al. (1987), 80% of the sample of infertile couples reported that their experience of infertility was either st ressful or extremel y stressful. The acknowledgement that psychological dist ress is one outcome of infertility and

PAGE 132

118 infertility treatment has led many res earchers and authors to advocate for the provision of counseling as an essential component in ART repertoires (Mahlstedt et al., 1987; Dennerstein & Morse, 1985; Connolly et al., 1993). Accumulating evidence suggests that st ress may influence the outcome of infertility treatment. While results of thes e investigations have led to conflicting conclusions, prospective, longitudinal studies investigating the relationship between stress and ART treatment outcomes i ndicate that the higher levels of negative psychological sym ptoms reported by fema le IVF patients may be associated with lower conception rates. Salvatore et al. (2001) conducted a study comparing the psychopathology personality features, and marital relationships of women undergoing IVF with those of a control group of patients as well as comparing infert ility patients participating in their first IVF treatment cycle with those who were participating in their second or more IVF treatment cycle. Results of this study indicat ed that while there were no significant associations between any special psyc hopathological, personality, or marital characteristics and pregnancy rates, wo men who achieved pregnancy through IVF showed a statistically significant di fference in their less exclusive closeness with partners and their search for friendships and social contacts. Demyttenaere et al. (1998) found in a series of studi es that pre-cycle levels of negative emotions were significantly associat ed with lower pregnancy rates. Further analysis revealed that women with a lower score (i.e., fewer negative emotions) than the median had a 31.3% chance of conceiving as compared to a 14.9% chance of conceiving for women who had a higher score than the median. A

PAGE 133

119 similar study conducted by Boivin and Ta kefman (1995) also supports the theory that stress is related to IVF outcome. Their results demonstrated that women who did not conceive as a result of the IVF cycle reported higher levels of stress during specific stages of the treatment cycle and had a poorer biologic response to treatment than those wo men who did conceive. Thiering et al. (1993) found that women starting their second or mo re attempt (referred to as veterans) scored significantly higher than those women initiating th eir first attempt with IVF (referred to as inductees). In additi on, the veteran women had a significant association between depression levels and conception rates during the one-year follow-up period; 13% of depressed women conceived as compared to 29% of nondepressed women. The results of tw o studies conducted by Sanders and Bruce (1997, 1999) provided similar result s demonstrating a significant difference among women participating in an IVF-ET tr eatment cycle in the levels of state anxiety between those who conceived and those who did not conceive. CreachLe Mer et al. (1999) reported that an incr ease in increments of 10 points on an anxiety scale was associated with a twofol d increase in the probability of failure for conception, leading the authors to assert that the treatment of anxiety in IVF patients should increase conception rates. Additional studies support these findings including studies conducted by K ee et al. (2000), Smeenk et al. (2001), and Klonoff-Cohen et al. (2001). Kee et al. (2000) found that among women entering an IVF program in Korea, those who failed to conceive had significantly higher pretreatment levels of anxiety and depression than those women who did conceive. Similarly, Smeenk et al. (2001) and Verhaak et al. (2001) found a

PAGE 134

120 significant relationship between baseline psychological factor s and the probability of conception. Additionally, KlonoffCohen et al. (2001) found a significant relationship between baseline measures of psychological state and subsequent pregnancy rates. Gallinelli et al. (2001), in an investigation of the influence of anxiety as well as coping skills on conc eption rates, found that anxiety state score was significantly lower in patients achieving implantation than in the group who failed to conceive. In addition, women who had successful implantation had significantly lower systolic blood pressu re and heart rates during the stress test than those women who failed to achieve implantation. Researchers approaching the investigat ion of the relationship between stress and ART outcomes through meas urements of psychoendocrinologic responses have also found similar results. In a series of studies, Demyttenaere et al. (1991, 1998) found that women with high anticipatory state anxiety levels and high anticipatory cortisol levels hav e lower pregnancy rate s. Similarly, Strauss et al. (1992) found a reduced probability for patients with psychological impairment to achieve pregnancy. Facchiniett i et al. (1997) found similar results. This study investigated a variety of variables as predictors for achieving pregnancy. The data provided evidenced a negative correlation between stress and outcome following an IVF-ET cycle. However, a number of in vestigators exploring t he relationship between stress and ART outcomes have reported findings that contradict the aforementioned studies. Lovely et al. ( 2003) recently conducted a study that measured cortisol and 6-sulfatoxy-mela tonin in women during an IVF treatment

PAGE 135

121 cycle. In addition to these measures of the physiologic stress response, participants completed the State-Trait Anxi ety Inventory (STAI). The authors of this study did not find any statistically si gnificant difference in any of the three variables in a comparison with a contro l group of donor IVF patients, leading to the conclusion that stress does not play a role in IVF outcome. In another study conducted by Harlow et al. (1996), while women who did not achieve pregnancy as a result of their IVF cycl e reported higher levels of anxiety during the follicular and pre-operative phases, these findings were not statistica lly significant. Merari et al. (1992) found similar results in t heir investigation of 113 Israeli couples initiating their first IVF cycl e. In this study, psychological assessments were conducted at three points in the treatment cycle: prio r to beginning the cycle, the morning of oocyte retrieval, and the morning of embryo transfer. Results of this investigation found no significant diffe rences between women who conceived and those women who did not conceive at any point in psychological assessments. In addition to these studi es, Slade et al. (1997) also found no significant relationships between emot ional state and conception rates. Furthermore, a recent study conducted by Csemiczky et al. (2000) also found no significant relationships between state anxie ty levels and IVF conception rates for women presenting with tubal infertility. Few studies address the provision of psychoeducational interventions for couples in the midst of their inferti lity procedures (Domar & Seibel, 1997; McNaughton-Cassill, 2000). In addition, a lack of systematic appraisal of the most effective form of p sychological counseling or support is an impediment to

PAGE 136

122 research. Traditional psychological treatment has included individual psychotherapy and support groups. While cl inical case studies point to the benefits of these approaches, we ll-controlled studies that support the efficacy of a traditional psychotherapeutic approach are very limited. In addition, while there have been anecdotal reports of an increase in conception rates, little convincing empirical evidence exists in the literat ure (Domar et al., 1990). Most studies focusing on infertility counseling are limit ed in both the size and scope of the examination and is based upon the asserti ons of counselors rather than any systematic evaluation of the problems experienced by couples and methods employed to resolve them. However, when investigations have been conducted, the benefits of psychoeducational intervent ions to infertility couples have been reported. In one report by Karahasanglu et al. (1972) examining the efficacy of the use of counseling, four major areas were addressed: 1) screening and initial evaluation of the couple, 2) relations hip improvement, 3) sexuality, and 4) supportive counseling. In this inve stigation, marked im provement in the spontaneity and frequency of marital sexual intercourse occu rred; attitudes became more positive; tension bet ween the couple diminished; husbands became more actively interested; and the sense of isolation was decreased in both the husband and wife (Sei bel & Taymor, 1982; Kar ahasanglu et al., 1972). Additional evidence concerning the t herapeutic effectiveness of counseling comes from investigations by Bresnick & Taymor (1979) who investigated the effects in 62 infertile couples of five to six sessions of counseling on emotional symptoms, relationship issues and the atti tudes to infertility. Results of this

PAGE 137

123 investigation demonstrated so me improvement in sympt omatology, leading the authors to conclude that in fertility counseling improves the quality of life for many dealing with the crisis of in fertility. However, Cooper (1979), investigating the effectiveness of a counseling support gr oup offered through Resolve, Inc. as compared to a control group receiving no tr eatment, reported different results. Measures of locus of control, self -concept and body-image were taken before and after a 15 week therapeutic interventio n. No differences between the two groups were observed. However, Cooper asserts that 15 weeks may not have been adequate time for adaptation to occu r (Bresnick & Taymor, 1979; Cooper, 1979; Edelmann & Connolly, 1986). Exam inations of the benefits of group sessions have also demonstrated a dec rease in the amount of distress and feelings of personal failure experienced by infertile couples when they were able to vent their feelings and anxieties with others who were dealing with infertility. Another report focusing on therapeutic techniques that included patient education, encouragement, and behaviora l techniques documented similar benefits (Abarbanel & Bach, 1959; McGuir e, 1975; Berger, 1977; Seibel & Taymor, 1982). Benefits of similar progr ams are documented in a report by McNaughtonCassill et al. (2000). A brief couples’ t herapy group program wa s implemented at Wilford Hall Medical Cent er in Texas as an option for couples undergoing IVF treatment. In this study, participants were recruited by the nurse coordinating their IVF cycle. The groups met twice a week for 1.5 hours for the duration of their treatment cycle with one or two facilitators: a psychiatrist and a

PAGE 138

124 psychologist. The groups used a Cognitive Behavioral format in which cognitions or thoughts are believed to play a major role in one’s evaluation of stressful events and perceived coping resources. Within this context, participants were encouraged to identify their own cognition s about their infertility and to explore the associations between their irrational beliefs and expectatio ns and emotional distress. In addition, participants were introduced to techniques for reframing attributions and generating alternat ive thoughts and solutions for common problems. A number of co mmon infertility-related t hemes including concerns about isolation, problematic interacti ons with family and friends, compromised marital relations, jealousy and inadequacy when confronted with normal fertility, religious faith, and the ethics and moralit y of IVF procedures emerged through the progression of t he program. Participants frequently remarked that talking to others going through similar experiences was reassuring. In addition, participants frequently wanted to compare notes about the procedures they were undergoing, including side effects of medi cation and the stress they experienced as a result of either giving or receivin g injections. Couples engaged in animated conversations about medical procedures, the outcomes of previous attempts, and how previous experiences related to their current experience. Couples shared information with each other regar ding national support networks as well as the specifics about adoption options. Another benefit resulting from this support group included the bonds fo rmed among the participants extended beyond the confines of the formal meetings into social interactions outside of the program. In addition to thes e benefits realized as a result of the program, overall

PAGE 139

125 results of this study found that both males and females valued the social support derived from the group and t hat the groups helped them to deal with the stress of IVF treatment. Couples also indicated that the program w ould be beneficial to others undergoing similar proc edures and that participating in the support group was crucial to the management of their stress associated with the IVF treatment (McNaughton-Cassill et al., 2000). In addition to these investigations, increasing evidence exists demonstrating that psychoeducational pr ograms may be efficacious in the treatment of the emotional effects of infertility and may lead to increased conception rates. Results demonstrate that in women wit h idiopathic infertility, the elicitation of the re laxation response reduces anxiety, depression, anger, and fatigue while increasing a sense of well-bei ng (Domar et al., 1990). Furthermore, the results of one study conducted by Domar et al. (1990) support the hypotheses that the reduction in stress incr eases the potential for conception. In addition to the documented and subjective improvement in psychological and physiological symptoms, 34% of the parti cipants conceived wit hin 6 months of completing the program (Domar, 1990). Similar findings were reported in a subsequent study conducted by Doma r et al. (1992) with participants demonstrating statistically significant reductions in every factor measured including anxiety, depression, and anger Again, 34% of the participants conceived within 6 months of completi on of this program. Among those who conceived, 37% attempting in vitro fert ilization conceived on their first attempt (Domar et al., 1992; Domar & Seibel, 1997).

PAGE 140

126 Methodological Issues Several problems have been identified in the studies attempting to show a causal relationship between psychosocial st ress and infertility. One criticism is that various types of fertility disorders are lumped together. Treating all fertility disorders in one study may dilute the m agnitude of the effects of stress because infertility is a complex entit y. Moreover, not all stre ssors that induce a stress response are relevant to reproductive failu re. A second criticism revolves around the inability to separate cause from effe ct. Infertility and emotional distress can be viewed as a circular series of life crises and a variety of emotions ranging from hope to despair. Based on availa ble evidence, the relationship between infertility and emotional distress is best descr ibed as reciprocal: infertility leads to emotional distress; emotional distress, in turn, may make conception less likely which leads to still more stress. Althoug h a positive correlation between infertility and emotional distress has been found one c annot identify whether infertility is the cause of the emotional distress or whether the emotional distress is an etiological factor in infertility. Similar to the criticisms about stress re search in its totality, the term stress in the field of infertility has been a sour ce of much conceptual confusion and has been defined in several different ways. The first common conceptualization of stress in infertility research is as an ev ent defined as a distressing circumstance external to the person. Stress in infe rtility research has also been conceptualized as a response defined as a disturbance of a person’s normal state. Finally, a third psychological conceptualization of stress in infertility research postulates

PAGE 141

127 that stress is determined neither by events nor by response variables. Alternatively, stress is viewed as a co mbination of many factors including the perceived meaning of an event and a selfappraisal of the adequacy of coping resources. Similarly, chronic stress has been defined as a set of related events and conditions that are perceived to threat en important social roles or “domains” that persist over a given length of time. In addition to the lack of a clear, cons istent conceptual framework of the stress process and its relationship to infe rtility, methodological problems in the design and analysis of studies have hindered infertility stress research efforts. One limitation cited is the use of c onvenience samples and/or small sample sizes. Other methodological issues ma y have contributed to conflicting results include homogenous groups of women in terms of ethnicity, age, and socioeconomic status included in the sample s. The limited representativeness of infertility patients seeking medical tr eatment has severely threatened the generalizability of these inve stigations. Because most of the psychologically oriented literature on infertility has focused on couples seeking medical treatment, most of which participate in IVF treatment protocols, the samples included in the studies represent only a se lect group of infertility patients and do not necessarily represent t he treatment population as a whole. Furthermore, it can be argued, that these participants may in fact represent the most physically, psychologically, socially, and financially fi ttest group of infe rtility patients, with perhaps the most stable marriages (Schneider, 2000).

PAGE 142

128 In addition, much of the res earch on this topic is based on nonexperimental and retrospective designs Several statistical models using combinations of biomedical factors in re lation with in vitro fertilization outcomes demonstrate limited external validity because they are based only on stable variables. Additionally, in the area of infertility, even welldesigned, randomized, controlled trials rarely have sufficient st atistical power to demonstrate small but clinically significant differences betw een control and treatment groups (Hughes, 1992; Smeenk et al., 2001; Wilson & Kopitz ke, 2002). Furthermore, empirical evidence collected through a multitude of st udies demonstrates that stress levels are not within the clinical range (B erg & Wilson, 1990; Benazon, 1992; Schneider, 2000). Supporting the cris is model proposed by Menning, psychological distress may be elevated at particular points in time and vary throughout the treatment process. Howeve r, overall distress levels have not been found to be extreme or debilitating. Therefore, psychological strain as opposed to psychiatric morbidity has generally characterized the distress experience by infertility patients (Daniluk, 1988; Seibel & Taymor, 1982; Berg & Wilson, 1990; Schneider, 2000) Based on this accumulated evidence, the use of psychological measures designed to detect psychopathology not specific to infertility may be inappropriate for infertilit y related investigations. The use of inappropriate measurement instruments may reduce t he likelihood of detecting important differences. While attempts have been made to develop instruments measuring stress specific to the infertilit y experience, these measures may also be limited in capturing the multiple dimensio ns of infertility distress. In addition,

PAGE 143

129 traditional psychological measures of depression, anxiety, and marital adjustment appear to be the measures of choi ce by infertility researchers. Meta-Analysis in Research Definition, Development and Use of Me ta-Analysis in Secondary Research Research literature is growing at an exponential rate. As research results continue to accumulate, interpreting re sults and finding the knowledge contained in this flood of information becomes incr easingly difficult. In the traditional method of integrating research studies a reviewer provided a narrative and a chronological discourse on previous findings However, this method is flawed and inexact: This method is unable to deal with a large number of studies on a given topic. Therefore, reviewers focus on a small subset of studies, often without describing how those studies in the subset were selected. The conclusions of previous revi ews conducted are of ten cited without a critical examination of the original reviewers. Reviewers are generally active and prom inent in the fiel d under review. Therefore, evidence contr adictory to their own positions may be ignored, introducing bias as a factor in a reviewer’s conclusions. Addressing these issues, Gene Glass proposed a method to integrate and summarize the findings from a body of rela ted research in 1976. He first used the term “meta-analysis” which he assert s is a philosophy, not a statistical technique. Glass argued that a review of the literature should be as systematic as primary research and should interpret t he results of individual studies in the

PAGE 144

130 overall context of the findings. In a meta -analysis, relevant research studies are collected, coded, and interpreted using stat istical methods similar to those used in primary data analysis. The result is an integrated review of findings that is more objective and exact than traditional re views. According to Hunt (1997), several benefits of conducting a me ta-analysis have been identified: Physicians can now make decisions regarding the use of therapies or diagnostic procedures on the basis of a single article that synthesizes the findings of a multitude of clinical studies. Researchers in every field can ga in a coherent view of the central reality behind the multifar ious and often discrepant findings of research in their field of interest. Meta-analysis of a series of small clinical investigations of a new therapy often provides a finding that physicians can confidently begin using without waiting long years for a large-scale trial to be conducted. Meta-analysis can generally synthes ize differing results. When it cannot, it can often identify the moder ator and mediator variables that account for the differences leading to the identification of precise areas in which future research is needed. In the context of social problems and the development of social policy, meta-analysis offers policymakers easily assimilated syntheses of relevant research that policymaker s generally do not have the time or the training to evaluate independently.

PAGE 145

131 Over the past three decades, me ta-analysis has grown from an unheard preoccupation of a very small group of statisticians working on problems of research integration in education and psychotherapy to a minor academic industry encompassing an assortment of procedures used in a variety of disciplines. Its popularity in the so cial sciences and education pales in comparison to its influence in medical research (Rosenthal, 1978; Hedges, 1985; Hunt, 1997; Hoffert, 1997). Meta-Analysis in Medical Research Evidence-based medicine has been gi ven increasingly more emphasis over recent years. Evidence-based m edicine focuses on the examination of empirical evidence from clinical research for sound medical decision-making. It is especially useful in providing info rmation regarding the st rength and quality of empirical evidence either supporting or refuting a medical practice as well as for developing practice guidelines. In m edical research, meta-analysis uses the accumulated evidence about a treatment or procedure to provide guidance to clinicians and to suggest directions for fu ture study. The use of meta-analysis in the medical literature has generated considerable interest and has proven to be a powerful tool in the field of perinatology. Likewise, obstetrics is one field leading other medical specialties in the attempt to systematically review all randomized trials conducted in its discipline (Hughes 1992; Peipert & Bracken, 1997). There are several advantages to conduc ting a meta-analysis, particularly in medical research. First, the researcher is able to identify gaps, problems and limitations in the primary research base. Since research regarding the

PAGE 146

132 relationship between stress and ART treatme nt outcomes have led to conflicting results, this technique can help researchers to resolve some uncertainty and controversy in this field. Through in creasing sample size, meta-analysis can provide stronger evidence fo r or against a treatment ef fect than can be derived from any of the individual studies because a more precise estimate of the effect size or measure of association is generat ed. This also increases statistical power allowing small but clinically signifi cant differences to be detected. In addition, meta-analysis allows researchers to investigate research questions not posed at the start of an individual tr eatment trial. Finally, another major advantage of this type of research review is the opportunity for others to judge the quality of its conclusions (Hughes 1992; Peipert & Bracken, 1997). Summary Although sometimes seen as a social condition of childlessness, infertility is a significant health problem. The ro le of stress on infertility and treatment outcomes is complex, often leading resear chers to conflicting conclusions. The psychological impacts of infertility have been well documented in the literature. While research findings regarding the prevalence of distress and depressive symptoms in infertile women are inconsist ent, there is evidence to support that at least some women who confront infertilit y are at risk for heightened distress and depressive symptoms. Furt hermore, infertility treatment protocols including ART programs are considered the most invasi ve and expensive treatments available to couples. Because ART treatment progr ams can exact a high toll on couples physically as well as financially and ar e generally the last treatment option

PAGE 147

133 offering hope to couples trying to c onceive a child, these patients may experience the greatest amount of distre ss. Psychoeducati onal interventions may provide an important com ponent to the treatment of in fertility. In addition, these programs may prove to be an effe ctive intervention in preventing the anticipated increase in psychological di stress as the duration of infertility increases. While several theoretical m odels postulate the effects of stress on infertility and infertility treatment outc omes as well as the efficacy of psychoeducational interventions, a synt hesis of the accumulated data incorporating a qualitative assessment of the methodology of reviewed studies as well as a quantitative method of combin ing and analyzing the data examining the effects of stress on ART treatment outco mes is nonexistent. Therefore, this study will investigate through a meta-analytic review of primary research studies conducted to date the impact of stress on the success of ART treatments and to determine whether psychoeducational interv entions mitigate the impact of stress during ART treatments.

PAGE 148

134 CHAPTER THREE: METHOD Problem Statement The psychological impacts of infert ility have been well documented in the literature. While research findings r egarding the prevalence of distress and depressive symptoms in infertile women ar e inconsistent, there is evidence to support that at least some women who confront infertility are at risk for heightened distress and depressive symptom s. Psychoeducational interventions may provide an important component to the treatment of infert ility. In addition, these programs may prove to be an effe ctive intervention in preventing the anticipated increase in psychological di stress as the duration of infertility increases. While several theoretical model s postulate the effects of stress on infertility and infertility treatment outc omes as well as the efficacy of psychoeducational interventions, a synt hesis of the accumulated data incorporating a qualitative assessment of the methodology of reviewed studies as well as a quantitative method of combin ing and analyzing the data examining the effects of stress on ART treatm ent outcomes is nonexistent. Research Purpose, Questions and Hypotheses Although sometimes viewed as a so cial condition of childlessness, infertility is a significant health problem The role of stress on infertility and treatment outcomes is complex, oft en leading researchers to conflicting

PAGE 149

135 conclusions. The purpose of this study wa s to investigate the impact of stress on the success of ART treatments and to determine whether psychoeducational interventions mitigate the impact of stress during ART treatments through a synthesis of the research. According to Cooper (1982), the goal of an integrated review of the research is to summariz e the accumulated knowledge concerning the relations of interest. An additional goal of an integrat ed review is to highlight important issues that research has left unr esolved. The goal of this review was to answer two research questions: 1. What is the relationship betw een stress and Assisted Reproductive Technology (ART) treatment outcomes? 2. Do psychoeducational interventions mitigate the distress experienced by patients participat ing in an Assisted Reproductive Technology (ART) treatment regimen? Therefore, the two hypotheses that were tested in this meta-analysis include: 1. Increased levels of stress will reduce the likelihood of Assisted Reproductive Technology (ART) treatment success, and 2. Psychoeducational interventions provided to patients receiving infertility treatment will mitigate the effects of stress during Assisted Reproductive Technology (ART) treatment.

PAGE 150

136 Method Formulation of the Research Problem The primary goal of research synthesis is the integration of empirical evidence for the purpose of producing gener alizations. There are three purposes inherent in an integrated research review: 1. A critical analysis of the resear ch, paying particular attention to relevant theories, 2. An attempt to resolve c onflicts in the literature, and 3. An attempt to identify current issues to be addressed in future research. There are four primary pr ocesses that transpire in a research synthesis: 1. The formulation of the problem, 2. The collection of dat a and research studies, 3. The evaluation of the data, and 4. The analysis and interpretation of results. Because research synthesis is an int egration of studies, primary research on a topic must exist. The formulation of the problem can arguably be the most important aspect of a res earch synthesis, requiring considerable thought and planning. The considerat ion and decisions made dur ing the formation of a problem will directly influence data collecti on, the evaluation of the data, as well as the presentation of the findings. One key issue to be addressed when developing a problem to research through a meta-analysis is the conceptualization and definition of the problem. This includes considerations of a

PAGE 151

137 number of characteristics i nherent in the problem. Fo r example, is the problem conceptually a broad topic or is it a narro wly defined topic? In addition to the definition of the probl em, the researcher must also specify the universe to which the researcher wishes to generalize. T here are two models of generalizations in which a researcher can approach a meta-analysis. The most common approach in quantitative research synthesis is the fi xed effects (condition al) model. In the fixed effects model, the generalizations and inferences are made only of the studies that have actually been conducted and are observed in the meta-analysis sample. The second model is the random effects (unconditional) model which presumes that the sample of studies included in the meta-analysis represents a sample from a hypothetical collection of studies. Therefore, generalizations and inferences can be applied to other studies or other situations that could have been studied (Hedges, 1994). In this study, both hypotheses were analyzed through a random effects model. The random effects model is conceptua lly justified for this study in that it was expected that the studi es included in the analysis will differ from one another in study characteristics and in effect si ze parameter. In addition, based on prior criticism regarding homogenous groups represented in t he samples of infertility research, a random effects model asserts that the studies to be included in the analysis differ from the possible studies in the universe as a consequence of sampling procedures. The conceptualization of this study implies that the studies included in the analysis are different fr om one another in ways too complex to capture by the inclusion of simple study characteristics. Therefore, inferences

PAGE 152

138 can be generalized to other studies or other situations that could have been studied. It provided this researcher with the ability to provide information regarding the likely relationship betw een stress and ART treatment outcomes. Another fundamental issue in the forma tion of a problem to be addressed with a meta-analysis is the nat ure of the effect size pa rameter to be estimated. Theoretical effect sizes, which remove biases due to artifacts of study design, can systematically influence effect size. Operational effect size parameters represent the true or population rela tionship measured between variables measured in a study without correcting for biases. Therefore, operational effect size parameters can be systematically infl uenced by artificial sources of bias such as restriction of range and measurement error present in a particular study. Although most research syntheses employ operational effect size parameters, some researchers use theoretical effe ct sizes to enhance the comparability and combinability of estimates across studies whose operational effect size would be influenced substantially because of bias es or study design and features. Although some authors have argued that a meta-analyst should not correct for study imperfections or artifacts and esti mate only the operational effect size parameters, Hunter and Schm idt (1994) contend that these artifacts are artificial in nature, stemming from imperfection in research methods. Therefore, they recommend estimating theoretical effect size parameters arguing that correcting these imperfections is e ssential to the development of cumulative knowledge (Hedges, 1994). As noted by Hunter and Sc hmidt (1990), the largest source of the variability across study effect sizes is sampling error. The removal of the

PAGE 153

139 variability across studies that can be a ttributed to sampling error will therefore provide a more accurate estimate of the true variance across study effect sizes. For the purposes of this meta-analysis, co rrection to the statistical artifact of sampling error was made, estimating the theoretical effect sizes. Although it is important to control and estimate bi as in primary research studies, the reduction of bias and simila r errors is of great importance when conducting a meta-analysis over a culminatio n of studies. Although corrections of problems in the design of the prim ary studies cannot be made, methods for measuring errors in study findings due to study imperfection and ultimately correcting these errors in a meta-analys is should be conducted. Errors in primary research studies can be either syst ematic or unsystemat ic. Examples of unsystematic errors include sampling error and bad data such as errors in statistical analysis and improper administra tion of the measurement instrument. An example of a systematic error in cludes attenuation of the population correlation. To address systematic errors in this meta-analysis, primary research studies were weighted by sample si ze and an index of study quality when combining the observed effect sizes across studies as described later in this chapter (Wortman, 1994.) Identification of Studies Once the researcher has formulated and clearly defined the question to be addressed with the meta-analysis, a literature review for t he collection of relevant data and studies is conducted. The purpose of data collection is to locate studies that are representativ e of the intended universe of studi es related to the topic of

PAGE 154

140 interest. Therefore, it is essential that the search crit eria are consistent with the definition of the probl em. If the search criteria used are consistent with the definition, the studies yiel ded are considered to be an exhaustive sample which is a representative sample of studies of t he universe. The point of conducting an exhaustive review of the lit erature is to avoid missing a useful study. One way in which to improve the yield in a research re view is to search multiple bibliographic databases. A second approach to improving t he identification of relevant studies is through citation searches which pr ovide papers comparable in utility and centrality to those produced through the dat abase search. In addition to citation searches, footnote chasing can provide addi tional primary studies of interest. Finally, consultation through informal conv ersations with others in the field can also be helpful in locating usable studies. The ideal literature review will provide the best possible pool of prim ary studies from which to se lect those studies finally included in the analysis (White, 1994). Researchers conducting a literature re view for a meta-analysis must recognize potential threats to the validit y of the studies included in the final analysis. One such threat is publication bias. In many cases, the decision to publish will be influenced by w hether or not the findings yielded statistically significant results, with significant results more likely to be published. Therefore, a preponderance of the publicat ions retrieved through an exhaustive literature review is statistically significant. If the meta-analysis includes only those published studies, there is a risk that it will lead to biased conclusions. Researchers can employ two techniques to avoid the problem of publication bias.

PAGE 155

141 First, this problem can be circumvented by restricting the meta-analysis to include only those studies selected on t he basis of a sampling frame such as through the use of a registry of prospective trials. The use of registered trials requires the inclusion of dat a from the registered st udies that are unpublished and are unaffected by the results of t he study. However, the most common method for avoiding publication bias is by attempting to track down all relevant unpublished studies on the topic. This c an be accomplished through a variety of techniques such as following up on published abstracts and contacting knowledgeable researchers in the field for leads on studies known to have been conducted. This would include s earching for relevant studies among dissertations and master’s theses as we ll as investigating presentations of unpublished studies at meetings and conf erences. Although either of these methods will assist the researcher in av oiding publication bias, even with the greatest of care given to identifying unpublished research, the researcher can never be sure that all or even most of the unpublished work has been located. Analytical tests have also been developed to correct for public ation bias. One such method is the file drawer test which determines how many studies with nonsignificant findings would be necessary to negate the meta-analysis findings. Most recently, more sophist icated methods using a we ighted distribution theory to correct for publication bias have been developed. Although these analytic techniques are available to identify and co rrect publication bi as, the statistical properties of these methods have yet to be subjected to scrut iny. Therefore, these methods are generally not consider ed to be standard methods employed in

PAGE 156

142 a meta-analysis (White, 1994; Rosenthal, 199 4; Dickersin, 1994). In this study, several methods were utilized in an effort to avoid and diminish publication bias. Studies included in this meta-analytic review were first located through an exhaustive comprehensive search of the English languag e literature from January 1985 through December 2003. Elec tronic searches were performed on MEDLINE/PubMed (1985 – 2003), ERIC (1 985 – 2003), Psychinfo (1985-2003), and Dissertation Abstracts Online (1985 – 2003) databases. (A description of each database may be found in Appendix A) Although the search strategy varied depending on the database, search terms for the first hypo thesis included: “infertility” and “psychologic al stress” or “anxiety”; and any terms related to “assisted reproductive technology” such as “In Vitro Fert ilization”, “Gamete Intrafallopian Transfer”, or “Zygote In trafallopian Transfer” and “stress” or “anxiety”. Search terms for the second hypothesis included: “infertility” and any terms related to coping such as “p sychological adaptation” or “group psychotherapy”; and “conception” or any te rms related to “assisted reproductive technology” such as “In Vitro Fertilization” “Gamete Intrafall opian Transfer”, or “Zygote Intrafallopian Transfer” and “s tress management” or “support groups” or any terms related to “coping“, such as “psychological adaptation” or “group psychotherapy”. Following the search of relevant databases, manual scans of reference lists and other publications, as well as branching from primary studies and review articles was conducted to identify any additional studies left uncovered in the original search.

PAGE 157

143 To address potential publication bias, a s earch for registered clinical trials being conducted in the field of reproduc tive endocrinology was conducted. Specifically, a search identifying any clinic al trial involving either the relationship between stress or distress and ART treat ment outcomes or the impact of psychoeducational interventions on stress and distress experienced by patients participating in an ART treat ment program was conducted. The second method addressing publication bias included contacting professionals active in conducting research in this field. Co rrespondence with these professionals was initiated in an effort to identify any ot her unpublished research studies in this field. In addition, efforts to track or iginal study results on published abstracts from professional conferenc es including the American So ciety of Reproductive Medicine annual conference was made. In order to reduce selection bias and increase internal validity, this researcher reviewed each article identifi ed in the literature review. The purpose of this review was to ensure all articles and studies meeting specified criteria for inclusion in the meta-analysis were ident ified. For each hypothesis, each study met all of the following inclusion criteria: 1. For hypothesis one, the study must involve situations where women were participating in an ART treat ment program and the focus of the study was on the relationship between stress and ART treatment outcomes. For hypothesis two, the study must involve women participating in a psychoeducational intervention program and an ART

PAGE 158

144 treatment program. In addition the study must focus on the relationship between the psychoeduc ational intervention and stress. 2. The study must have been cond ucted between January 1985 and December 2003 and be prospective in design. 3. The study must report outcome m easures of stress or anxiety and treatment outcomes. For hypothes is one, treatment outcome is defined as achieving pregnancy or failure to achieve pregnancy. For hypothesis two, treatment outcome is defined as the post-treatment score. Studies with insufficient data fo r effect size calculations will be excluded. Using the above inclusion cr iteria, studies from electronic searches, references from primary studies and review articles were examined to identify potential studies for inclusion. Following this procedure, the collected studies were reviewed by this researcher to identify all studies meeting these criteria for inclusion in the meta-analysis. In attempting to obtain all of the research, published and unpublished, on the topic of interest, it is inevitable that similar reports based on the same research study may be retrieved. For exam ple, after completing research for the completion of a dissertation or thesis, a researcher will often present the results at a conference or publish the research in a professional journal. In the event that reports appeared to be based on t he same research, the author was contacted to verify the uni queness of these reports. If a choice was presented between an unpublished version of the rese arch such as in a dissertation or

PAGE 159

145 thesis versus a published report of the re search, the report c ontaining the most information in terms of reporting outcome s was given priority and was used as the reference of choice in this review. Evaluation of Data Coding the Quality of the Studies Once the studies were identified and adequate information was available, each study was subjected to a structured re view of the quality of the study. This review applied the validity framework dev eloped by Campbell and his associates, providing a matrix of designs and study features or “threats to validity” (Wortman, 1994). The design features of this framework incorporat e four categories. First, internal validity refers to the truthfulne ss regarding statements that can be made about whether there is a causal relations hip from one variable to another in the form in which the variables were manipulat ed or measured. External validity is defined as the “approximate” truthfulness in the generalizations made about the presumed causal relationship across diffe rent persons, settings, and times. Next, statistical conclusion va lidity refers to the trut hfulness of the conclusions drawn about covariation between the identified independent and dependent variables. Finally, construct validity refe rs to the approximat e truthfulness with which a researcher can make generaliz ations regarding the higher-order constructs from the research operations. In this study, cons truct and external validity was used to determine if a particu lar study satisfied the hypothesis with respect to the cause, effect (s), participants, and setting. In this synthesis, the

PAGE 160

146 construct of cause for hypothesis one was identified as stress while the construct of effect included ART treatment outco mes defined as whether or not pregnancy was achieved. For hypothesis two, the c onstruct of cause was identified as the psychoeducational intervention while the construct of effect was defined as the stress experienced. The participants exam ined for both hypotheses in this metaanalysis were infertile women participating in ART treatment progr ams. If a given study met these criteria, then it was consi dered to be relevant for the inclusion in this synthesis. Internal validity was established through identifying only those studies employing prospective designs. Statistical conclusion validity was established by excluding those studies t hat applied inappropriate statistical tests that cannot be corrected by the meta -analyst or inappropriate grouping or comparisons in the analyses. The structur ed review of the studies for this metaanalysis is presented in Appendix B. Each question was scored as 1 for yes responses and 0 for no responses. The investigator then generated a summary score for each group of items and used the average of these summary scores to indicate the quality of each study. Coding the Studies Once all relevant studies have been retrieved, the meta-analysis researcher must then determi ne the characteristics of interest within the studies. Conventions for coding these characteri stics must then be dev eloped. Finally, coding forms capturing the characteristi cs present in each study must then be constructed. The purpose of coding study c haracteristics is to ensure the reliable and orderly extraction of information from each of the studies. This provides

PAGE 161

147 information regarding the quality of each study as well as the adequacy of the information in each study. In addition, the synthesist should seek independent verification and compute reliability estima tes. Key items to include in a coding system are as follows: Report identification such as t he author, country, year, source of publication and coder of the study, Setting describing the general conditi ons of the study such as the scope of sampling, the involvement of special populations, and demographic information, Subjects including specific char acteristics of the sample(s) and subsample(s) participating in the study, Methodology which describes the res earch design, details related to sampling and attrition, and the pr esence or absence of threats to internal validity, Treatment identifiers in cluding the theoretical or ientation motivating a treatment to be investigated, spec ific components of a treatment, the nature of the control groups, the dur ation of the treatment, and the mode of the treatment delivery, and The quantitative information required to estimate an effect size as well as items that describe the nature of the outcome measures (Stock, 1994). Additionally, coding outcomes and st udy features will identify those methodological and substantive characte ristics that may be responsible for

PAGE 162

148 significant variations in the findings. Fo r this study, characteristics reviewed and coded are presented in Appendix C, while the coding sheet appears in Appendix D and the coding manual and refer ence guide appears in Appendix E. A pilot test of the master code s heet was conducted. Two professional evaluators meeting the follo wing qualifications were re cruited to participate: Completed a Doctor of Philo sophy degree in Measurement and Research, and Current professional in the fiel d of research and evaluation. Training and implementation of the codi ng process were as follows: First, the two independent evaluators were furnished with a copy of a coder training manual and reference guide. Each c oder used the manual and reference guide to code a single article independently. Next, the researcher met with each evaluator independently to discuss probl ems encountered in using the guide and the coding sheet, and to make any adjustm ents or changes to the guide, the coding sheet, or both as required. The two evaluators were then assigned to a random sample of all of the articles to code for each hypothesis. For hypothesis one, each evaluator received the same six ar ticles to code. For hypothesis two, each evaluator received the same two articles to code. The code sheets submitted by the two evaluators were t hen compared to the master code sheet developed by the researcher. Items scored differently by either the researcher or by the two independent evaluators were di scussed by the team. Item scores in disagreement were coded by majority decision between the three evaluators.

PAGE 163

149 The percentage of agreement between c odes assigned by a given coder and those established by the master code sheet, was then calculated. Analysis and Interpretation Study coding and data was compile d using Microsoft Excel XP spreadsheet to collect and organize the info rmation. An Excel form was created to prompt for all needed features of each study and automatically save the information in a database from which t he analysis was conducted. SAS 8.2 was used to perform the statistica l tests needed for this study. Effect Size Studies that were selected for in clusion in the meta-analysis were analyzed to generate an effect size, a meas ure of the magnitude of the score change between the pre and post treatment, or conceived versus failure to conceive groups. For hypothesis one of this study, the effect size represents the direction, positive or negative, and magnitu de of the influence of stress on ART treatment outcomes. For hypothesis two of this study, the effect size represents the direction, positive or negative, and magnitude of the influence of psychoeducational interventions on an individual’s level of stress during ART treatment. The computati on of effect sizes for outcomes comparison was based on one of the following: 1. Direct computation, 2. Results of significance te sts reported in the study, and 3. Results from significance levels.

PAGE 164

150 The effect size metric that wa s used for each analysis was Cohen’s d. Cohen’s effect size provides a common scale for comparison of study outcomes reflecting the difference between estimated population means divided by average estimated population standard dev iation. Effect sizes were computed using the most appropriate option listed in Table 2, depending on the data available in each study. Because of the complexity of conceptualization of stress in infertility research, the results of measures of stress may represent several constructs, multiple measures of the same construc t, or both. When the results represent several different constructs, the effect sizes on all constructs and measures were coded. For example, results reported fo r anxiety as a measure of stress may utilize the State-Trait Anxiety Inventory (STAI). The STAI measures two distinct constructs: 1. State Anxiety evaluates how respondent s felt at a particular time in the recent past and how they anticipate t hey will feel either in a specific situation that is likely to be encounter ed in the future or in a variety of hypothetical situations. In addition it assesses the level induced by stressful experimental procedures and by unavoidable real-life stressors such as imminent surgery, dental treatment, job interviews, or important school tests. It is a sensitive indicator of changes in transitory anxiety experienced by c lients and patients in counseling, psychotherapy, and behavior modification programs. 2. Trait Anxiety is typically us ed for screening high school and college students and military recruits for anxie ty problems, and for evaluating

PAGE 165

151 the immediate and long-term outcome of psychotherapy, counseling, behavior modification, and drug-treat ment programs. It has been proven useful for identifying persons with high levels of neurotic anxiety and for selecting subjects for psychological experiments who differ in motivation or drive level. Therefore, effect sizes were computed fo r both state anxiety and trait anxiety. Several different constructs were ex pected to be represented in the primary research studies including: state anx iety, trait anxiety, depression measures, mood assessments, and physiological meas ures of the stress response. When results from multiple measures of stre ss representing the sa me construct were reported for the same sample group in a study, Cohen’s effect sizes were calculated for each measure and then av eraged and weighted to provide one statistic for each study. Finally, many st udies may assess the impact of stress measured at different times throughout the ART treatment program. For example, the STAI may be administer ed at the beginning of the treatment program before any medications have been administered, at the time oocyte retrieval, at the time of embryo transfer and at the time of the pregnancy test. This time-series information provided an in teresting analysis of temporal patterns in ART outcomes. Effect sizes for studies providing measures at multiple points during the treatment program were co mputed separately and compared across studies using univariate procedures.

PAGE 166

152 Table 2 Effects size computation Data Supplied Effect Size Formula Variable 1. Group means, standard deviation 2 1d 1= mean score of group 1 2= mean score of group 2 = pooled standard deviation 2. F statistic, df, N n n df n n F d2 1 2 1 F F statistic n1 = sample size of group one n2= sample size of group two dfdegrees of freedom 3. Chi-square, N 2 24 N d 2chi-square N total number 4. Correlation r r d21 2 r correlation coefficient 5. Significance level df t d 2 dfdegrees of freedom Incomplete Reporting of Results Even the most carefully planned res earch synthesis will encounter problems emerging from studies that do not provide comparable information. The problem of missing data occurs in an integrated review when studies do not report the relevant statis tics on the outcome data or adequate descriptions of the methods needed in order to apply quantitat ive techniques for combining the results across studies. Three kinds of data required for conducting an integrated review may be missing. The first type of missing data includes studies that are

PAGE 167

153 unavailable for use in a synthesis. In an ef fort to avoid this type of missing data, an exhaustive search of the liter ature including publis hed and unpublished reports on the topic of interest as de scribed earlier was conduc ted. In addition, studies that are to be included in a res earch synthesis may also lack relevant information on the variables that are thought to moderate effect size. Finally, studies identified and collected in the literature review may be missing the relevant information necessary for calcul ating a measure of the study’s effect size. This problem arises when studi es report no statistics or an inadequate amount of information about the outcome score s. In this meta-analytic review, studies that reported an insufficient amount of data to compute an effect size, the researcher was contacted in order to obtain all relevant information. In the event that information to compute an effect si ze is unobtainable, if the researcher describes the findings in terms of “non-significant” wit hout reporting the associated statistic, an effect size of 0 was assumed. The use of this convention provided a conservative impact on the quant itative review results. For those studies describing a statistical test as reaching a particular level of significance rather than stating the ex act probability associated with the outcome of the inference test, the probability was assumed to be equal to t he stated value. This convention provided a conservative estima te of the significance level in each case (Piggott, 1994). Combining Estimates of Effect Size Because this study assumes a ra ndom effects model a priori, the population effect size for the i th study ( i) is assumed random with its own

PAGE 168

154 distribution. Therefore, the total variability of an obs erved study effect size estimate (vR i) reflects both conditional variati on of that effect size around each population ( i) and random variation of the individual I around the mean population effect size. Therefore, the overall mean and variance of each effect size were determined, weighted by the sample size and study quality using the following formulas: k i R i i k i i R i iw q T w qT1 1., where T= an unbiased estimate of the population parameter wR i = weight assigned to the i th study computed as vi R iw*1 qi = the i th studys score on the quality index, and Ti = one observed effect size in the i th study with a population effect size of i and variance vi. Using the weights described ab ove, the conditional vari ance of the average effect size Twas computed as follows: w wF i F iq i q iT2 2 1)var(

PAGE 169

155 where vi F iw1 where n n d n n n ni i i i i i i iv2 1 2 2 1 2 12 where ni1 = within-study sample size in group 1 of the i th study, ni 2 = within-study sample size in group 2 of the i th study, and di = estimates the population parameter i. The specific effect size statistic used to estimate T i is the standardized mean differences. The following formula was used to compute the standardized mean differences: s X Xi i i id2 1 where Xi 1 = the mean of group 1 in the i th study, Xi 2 = the mean of group 2 in the i th study, and si = the pooled standard devia tion of the two groups. The following formula was used to compute the variance: vi iv 2*,

PAGE 170

156 where 2 = between-studies variance or the variance component which serves as an estimate of the weight ed sample estimate of the unconditional variance Ti2 and computed as follows: c k Q/ 1ˆ2 where c = k i F i k i F i k i F iw w w1 1 2 1, and vi = within-study variance or the co nditional variance and computed as follows: n n d n n n ni i i i i i i iv2 1 2 2 1 2 12 where ni 1 = within-study sample size in group 1 of the i th study, ni 2 = within-study sample size in group 2 of the i th study, and di = estimates the population parameter The following formula was used to correct for sampling error, providing for an unbiased estimate of the variance: 8 d 2 1 4 3 1) (N N Ne

PAGE 171

157where N= mean sample size and d = estimates the population parameter The unbiased estimate of the variance was then computed as follows: vev)( ivu To determine whether or not the variance component differs significantly from zero, the following homogeneity te st statistic was computed: k i k i F i i k i F i i F iw T w T wQ1 1 1 2 2 Q was rejected if the upper-tail critical value of chi-square at k-1 degrees of freedom (where k = number of groups) is significantly greater than what we would expect by chance if all studies s hared a common population effect size. If the Q test for homogeneity of effect size was rejected, an estimate of the magnitude of the variance component was investigated. The standard error of the estimate of the combined effect size is equal to the square root of v. Multiplying the standard error by an appropriate critical value C (1.96, = 0.05), and adding and subtracting t he resulting product to T provided a 95% confidence interval for as follows: L = T C( )var( T), U = T + C( )var( T) The estimate of the variance component, 2, provides a nonzero estimate of the variance component only if the homogeneity statistic Q is larger than its expected

PAGE 172

158value under the null hypothesis that 2 = 0. In addition, if the Q statistic was rejected, then study effect sizes were disaggregated in to appropriate categories based on characteristics of the studies Graphical displays depicting the frequency distribution of effect sizes as well as the confidence interval are provided for each individual study as well as for combined results. In the event that studies have been disaggregated into categories, the studies were then examined for any moderating e ffects that may be present. Identification of Outliers Following the computation of t he overall mean and variance across studies, the data was analyzed to identify pot ential effect sizes for outliers. The identification of outliers is important in that these dat a could result in a notable increase in the observed variance and a distortion in the mean, altering the conclusions reached in this meta-anal ysis. Although the methodology for detecting outliers in meta-analytic review s is sparse, Huffcutt and Arthur (Arthur et al, 2001) have proposed the Sample -Adjusted Meta-Analytic Deviancy (SAMD) statistic. The SAMD statistic co mpares the value of each study effect size computed without that ef fect size in the analysis and adjusts the difference for the sample size of the study, result ing in one SAMD statisti c for every primary study included in the analysis as well as a distribution of SAMD statistics that approximate a t distribution. Bec ause SAMD values approximate the t distribution, values greater than 3.0 we re considered extreme and identified as

PAGE 173

159potential outliers (Arthur et al., 2001). The equation for computing the variance is as follows: k N N Nout w dv* 3 8 d 1 1 42 study / where d = mean effect size, N = average sample size of the studies, and k = the number of studies used to compute the mean effect size. The SAMD statistic is then comput ed using the following equation: v dd o w i iSAMD v di study / where d = mean effect size and vd vi = standard error. Statistical Analysis A formal analysis was completed for the effect sizes utilizing a mixed effects linear regression model and weig hted least squares estimation. The prediction model is as follows: e u Ti i ip p i i i ...2 2 1 1 0 where 0= the model intercept,

PAGE 174

160 ip i,.....,1 = the coded characteristics of studies hypothesized to predict the study effect size i, p,.....,1 = regression coefficients capturing the association between study characteristics and effect sizes, and e ui i = the random effect of study i (error term). (T he deviation of study i ’s true effect size from the val ue predicted on the basis of the model. Each random effect is assumed independent with a mean of zero and a variance vi 2.) The regression was computed with t he effect estimates as the dependent variable and the predictor variables as independent variables with weights defined by the reciprocal of the sampling variance. Therefor e, a weighted least squares approach was used to determi ne the optimal weights as follows: viwR i 21. In order to estimate 2, and, therefore, the weights, wR i, the method of moments procedure was used. In this proc edure, provisional estimates of ’s was computed in three steps as follows: 1. Computation of estimates ˆ ,...., ˆp o using ordinary least squares regression to yield a by-product of the residual sum of squares: ip p o iTRSS ˆ ˆ ˆ...12 2. Estimate 2 as follows:

PAGE 175

161 2* 2 constant 1 constantRSS Eo This equation is solved for 2 by substituting the observed sum of square residuals, RSS for the expe cted E(RSS), which led to the following estimate: 2 constant 1 constant RSS2. (When a negative number is produced, 2 will be set to zero.) 3. The new estimates of the regres sion coefficients were computed using weighted least squares regr ession with the weights provided as follows: 2 21 vi iw. Two hypothesis tests were conducted. First, the null hypothesis for the regression coefficient q for any q = 0, …., p is as follows: 0 :0 q H In order to test this hypothesis, the rati o of the estimate to its standard error, depicted as follows: ˆ ˆq qSt, where S(ˆq) is the estimated standard error of ˆq which are produced by the weighted least squares regre ssion procedure. The obtained t was then compared with the critical values of t with k – p – 1 degrees of freedom.

PAGE 176

162Finally, the null hypothesis for the random effects variance was tested and is as follows: 0 :2 0 H This hypothesis was tested by computi ng a weighted least squares regression with weights equal to v wiF i1 The weighted residual sum of squares was compared with the critical values of chi-square distribution with k – p – 1 degrees of freedom. Interpreting the Results When interpreting the overall result of the meta-analyses, a confidence interval was constructed to determine whether the average effect size under investigation encompasses zero. In additi on, the use of Cohen ’s guidelines was used to evaluate significance of effect size defined as: 0.2 = small association, 0.5 = medium association, and 0.8 = large association observed. Effect sizes are also reported for each variable. For hypothesis one, effect size comparisons were made with respect to the following variables: time of measure (baseline/pre-treat ment, follicular phase, oo cyte retrieval, embryo transfer, and luteal phase), construc t (acute stress, chronic stress, and depression), duration of infe rtility and country in which the study was conducted. For hypothesis two, effect size compar isons were made for construct measured (acute stress, chronic stress, and depression).

PAGE 177

163Generalizability One strength of this study is that it allows the relationship between stress and achieving pregnancy through ART treat ment protocols to be estimated across populations, at different times of the ART program, and study designs. It was hoped that the results of the first analysis w ould provide adequate information to resolve the controversy regarding the relationship between stress and ART treatment outcomes and the effica cy of psychoeducational interventions for patients participating in ART programs. However, the results of this study are only applicable to women participating in an ART treatm ent program for the purposes of resolving infertility and achieving pregnancy. The extent to which these results apply to other populations has not be determined. T herefore, generalizations about the relationship between stress and fertility among normal women should not be made. Furthermore, the efficacy of psyc hoeducational interventions such as the application of the relaxation response in order to increase the likelihood of conception to women who have not been diagnosed with infertility cannot be determined. Finally, while this study wil l examine the relationship between stress and ART treatment outcomes, it did not ex plore other factors, such as coping skills, on the success of ART treatment.

PAGE 178

164CHAPTER FOUR: RESULTS Problem Statement The role stress plays in infertility and infertility treatment has been a common topic of research in the area of reproductive health and medicine. Historically, researchers have focused on whether stress caused infertility. While empirical research has not provided any evidence of this causal relationship, recent researchers have renewed this focus through the modification of the psychogenic hypothesis. While evidence establishing a causal relationship between stress and infertility is yet to be pr ovided, the literature is replete with evidence documenting the stress caused by infertility and infertility treatment. Because of the salient and individualisti c properties of stress, research has demonstrated that while general cyclical pa tterns of emotional reactions to the stress of infertility are present, every indi vidual responds to the stress of infertility in varying degrees and may progress th rough the various stages of emotion in varying sequences. Most recent research focuses on the affect stress may have on the success of infertility treatment. Ho wever, results of these studies are varied leaving this issue to debate among researchers and physicians. Additionally, evidence regarding psychoeducat ional interventions in mitigating the impact of stress on the success of infertility treatment is in its’ infancy. While mounting evidence suggests that psychoeducational interventions may be an

PAGE 179

165effective method addressing the emoti onal needs of infertility patients, a synthesis of accumulated data examini ng the relationship between stress and infertility treatment outcomes as well as the mitigating properties of psychoeducational interventions on infert ility treatment success was nonexistent until the conduct of this study. Research Purpose and Questions Although not an epidemic, infertilit y is a significant health problem affecting millions of American couples. The role of stress on infertility and infertility treatment outcomes is complex, often leading researchers to conflicting conclusions. The purpose of this study wa s to investigate the impact of stress on the success of Assisted Reproductive Techno logy (ART) treatment outcomes. In addition, this study sought to determine whether psychoeducational interventions mitigate the impact of stress during ART treatments. Theref ore, this study addresses two questions through a synt hesis of the existing research: 1. Do increased levels of stress r educe the likelihood of ART treatment success? 2. Do psychoeducational interventions provided to patients participating in infertility treatment mitigate the effects of stress during ART treatment?

PAGE 180

166Collection and Evaluation of Studies Literature Review After identification of the research questions, studies to be included in the meta-analysis must be identified. Ther efore, an exhaustive search of the literature was conducted in an effort to identify every primary research study conducted. Studies to be included in th is meta-analysis were identified through the following methods: A comprehensive search of the English language literature from January 1985 through December 2003 through electronic database searches including MEDLINE/PubMed, ERIC, Psychinfo, and Dissertation Abstracts Online; Electronic branching from identified research studies located through the database searches; Manual scans of primary re search studies identified; Contacting professionals active in conducting research in this field; A search all clinical trials completed; and A search of published abstracts from professional conferences including the American Societ y of Reproductive Medicine. The results of this comprehensive s earch produced a total of 419 published articles, 1 conference abstract and 3 doc toral dissertations focusing on the relationship between stress and ART treat ment outcomes, hypothesis one. Contact with Alice Domar, Jackie Bo ivin and Pauline Slade identified an

PAGE 181

167additional 8 studies. The abstracts of each study and article were then read to identify those studies t hat examined the relationship between stress and ART treatment outcomes on at least one outcome measure. Of the initial pool of studies and articles located, 32 public ations, one conference abstract and two dissertations were empirical investigat ions reporting research study results. The search method for hypothesis two used the same searching methods described above, locating a total of 62 published studies and one dissertation focusing on the efficacy of psychoeducatio nal interventions in mitigating the infertility stress experienced. Among these studies initially located, 21 studies and one dissertation reported the results of primary, empirica l investigations. Identification of Studies for Inclusion in the Meta-Analysis In order to reduce selection bias and increase internal validity, this researcher reviewed each article ident ified presenting empirical data in the literature review. The purpose of this review was to ensure all articles and studies meeting specified criteria fo r inclusion in the meta-analysis were identified. For each hypot hesis, each study met all of the following inclusion criteria: 1. For hypothesis one, the studies involved women participating in an ART treatment program. The focu s of these studies was on the relationship between stress and AR T treatment outcomes. For hypothesis two, the study invo lved women participating in a psychoeducational intervention program and an ART treatment

PAGE 182

168program. In addition, the study fo cused on the relationship between the psychoeducational intervention and stress. 2. The studies were conducted between January 1985 and December 2003. 3. The design of the study was prospective. 4. The studies reported outcome m easures of stress or anxiety and treatment outcomes. For hypothes is one, treatment outcome was defined as achieving pregnancy or failure to achieve pregnancy. For hypothesis two, treatment outcome was defined as the post-treatment score. Studies with insufficient data fo r effect size calculations were excluded. Among the 35 empirical studies locat ed examining the relationship between stress and ART treatment outcomes, a tota l of 13 (37%) met the criteria for inclusion while the remaining 22 studies (Appendix F) were excluded from the meta-analysis for the following reasons: 1. Three (14%) included women in the sample that were participating in other types of infertility treatment such as Artifi cial Insemination and Intra-Uterine Insemination; 2. Eleven (50%) did not focus on the relationship between stress and ART treatment outcomes; 3. One (5%) was retr ospective in design; 4. Six (27%) reported insufficient data or statistical analysis; and

PAGE 183

1695. One (5%) study was a copywrit er-protected dissertation and was unobtainable. Of these 13 studies included in the me ta-analysis for hypothesis one, two (15.38%) were conducted in Australia, th ree (23.08%) in Asia and the Middle East (Southeast Asia), seven (53.85%) in Europe, and one (7.69%) in North America. Data regarding t he characteristics of the sa mple included in each study were collected. Table 4 below depict s the number and percentage of studies reporting sufficient information for descr ibing the participants included in the meta-analysis and for coding study characte ristics as described in Appendix E. Table 4 Studies reporting characteristics of sample for hypothesis one Characteristic N % Age 11 84.46 Education 3 23.08 Employment 3 23.08 Economic Status 1 7.69 Classification 3 23.08 Etiology 11 84.46 Duration of Infertility 8 61.54 Type of ART 13 100.00 Number of Previous ART Attempts 4 30.77 Descriptive information for the female participants incl uded in the meta-analysis is presented in Table 5.

PAGE 184

170Table 5 Demographic information for studies reporting the effects of stress on ART treatment outcomes Characteristic N X ( SD) % Overall Sample Size 1348 103.69 ( 80.99) Overall Age 1195 32.99 ( 1.21) Employment Status Unknown 723 53.64 Not Employed 163 12.09 Employed 462 34.27 Classification of Infertility Unknown 1001 74.26 Primary 304 22.55 Secondary 43 3.19 Etiology Unknown 178 13.20 Female Factor Only 22 1.63 Male Factor Only 0 0 Combination 1148 85.16 Duration of Infertility 1117 5.39 ( 1.17) Type of ART IVF 1165 86.42 IVF or ICSI 127 9.42 IVF of GIFT 56 4.15 Previous ART Attempts Unknown 935 69.36 0 268 19.88 1 or 2 138 10.24 3 or More 41 3.04

PAGE 185

171Among the 22 empirical studies loca ted examining the effects of psychoeducational interventions on the stress experienced by women participating in an ART program, a total of 4 (18%) studies met the inclusion criteria for hypothesis two. Of the 22 studies reporting empirical data for hypothesis two, a total of 18 were ex cluded from the meta-analysis for the following reasons: 1. Twelve (55%) focused on women diagnosed with infertility, but not participating in an ART treatment pr ogram or participating in other types of treatment programs such as Intra-Uterine Insemination; 2. One (5%) focused on men only; 3. Four (18%) focused on outcome measures other than the impact on stress; and 4. One (5%) was retr ospective in design. Of these four studies included in t he meta-analysis for hypothesis two, one (25%) was conducted in Asia, two (50% ) in Europe, and one (25%) in North America. Data regarding t he characteristics of the sa mple included in each study were collected. Table 6 below depict s the number and percentage of studies reporting sufficient information for descr ibing the participants included in the meta-analysis and for coding study characte ristics as described in Appendix E.

PAGE 186

172Table 6 Studies reporting characteristics of sample for hypothesis two Characteristic N % Age 3 75.0 Education 1 25.0 Employment 1 25.0 Economic Status 2 50.0 Classification 1 25.0 Etiology 1 25.0 Duration of Infertility 4 100.0 Type of ART 3 75.0 Number of Previous ART Attempts 2 50.0 Descriptive information for the female participants incl uded in part two of the meta-analysis is presented in Table 7. Table 7 Demographic information for st udies reporting the effects of psychoeducational interventions on stress experienced by women participating in an AR T treatment program Characteristic N X ( SD) % Overall Sample Size 498 124 ( 90.94) Treatment Group Sample Size 203 50.75 ( 39.76) Control Group Sample Size 225 56.25 ( 50.96) Overall Age 369 33.47 ( 1.29) Employment Status Unknown 438 88.0 Not Employed 25 5.0 Employed 35 7.0 Classification of Infertility Unknown 346 69.5 Primary 89 17.9 Secondary 63 12.7 Continued on the next page

PAGE 187

173Table 7 (Continued) Etiology Unknown 397 79.7 Female Factor Only 57 11.5 Male Factor Only 26 5.2 Combination 18 3.6 Duration of Infertility 369 4.98 ( 1.04) Type of ART IVF 498 100.0 IVF or ICSI IVF of GIFT Previous ART Attempts Unknown 257 51.6 0 241 48.4 1 or 2 3 or More Psychoeducational Intervention Counseling 177 87.2 Support Group Cognitive Behavioral Format 26 12.8 Other Frequency of Psychoeducational Intervention 1 – 3 sessions 147 72.4 4 – 5 sessions 30 14.8 Other 26 12.8 Length of Psychoeducational Intervention Unknown 110 54.2 1 hour 37 18.2 1.5 hours 26 12.8 2 hours Other 30 14.8 The final sample to be included in the me ta-analysis consisted of a total of 13 studies meeting the inclusio n criteria for hypothesis one and four studies meeting the inclusion criteria for hypothesis two (Appendix F).

PAGE 188

174Coding the Characteristics of Included Studies Once all relevant studies were retr ieved, the researcher determined the characteristics of interest within the studies. Conventions for coding these characteristics were developed. Finally, coding forms capturing the characteristics present in each study we re then constructed. The purpose of coding study characteristics was to ensur e the reliable and order ly extraction of information from each of the studies. For this study, characteristics reviewed and coded are presented in Appendix C, while the coding sheet appears in Appendix D and the coding manual and refer ence guide appears in Appendix E. In addition to coding the characteristi cs of interest, the quality of each study was reviewed. This review appli ed the validity framework developed by Campbell and his associates (Wortman, 1994), providing a matrix of design and study features. This structured review of each study’s quality for this metaanalysis appears in Appendix B and was included in the coding process for each study. Based on the responses, a score of each study’s quality index was generated. Each question was scored as 1 for yes responses and 0 for no responses. The quality index score was determined as follows: 1. Each category was summed and averaged and 2. The average scores from each category were averaged. A pilot test of the master coding sheet, developed by this researcher, capturing the relevant study characterist ics as well as the quality characteristics was conducted. Two professi onal evaluators meeting t he following qualifications were recruited to participate:

PAGE 189

175 Completed a Doctor of Philo sophy degree in Measurement and Research and Current professional in the fiel d of research and evaluation. Training and implementation of the codi ng process were as follows: First, the two independent evaluators were furnished with a copy of a coder training manual and reference guide. Each c oder used the manual and reference guide to code a single article independently. Next, the researcher met with each evaluator independently to discuss probl ems encountered in using the guide and the coding sheet, and to make any adjustm ents or changes to the guide, the coding sheet, or both as required. The two evaluators were then assigned to a random sample of all of the articles to code for each hypothesis. For hypothesis one, each evaluator received the same six ar ticles to code. For hypothesis two, each evaluator received the same two articles to code. The code sheets submitted by the two evaluators were t hen compared to the master code sheet developed by the researcher. Items scored differently by either the researcher or by the two independent evaluators were di scussed by the team. Item scores in disagreement were coded by majority decision between the three evaluators. Interrater agreement was assessed by co mparing the values recorded by each coder for each of the variables of interes t. Raters were in agreement if all coders recorded identical values. The level of agreement obtained for the variables is presented in Table 8. As these result s indicate, the level of agreement was reasonably high with a mean over all agreement of 94.09%.

PAGE 190

176Table 8 Study coding agreement Coding Sections % Agreement Part I – Study Characteristics 98.00 Publication Characteristics 100.00 Ecological Characteristics 97.92 Methodological Characteristics 96.67 Results 100.00 Part II – Quality 92.65 Design 95.00 Participants 93.94 Controls/Implementation 91.67 Protocol 93.33 Outcomes 91.67 Statistics 83.33 Total Instrument 94.09 Findings Hypothesis One The effect size metric com puted for each study was Cohen’s d s X Xi i i id2 1 where Xi 1 = the mean of group 1 in the i th study, Xi 2 = the mean of group 2 in the i th study, and si = the pooled standard devia tion of the two groups,

PAGE 191

177providing a common scale for comparis on of study outcomes. When a study reported multiple measures of stress that represented the same construct, for the same sample group in a study, Cohen’s e ffect sizes were calculated for each measure and then averaged and weighted to pr ovide one statistic for each study. When a study reported multiple measur es of stress representing different constructs as defined in Appendix G, an effect size estimate was computed for each construct. Finally, for studies prov iding outcome measur es at multiple points during the treatment program, an effect size estimate was computed for each point measured. Table 9 lists the information extr acted from each study as well as the index of effect size and 95% Confidence Interval while Appendix H provides the SAS code. Table 9 Studies reporting the effects of stress as it relates to ART treatment outcomes 95% CI Study N Country X Age Etiologya X Duration Timeb Constru ctc d Lower Upper Gallinelli et al. (2001) 40 Italy 1 3 1 0.459 -0.2423 1.1603 Gallinelli et al. (2001) 40 Italy 1 3 2 0.080 -0.6143 0.7743 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 1 1 -0.071 -0.6981 0.5561 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 2 1 0.573 -0.0664 1.2123 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 3 1 0.806 0.1547 1.4573 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 4 1 0.643 0.0005 1.2855 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 5 1 0.866 0.2110 1.5210 Biovin & Takefman (1995) 40 Canada 33.3 0 4.4 years 1 2 0.190 -0.4383 0.8183 Kee et al. (2000) 138 Korea 32. 8 0 5.5 years 1 1 1.412 0.0395 1.0225 Kee et al. (2000) 138 Korea 32. 8 0 5.5 years 1 2 1.787 1.3766 2.1974 Kee et al. (2000) 138 Korea 32. 8 0 5.5 years 1 3 0.949 0.5796 1.3184 Thiering et al. (1992) 312 Australia 33.7 1 1 1 -0.630 -0.8869 -0.3731 Thiering et al. (1992) 312 Australia 33.7 1 1 2 -0.061 -0.3131 0.1911 Thiering et al. (1992) 312 Australia 33.7 1 1 3 -0.117 -0.3693 0.1353 Continued on the next page

PAGE 192

178Table 9 (Continued) Verhaak et al. (2001) 127 Netherlands 33.4 1 3.7 years 1 1 0.199 -0.1788 0.5768 Verhaak et al. (2001) 127 Netherlands 33.4 1 3.7 years 1 3 0.326 -0.0532 0.7052 Ardenti et al. (1999) 200 Italy 33.8 1 6.1 years 3 1 -0.616 -0.9744 -0.2576 Ardenti et al. (1999) 200 Italy 33.8 1 6.1 years 3 2 0.034 -0.3193 0.3873 Sanders & Bruce (1999) 56 Australia 32.6 1 1 1 0.498 -0.0392 1.0352 Sanders & Bruce (1999) 56 Australia 32.6 1 1 2 0.579 0.0390 1.1190 Csemiczky et al. (2000) 22 Sweden 33.4 2 1 1 0.856 -0.0225 1.7525 Csemiczky et al. (2000) 22 Sweden 33.4 2 2 1 0.598 -0.2701 1.4661 Csemiczky et al. (2000) 22 Sweden 33.4 2 5 1 1.528 0.5656 2.4904 Facchinetti et al. (1997) 49 Italy 33.9 1 6.3 years 1 1 0.241 -0.3580 0.8400 Facchinetti et al. (1997) 49 Italy 33.9 1 6.3 years 3 1 0.437 -0.1663 1.0403 Facchinetti et al. (1997) 49 Italy 33.9 1 6.3 years 1 2 0.629 0.0191 1.2389 Demyttenaere et al. (1993) 40 Belgium 32.4 1 6.1 years 2 1 0.473 -0.2502 1.1962 Demyttenaere et al. (1993) 40 Belgium 32.4 1 6.1 years 3 1 0.190 -0.5269 0.9069 Demyttenaere et al. (1993) 40 Belgium 32.4 1 6.1 years 4 1 0.000 -0.7157 0.7157 Demyttenaere et al. (1993) 40 Belgium 32.4 1 6.1 years 2 3 2.723 1.7912 3.6548 Merari et al. (1992) 113 Isr ael 2 1 -0.212 -0.6708 0.2468 Merari et al. (1992) 113 Isr ael 3 1 -0.177 -0.6355 0.2815 Merari et al. (1992) 113 Is rael 4 1 0.065 -0.3930 0.5230 Merari et al. (1992) 113 Is rael 5 1 0.107 -0.3512 0.5652 Merari et al. (1992) 113 Is rael 1 3 0.048 -0.4100 0.5060 Merari et al. (1992) 113 Isr ael 2 3 -0.328 -0.7879 0.1319 Merari et al. (1992) 113 Isr ael 3 3 -0.306 -0.7657 0.1537 Merari et al. (1992) 113 Is rael 4 3 0.067 -0.3910 0.5250 Merari et al. (1992) 113 Isr ael 5 3 -0.007 -0.4649 0.4509 Demyttenaere et al. (1998) 98 Belgium 29.7 1 4.1 years 1 3 0.041 -0.4262 0.5082 Merari et al. (1996) 113 Israel 33.9 1 6.9 years 1 1 0.024 -0.4340 0.4820 Merari et al. (1996) 113 Israel 33.9 1 6.9 years 1 2 0.050 -0.4080 0.5080 Merari et al. (1996) 113 Israel 33.9 1 6.9 years 1 3 -0.332 -0.7920 0.1280 a Etiology is coded as follows: 0 = Unknown, 1 = Mixed (combination of female factor and male factor), 2 = Female factor only b Time is coded as follows to reflect the time the measure was adm inistered: 1 = Baseline/Pre-Treatm ent, 2 = Follicular Phase ( day 3 – day 14), 3 = Oocyte Retrieval, 4 = Embryo Transfe r, 5 = Luteal Phase (approximately days 21 – 28) c Construct is coded to reflect the construct of the measure as defined in Appendix G: 1 = Acute Stress, 2 = Chronic Stress, and 3 = Depression. Note. Effect sizes are scored so that positive numbers reflect great er amounts of stress in the group who failed to become pregnant a nd negative numbers reflect greater amounts of stress in the group who achieved pregnancy.

PAGE 193

179Figure 1 displays the 13 studies included in the meta-analysis for hypothesis one listed in Table 9, rank order ed by the magnitude of the effe ct size by the time at which the measure was administered and the construct for which the measure reports. As illustrated in Figure 1, resu lts across primary studies vary greatly. Among studies reporting acute stress meas ures at the time of baseline/pretreatment, effect size estimates r anged from –0.630 to 1. 412 with a mean of 0.290. Likewise, similar findings are s hown in primary studi es reporting acute stress measures during the follicul ar phase (range= -0.212 to 0.598, x= 0.2579) and at oocyte retrieval (range= -0.616 to 0.806, x=0.1052) while measures of acute stress at the time of em bryo transfer (range= 0 to 0.643, x= 0.1977) and during the luteal phase (range= 0.107 to 1.528, x= 0.7031) were all positive. Among primary studies reporting chronic stress measures, results varied. Baseline/pre-treatment chronic stress measures ranged from –0.061 to 1.787 with a mean of 0.4491 while ch ronic stress measures taken at the time of oocyte retrieval ranged from 0.034 to 0.08 with a mean of –0.1874. Finally, measures of depression demonstrated a lar ge amount of variability ac ross studies as well. Measures of depression reported for base line/pre-treatment and follicular phase end-points ranged from negative to positive effect size estimates (range= 0.332 to 0.949, x= 0.1702; range= -0.328 to 2.723, x= 1.0629, respectively).

PAGE 194

180Figure 1. Dot plot for studies reporting the effect s of stress on ART treatment outcomes -2.5-2-1.5-1-0.500.511.522.533.544.5 Continued on the next page Baseline / Pre-Treatment Acute Stress Thiering et al. (1992) Boivin & Takefman (1995) Merari et al. (1996) Verhaak et al. (2001) Facchinetti et al. (1997) Sanders & Bruce (1999) Csemiczky et al. (2000) Kee et al. (2000) Subgroup Mean Baseline / Pre-Treatment – Chronic Stress Thiering et al. (1992) Merari et al. (1992) Merari et al. (1996) Boivin & Takefman (1995) Sanders & Bruce (1999) Facchinetti et al. (1997) Kee et al. (2000) Subgroup Mean Baseline / Pre-Treatment – Depression Merari et al. (1996) Thiering et al. (1992) Demyttanaere et al. (1998) Verhaak et al. (2001) Kee et al. (2000) Subgroup Mean

PAGE 195

181Figure 1 (Continued) -2.5-2-1.5-1-0.500.511.522.533.544.5 Continued on the next page Follicular Phase – Acute Stress Merari et al. (1992) Demyttanaere et al. (1993) Boivin & Takefman (1995) Csemiczky et al. (2000) Subgroup Mean Follicular Phase – Depression Merari et al. (1992) Demyttanaere et al. (1993) Subgroup Mean Oocyte Retrieval – Acute Stress Ardenti et al. (1999) Merari et al. (1992) Demyttanaere (1993) Facchinetti et al. (1997) Gallinelli et al. (2001) Boivin & Takefman (1995) Subgroup Mean Oocyte Retrieval – Chronic Stress Ardenti et al. (1999) Gallinelli et al. (2001) Subgroup Mean Oocyte Retrieval – Depression Merari et al. (1992) Embryo Transfer– Acute Stress Demyttanaere et al. (1993) Merari et al. (1992) Boivin & Takefman (1995) Subgroup Mean Embryo Transfer– Depression Merari et al. (1992)

PAGE 196

182Figure 1 continued -2.5-1.5-0.50.51.52.53.54.5To test data for outliers, the Samp le-Adjusted Meta-Analytic Deviancy (SAMD) statistic was computed using SAS 8.2. Programming code for this analysis appears in Appendix I while the results are displayed in Table 10. In addition, scree plots were constructed to visually identify outliers and are presented in Appendix J. Effect size s with a SAMD statisti c greater than 3.0 were considered extreme observations. Table 10 Test for outliers among effect sizes fo r studies reporting the effects of stress on ART treatment Time Construct Study d SAMD Baseline/Pre-treatment Acute Stress Kee et al. (2000) 1.412 7.77 Thiering et al. (1992) -0.630 5.44 Csemiczky et al. (2000) 0.865 1.88 Sanders & Bruce (1999) 0.498 1.73 Verhaak et al. (2001) 0.199 0.97 Facchinetti et al. (1997) 0.241 0.75 Boivin & Takefman (1995) -0.071 0.27 Merari et al. (1996) 0.024 0.02 Continued on the next page Luteal Phase– Acute Stress Merari et al. (1992) Boivin & Takefman (1995) Csemiczky et al. (2000) Subgroup Mean Luteal Phase– Depression Merari et al. (1992) Grand Mean

PAGE 197

183Table 10 (Continued) Chronic Stress Kee et al. (2000) 1.787 9.53 Facchinetti et al. (1997) 0.629 1.91 Sanders & Bruce (1999) 0.579 1.86 Thiering et al. (1992) -0.061 1.24 Boivin & Takefman (1995) 0.190 0.39 Merari et al. (1992) 0.048 0.11 Merari et al. (1996) 0.050 0.10 Depression Kee et al. (2000) 0.949 5.14 Merari et al. (1996) -0.332 1.84 Verhaak et al. (2001) 0.326 1.59 Thiering et al. (1992) -0.117 1.32 Demyttenaere et al. (1998) 0.041 0.05 Follicular Phase Acute Stress Boivin & Takefman (1995) 0.573 1.58 Merari et al. (1992) -0.212 1.39 Demyttenaere et al. (1993) 0.473 1.27 Csemiczky et al. (2000) 0.598 1.22 Chronic Stress Demyttenaere et al. (1993) 2.723 7.63 Merari et al. (1992) -0.328 2.26 Oocyte Retrieval Acute Stress Ardenti et al. (1999) -0.616 3.91 Boivin & Takefman (1995) 0.806 2.47 Facchinetti et al. (1997) 0.437 1.50 Gallinelli et al. (2001) 0.459 1.42 Merari et al. (1992) -0.177 0.82 Demyttenaere et al. (1993) 0.190 0.60 Chronic Stress Gallinelli et al. (2001) 0.080 0.21 Ardenti et al. (1999) 0.034 0.09 Embryo Transfer Acute Stress Boivin & Takefman (1995) 0.643 1.60 Merari et al. (1992) 0.065 0.66 Demyttenaere et al. (1993) 0.000 0.31 Luteal Phase Csemiczky et al. (2000) 1.528 3.09 Boivin & Takefman (1995) 0.866 2.25 Merari et al. (1992) 0.107 0.44 The following five studies were identified as outliers: 1. Kee et al. (2000) 2. Thiering et al. (1992)

PAGE 198

1843. Demyttenaere et al. (1993) 4. Ardenti et al. (1999) 5. Csemiczky et al. (2000). These studies were reviewed to dete rmine if there were any identifiable characteristics in the populations or conditions that may account for the differences observed. While there were no clear or obvious differences in the sample population or conditions in the study reported by Th iering et al. (1992), three of the studies did have characte ristics unique to their populations or conditions. The study conducted by Kee et al. (2000) was the only study completed in Asia, specifically South Korea. More notably, in the introduction of the study, it was stated In the past, traditionally if Korean women were infertile, they were regarded as having one of the “seven largest sins.” Thus, they were stressed from the mistreatment at t he hands of their own family members. This historical cultural or religious di fference may have contributed to the extreme effect sizes observed. In the study conducted by Ardent i et al. (1999) the focus was specifically on stress measures tak en at the time of oocyte retrieval and embryo transfer. However, unlike any of the other studies included in the metaanalysis, it was reported that the women were hospitalized during these stages of the treatment. In addition, the mean duration of knowled ge of their infertility for this sample was reportedly 6.1 years, r anging from 1 to 22 years. While this mean duration is not the highest report ed average, the range reported was the largest. Similarly, the study conducted by Demyttenaere et al. (1993) reported a

PAGE 199

185mean duration of infertility of 6.1 years ( 3.1). Finally, in the study conducted by Csemiczky et al. (2000), unlike the other st udies included in the meta-analysis, it was noted that the participants in the st udy had been on a waiting list to receive treatment for an average of 4.3 years. Next, a homogeneity test was conduct ed to determine whether the sample effect sizes for the 28 data points in the meta-analysis were homogenous. SAS 8.2 was used to perform the homogeneity test (see Appendix I). As shown in Table 11, the Q -statistic for measures of depression at bas eline/pre-treatment, acute stress at oocyte retrieval, and ac ute stress at the luteal phase were significant at the .05 level, indicating t hat effect sizes are not homogeneous. Table 11 Test of homogeneity of effect size s including outliers for hypothesis one Source df Q -Stat p Baseline –Acute Stress 7 80.4367 >0.0001 Baseline – Chronic Stress 6 62.3665 >0.0001 Baseline Depression 4 27.5261 0.00002 Follicular Phase – Acute Stress 3 5.7607 0.12385 Follicular Phase – Depression 1 33.1169 >0.0001 Oocyte Retrieval – Acute Stress 5 21.4644 0.00066 Oocyte Retrieval – Chronic Stress 1 0.0134 0.90786 Embryo Transfer – Acute Stress 2 2.4592 0.29241 Luteal Phase – Acute Stress 2 8.4055 0.01495 A test of homogeneity was also conduct ed excluding the outliers identified with the SAMD statistic. Excluding the extreme observations, the Q -statistic was not significant at the .05 level for any of the measures as shown in Table 12 suggesting that the data are homogenous.

PAGE 200

186Table 12 Test of homogeneity of effect size s excluding outliers for hypothesis one Source df Q -Stat p Baseline –Acute Stress 5 4.6051 0.46594 Baseline – Chronic Stress 5 7.6442 0.17696 Baseline Depression 3 5.5882 0.13346 Follicular Phase – Acute Stress 3 5.7607 0.12385 Oocyte Retrieval – Acute Stress 4 4.6051 0.14317 Oocyte Retrieval – Chronic Stress 1 0.0134 0.90786 Embryo Transfer – Acute Stress 2 2.4592 0.29241 Luteal Phase – Acute Stress 1 3.4638 0.06273 A funnel plot investigati ng the properties of the effe ct sizes for the studies included in the meta-analysis was also c onstructed. Figure 11 displays these results. The shape of the funnel plot indi cates the presence of publication bias. In other words, it appears that studies with smaller sample sizes reporting a negative effect, indicating that participants experiencing greater levels of stress were more likely to become pregnant following ART treatment, were not published.

PAGE 201

187Figure 11. Funnel plot for studies reporting the effects of stress on ART treatment outcomes 0 50 100 150 200 250 300 350 -3-2-101234 Effect Size EstimateTotal Sample Size Finally, a regression analysis was c onducted to statistically analyze the effect-size estimates. Two hypothesis tests were conducted. First, the null hypothesis for the regression coefficient q for any q = 0, …., p is as follows: 0 :0 qH. The regression coefficients represent the variables time (baseline/pre-treatment, follicular phase, oocyte retrieval, and embryo transfer) and construct (acute stress and chronic stress). In this model, predictors are coded with depression measures taken during the luteal phase as reference categories. Results including outliers and excluding outliers are presented in Table 13. Although the results including study outliers at each time measured demonstrate a small to moderate negative relationship between stress and ART treatment outcome

PAGE 202

188(ranging from –0.1832 at the ti me of embryo transfer to –0.5037 at the time of baseline/pre-treatment), none of these coefficients are statistically significant. Similar results are found for the constr uct acute stress (-0.0839) demonstrating a small negative effect with ART treatment outcome while chronic stress (0.2682) demonstrates a small positive relationship with ART treatment outcomes. Again, neither acute stress nor chronic stress wa s statistically significant. When study outliers are excluded, the intercept demons trates a small relationship with ART treatment outcomes and is still statistically si gnificant. In addition, the effect size estimates for time (baselin e/pre-treatment, fo llicular phase, oocyte retrieval and embryo transfer) show a small to m oderate negative effect on ART treatment outcome, but are not statistically signi ficant. Furthermore, acute stress and chronic stress demonstrated a small to moderate positive effect on ART treatment outcome. However, neither of these coefficients was statistically significant. Table 13 Hierarchical random effects analysis for hypothesis one Including Outliers Excluding Outliers 95% CI 95% CI Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.7040 0.0280 0.09014 1.31790 0.36360 0.0438 0.01205 0.71510 Baseline / Pre-treatment -0.5037 0.0757 -1.06360 0.05625 -0.30100 0.0689 -0.62790 0.02592 Follicular Phase -0.2081 0.4431 -0.75910 0.34280 -0.25450 0.1111 -0.57360 0.06462 Oocyte Retrieval -0.4211 0.1193 -0.95900 0.11690 -0.26550 0.0868 -0.57350 0.04248 Embryo Transfer -0.1832 0.5125 -0.75210 0.38560 -0.08916 0.5697 -0.41260 0.23430 Acute Stress -0.0839 0.5174 -0.34740 0.17960 0.13190 0.1402 -0.04770 0.31160 Chronic Stress 0.2682 0.1514 -0.04441 0.58070 0.14070 0.1431 -0.05238 0.33390 Note. A total of 13 studies representing 43 effect sizes we re included in the analysis including outliers. A total of 12 studies representing 36 effect sizes we re included in the analysis excluding outliers.

PAGE 203

189The null hypothesis for the random effects variance: 0 :2 0 H. was tested. This hypothesis was tested by computing a weighted least squares regression with weights equal to vi iw1 The weighted residual sum of squares was compared with t he critical values of ch i-square distribution with k – p – 1 degrees of freedom. As shown in Ta ble 14, both the intercept and residual variance estimates are significant at the .05 level for results including outliers while only the residual variance estimate was significant for results excluding outliers. Table 14 Heirarchical random effects variance analysis for hypothesis one Including Outliers Excluding Outliers 95% CIa 95% CIa Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.23640 0.0200 0.10970 0.82910 0.05229 0.0507 0.02089 0.28810 Residual 1.91040 0.0002 1.18080 3.60900 0.57280 0.0012 0.32940 1.23570 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. The results provided by the significance test are considered to be reliable for interpretation. Note. A total of 13 studies representing 43 effect sizes we re included in the analysis including outliers. A total of 12 studies representing 36 effect sizes we re included in the analysis excluding outliers. SAS code for computing the coefficients and conducting the analysis is listed in Appendix K. Since the variables time (baseline/ pre-treatment, follicular phase, oocyte retrieval, and embryo transfer) and cons truct (acute stress and chronic stress) did not significantly cont ribute to the model, a null model including outliers and

PAGE 204

190excluding outliers was constructed. SAS code is presented in Appendix L. Results of this analysis found that the in tercept was not statistically significant when study outliers were included. Howe ver, when outliers were removed from the analysis, results indica te that stress has a sma ll effect on ART treatment outcome with an estimated mean effect size of 0.2012 and is statistically significant (see Table 15). Table 15 Hierarchical random effects null model analysis for hypothesis one Including Outliers Excluding Outliers 95% CI 95% CI Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.30470 0.0541 -0.00635 0.61570 0.20120 0.0252 0.03007 0.37240 Note. A total of 13 studies representing 43 effect sizes we re included in the analysis including outliers. A total of 12 studies representing 36 effect sizes we re included in the analysis excluding outliers. In addition, the analysis of the variance es timates including outliers as well as excluding outliers reveal statistically sign ificant results as outlined in Table 16. Table 16 Hierarchical random effects null model variance analysis for hypothesis one Including Outliers Excluding Outliers 95% CIa 95% CIa Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.20610 0.02000 0.09558 0.72310 0.05271 0.0403 0.02202 0.25080 Residual 2.17100 <0.0001 1.40180 3.80800 0.62790 0.0002 0.38560 1.20000 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. The results provided by the significance test are considered to be reliable for interpretation. Note. A total of 13 studies representing 43 effect sizes we re included in the analysis including outliers. A total of 12 studies representing 36 effect sizes we re included in the analysis excluding outliers.

PAGE 205

191 Finally, a moderator analysis was conduc ted to determine if either the duration of infertility for the study participants (Appendi x M) or the country in which the study was conducted (Appendix N) contribute to the model. The analysis of the coefficient estimates ar e presented in Table 17 while the random effects variance analysis is presented in Ta ble 18. None of the estimates for the variable duration of infertility were statis tically significant regardless of whether study outliers were included or excluded in the analysis. When examining the variable country in which the study was conducted, when study outliers are included, the coefficient estimate for studies conducted South East Asia was statistically significant at the .05 level. However, when outliers are excluded from the analysis, results dem onstrate that a small pos itive relationship between stress and ART treatment outcome is signific ant at the .05 level while none of the coefficients by country were statistically significant. Table 17 Hierarchical random effects analysis of moderators for hypothesis one Including Outliers Excluding Outliers 95% CI 95% CI Moderator Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Duration of Infertility Intercept 0.83920 0.4023 -1.38700 3.06550 0.67950 0.1233 -0.24830 1.60740 Duration -0.08705 0.6135 -0.44680 0.27270 -0.08716 0.2276 -0.23730 0.06296 Country Intercept 0.27130 0.1000 -0.06355 0.60620 0.25610 0.0175 0.05645 0.45580 Australia -0.24430 0.4170 -0.89400 0.40540 -0.09233 0.6076 -0.48490 0.30030 North America 0.30380 0.3883 -0.40570 1.01330 0.31990 0.1139 -0.08274 0.72240 South East Asia 1.08200 0.0160 0.25460 1.90940 South West Asia -0.35480 0.2301 -0.97830 0.26870 -0.33970 0.0708 -0.71490 0.03551 Note. A total of 8 studies representing 24 effect sizes were included in the analysis including outliers for the variable duration of infertility. A total of 7 studies representing 19 effect sizes were included in the analysis excluding outliers for the variable duration of infertility.

PAGE 206

192 Results of this analysis of the vari ance indicate that for duration of infertility, with outliers included in t he analysis, both the intercept and residual variances are statistically significant. However, when study outliers are excluded from the analysis, only the residual variance es timate is statistically significant at the .05 level. A comparison of the resi dual variance of the variable duration of infertility with the null model residual vari ance estimates reveals that the duration of infertility does not contri bute to the model or account for an increase in the amount of variability accounted for in the model. Theref ore, it appears that the variable duration of infertility does not ac t as a moderating variable. In the analysis of variance investi gating country in which the study was conducted as a possible moderator, only the residual varian ce is statistically significant for the analysis including study outliers and for t he analysis excluding study outliers. When comparing the variance estimates to the null model, the addition of the variable country in which the study wa s conducted slightly reduces both the intercept and residual variance estimates suggesting that the country in which the study is conducted may contribute to accounting for slightly more of the variability in this model. Table 18 Hierarchical random effects variance anal ysis of moderators for hypothesis one Including Outliers Excluding Outliers 95% CIa 95% CIa Moderator Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Duration of Infertilityb Intercept 0.25790 0.0695 0.09575 1.85750 0.02917 0.1820 0.00727 2.26800 Residual 2.95410 0.0019 1.65530 6.70360 0.77250 0.0065 0.40030 2.06890 Continued on the next page

PAGE 207

193Table 10 (Continued) Countryc Intercept 0.08274 0.1202 0.02566 1.38160 0.03235 0.1113 0.01034 0.45770 Residual 2.22190 <0.0001 1.42310 3.95040 0.61070 0.0003 0.37120 1.18830 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. The results provided by the significance test are considered to be reliable for interpretation. Note.bA total of 8 studies (24 effect sizes) contributed to the analysis including outliers and a total of 7 studies (19 effect sizes ) contributed to the analysis excluding outliers for the variable duration of infertility. cA total of 13 studies (43 effect sizes) contributed to the analysis including outliers and a total of 12 studies (36 effect sizes ) contributed to the analysis excluding outliers for the variable country. In order to test for moderating effects of the variable duration of infertility, a subset of studies were used due to many studies missing information regarding the duration of infertility fo r the women included in the sample. Therefore, a null model was constructed using this subset of studies for comparison purposes with the analysis for the variable duration of infertility as a moderator. Table 19 presents the null model estimates for t he intercept coefficient and Table 20 presents the results for t he variance components. T he results of this analysis show that the intercept is not statisti cally significant for the model including outliers as well as for the model excludi ng outliers. However, estimates for the residual variance component, are statistica lly significant for the model including outliers and for the model ex cluding outliers while the intercept variance estimate is not statistically signif icant for either model.

PAGE 208

194Table 19 Hierarchical random effects null model analy sis for studies reporting information on the duration of infertility for hypothesis one Including Outliers Excluding Outliers 95% CIa 95% CIa Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.36580 0.1024 -0.09467 0.82630 0.20730 0.0577 -0.00931 0.42390 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. The results provided by the significance test are considered to be reliable for interpretation. Note. A total of 8 studies representing 24 effect sizes we re included in the analysis including outliers for the variable duration of infertility. A total of 7 st udies representing 19 effect sizes were included in the analysis excluding outliers for the variable duration of infertility. Table 20 Hierarchical random effects null model variance analysis for studies reporting information on the duration of infertility for hypothesis one Including Outliers Excluding Outliers 95% CI 95% CI Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.23120 0.0619 0.08838 1.49220 0.03193 0.1385 0.00929 0.77780 Residual 2.92470 0.0018 1.64780 6.56570 0.78900 0.0053 0.41520 2.04250 Note. aA total of 8 studies representing 24 effect sizes we re included in the analysis including outliers for the variable duration of infertility. A total of 7 st udies representing 19 effect sizes were included in the analysis excluding outliers for the variable duration of infertility. Hypothesis Two The second part of this study investigated the relationship between psychoeducational interventions and the stress experienced by women participating in ART treatment regimens. S pecifically, this investigation sought to answer the following question: Do psyc hoeducational intervent ions mitigate the distress experienced by patients partici pating in an Assisted Reproductive

PAGE 209

195Technology (ART) treatment regimen? The first step in analyzing the relationship between psychoeducational inte rventions and stress was to compute an effect size. The metric used for this analysis was Cohen’s d providing a common scale for comparison of study outco mes. Table 21 lists the information extracted from each study as well as t he index of effect size and 95% Confidence Interval. Table 21 Studies reporting the effects of psychoeducat ional interventions as they relate to stress experienced during ART treatment regimens 95% CI Study N Country X Age Etiologya X Duration PIb Frequencyc Constructd d Lower Upper Connolly et al. (1993) 152 United Kingdom 32 1 5.4 years 1 1 1 .38 -0.0605 0.8171 Connolly et al. (1993) 152 United Kingdom 32 1 5.4 years 1 1 3 1.16 0.6928 1.6328 Emery et al. (2003) 282 Switzerland 34.4 0 3.8 years 1 1 1 Randomized 200 .32 -0.1729 0.8093 Non-Randomized 82 .26 0.0625 -0.2319 Emery et al. (2003) 282 Switzerland 34.4 0 3.8 years 1 1 2 Randomized 200 -.28 -0.7723 0.2087 Non-Randomized 82 .19 -0.3039 0.6747 Emery et al. (2003) 282 Switzerland 34.4 0 3.8 years 1 1 3 Randomized 200 .46 -0.0375 0.9503 Non-Randomized 82 .20 -0.2888 0.6900 McNaughtonCassill et al. (2002) 45 North America 34 0 5.75 years 3 2 1 0.80 0.1820 1.4010 McNaughtonCassill et al. (2002) 45 North America 34 0 5.75 years 3 2 3 0.10 -0.4955 0.6883 Terzioglu (2001) 90 Turkey 0 1 3 1 0.59 0.0771 1.1113 Terzioglu (2001) 90 Turkey 0 1 3 3 1.29 0.7367 1.8497 a Etiology is coded as follows: 0 = Unknown, 1 = Mix ed (combination of female factor and male factor) bPsychoeducational Intervention is coded as fo llows: 1=Counseling, 2=Support Group,3=Cognitive Behavioral Format cFrequency is coded as follows: 1=1-3 sessions, 2=4-6 sessions, 3=Other dConstruct is coded as follows: 1=Acut e Stress, 2=Chronic Stress, 3=Depression Note. Effect sizes are scored so that positive numbers reflect smaller amounts of stress in the group who participated in a psychoeducational intervention.

PAGE 210

196Figure 12 displays the 4 studies included in the meta-analysis for hypothesis two listed in Table 21, rank ordered by the magnitude of the effect size by the construct for which the measure reports. As shown in Figure 12, all of the studies are positive for the constructs acute stress and depression. However, one of the studies demonstrat es a small negative effect for the construct chronic stress. While the majority of studies had a small to moderate observed effect, two studies had a very large observed e ffect (1.29 and 1.16) for the construct depression. Figure 12. Dot plot for studies reporting the effect s of psychoeducational interventions on stress during ART treatment regimens -2.5-1.5-0.50.51.52.53.54.5 Acute Stress McNaughton-Cassill et al. (2002) Terzioglu (2001) Connolly et al. (1993) Emery et al. (2003) Emery et al. (2003) Subgroup Mean Chronic Stress Emery et al. (2003) Emery et al. (2003) Subgroup Mean Depression Terzioglu (2001) Connolly et al. (1993) Emery et al. (2003) Emery et al. (2003) McNaughton-Cassill et al. (2002) Subgroup Mean Grand Mean

PAGE 211

197To test data for outliers, the Samp le-Adjusted Meta-Analytic Deviancy (SAMD) statistic was computed using SAS 8.2. Programming code for this analysis appears in Appendix O while the resu lts are displayed in Table 22. In addition, scree plots were constructed to visually identify outliers and are presented in Appendix Q. Effect size s with a SAMD statisti c greater than 3.0 were considered extreme observations. Table 22 Test for outliers among effect sizes for hypothesis two Construct Study d SAMD Acute Stress McNaughton-Cassill et al. (2002) 0.796 2.18 Terzioglu (2001) .5942 1.77 Connolly et al. (1993) .3783 1.11 Emery et al. (2003) .3182 0.80 Emery et al. (2003) .2582 0.56 Chronic Stress Emery et al. (2003) -.2818 1.11 Emery et al. (2003) .1854 0.85 Depression Connolly et al. (1993) 1.1628 4.23 Terzioglu (2001) 1.2932 4.14 Emery et al. (2003) .4564 1.17 McNaughton-Cassill et al. (2002) .0964 0.18 Emery et al. (2003) .2006 0.14 Based on this analysis, two studies were identified as outliers: 1. Connolly et al. (1993) 2. Terzioglu (2001) These studies were examined for any differ ences in the populations or conditions that may account or explai n the extreme observations. In the study conducted by Terzioglu (2001), the treatment group received support from a nurse practitioner that worked with all of the participants throughout the duration of the

PAGE 212

198treatment. In addition, the nurse practitioner was present at the time of oocyte retrieval and embryo transfer. This was a distinct difference from the remainder of the studies who repor ted limited number of couns eling sessions with the patients. However, there were no notabl e differences in the study conducted by Connolly et al. (1993) as compared to the remainder of the studies. Next, a homogeneity test was conduct ed to determine whether the sample effect sizes for the 12 data points in the meta-analysis were homogenous. SAS 8.2 was used to perform the homogeneity te st (see Appendix P). As shown in Table 23, the Q -statistic for measures of depre ssion were significant at the .05 level, indicating that effect sizes ar e not homogeneous. Ex cluding the extreme observations as identified by the SAMD statistic, the Q -statistic was not significant at the .05 level for any of the measures, suggesti ng that the data are homogenous. Table 23 Test of homogeneity of effe ct sizes for hypothesis two Including Outliers Excluding Outliers Source df Q -Stat p df Q -Stat p Acute Stress 4 2.47318 0.64944 4 2.47318 0.64944 Chronic Stress 1 1.74720 0.18623 1 1.74720 0.18623 Depression 4 16.8918 0.00203 2 0.95471 0.62042 To analyze the effects of psychoeducat ional interventions on the stress experienced by participants in an ART tr eatment program, a r egression analysis was conducted (SAS code is presented in Appendix R). Two hypothesis tests

PAGE 213

199were conducted. First, the null hypot hesis for the regression coefficient q for any q = 0,1,2 is as follows: 0 :0 q H The regression coefficients represent the variable constr uct (acute stress and chronic stress). Results includi ng outliers and excluding outliers are presented in Table 24. Results including st udy outliers show a moderate positive relationship between psychoeducational inte rventions. However, neither the estimates for acute stress or chronic st ress was statistically significant. When the study outliers were excluded from t he analysis, results indicate that none of the coefficients are statistically significant at the .05 level. Table 24 Hierarchical random effects analysis for hypothesis two Including Outliers Excluding Outliers 95% CI 95% CI Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.6924 .03590 0.08591 1.29880 0.2681 0.1342 -0.15040 0.68660 Acute Stress -0.2285 0.3483 -0.77810 0.32110 0.1688 0.3620 -0.28710 0.62470 Chronic Stress -0.5885 0.1012 -1.33280 0.15580 -0.3157 0.1913 -0.87370 0.24220 Note. A total of 4 studies representing 12 effect sizes we re included in the analysis including outliers. A total of 4 studies representing 10 effect sizes were included in the analysis excluding outliers. The null hypothesis for the random effects variance: 0 :2 0 H was tested. This hypothesis was tested by computing a weighted least squares regression with weights equal to vi iw1. The weighted residual sum of squares was compared with the critical val ues of chi-square distribution with k – p – 1

PAGE 214

200degrees of freedom. As shown in Table 25, the residual variance estimate is significant at the .05 level for re sults including and excluding outliers. Table 25 Hierarchical random effects variance analysis for hypothesis two Including Outliers Excluding Outliers 95% CIa 95% CIa Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.03762 0.3007 0.00569 0.11096 0 Residual 1.87680 0.0349 0.80380 8.26460 0.73930 0.0307 0.32320 3.06240 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. Note. A total of 4 studies representing 12 effect sizes we re included in the analysis including outliers. A total of 4 studies representing 10 effect sizes were included in the analysis excluding outliers. SAS code for computing the coefficients and conducting the analysis is listed in Appendix S. Since construct (acute stress and chronic stress) did not significantly contribute to the model, a null model was constructed (see Appen dix R). Results of this analysis found that the intercept was not statistically significant when either study outliers were included or excl uded. These results indicate that while the estimates demonstrate a small pos itive effect of psychoeducational interventions on the stress experienc ed by women participating in an ART treatment program when outliers are exclud ed, this effect is not statistically significant (see Table 26). However, it is important to note that the number of studies included in this analysis was small, limiting the power of this analysis.

PAGE 215

201Table 26 Hierarchical random effects null model analysis for hypothesis two Including Outliers Excluding Outliers 95% CI 95% CI Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.53420 0.0589 -0.03736 1.10570 0.30710 0.0527 -0.00670 0.62090 Note. A total of 4 studies representing 12 effect sizes we re included in the analysis including outliers. A total of 4 studies representing 10 effect sizes were included in the analysis excluding outliers. In addition, the analysis of the variance es timates including outliers as well as excluding outliers reveal statistically sign ificant results for the residual estimates (1.9837 and 1.109, respectively) and are presented in Table 27. Table 27 Hierarchical random effects null model variance analysis for hypothesis two Including Outliers Excluding Outliers 95% CI a 95% CI a Parameter Estimate p-value Lower Upper Estimate p-value Lower Upper Intercept 0.07452 0.2171 0.01625 22.75280 0.00584 0.4230 0.00051 1.02300 Residual 1.98370 0.0187 0.92790 6.81030 1.10900 0.0275 0.49280 4.38120 aDue to different formulas used in the inferential statistics reported above, the 95% CI results may conflict with the significa nce test. Note. A total of 4 studies representing 12 effect sizes we re included in the analysis including outliers. A total of 4 studies representing 10 effect sizes were included in the analysis excluding outliers.

PAGE 216

202CHAPTER FIVE: CONCLUSIONS Purpose of Research The literature is replet e with the psychological impacts of infertility, providing evidence that at least some wo men who confront in fertility are at risk for heightened distress and depressive sympt oms. Although research over the past several decades documents the prev alence of distress and depressive symptoms in infertile women, research findings provide conflicting evidence regarding the effects of stress and depr ession on ART treatment outcomes. Based on mounting evidence about the stress of infertility and infertility treatment, some investigators have suggest ed that infertility treatment programs should incorporate a psychological tr eatment component. As early as 1959, acute psychological supports for infert ile couples as an adjunct to medical treatment can be found. Recently, national organizations such as Resolve, Inc. in the United States and the National A ssociation for the Childless in the UK offering referral support groups and infertility counseling have been established (Anderson & Alesi, 1997). Although the debat e regarding the impact of stress on ART treatment outcomes has yet to be resolved, it has been argued that biochemical treatment for infertility al ong with treatment for psychosocial stress could markedly improve overall reproduc tion outcomes (Wasser et al., 1999). These cultural trends along with the mount ing empirical evidence have led many

PAGE 217

203reproductive endocrinologists to acknowledge the importance of psychoeducational interventions, regardle ss of their impact on ART treatment outcomes. While several theoretical models postu late the effects of stress on ART treatment outcomes, a synthesis of t he accumulated data incorporating a qualitative assessment of the methodology of reviewed studies and a quantitative method of combining and analyz ing the data examining t he effects of stress on ART treatment outcomes was nonexistent until the conduct of this study. Although practitioners and researchers have postulated that psychoeducational interventions may provide an important com ponent to the treatment of infertility and may prove effective in preventing t he anticipated increase in psychological distress as the duration of in fertility increases, research on this topic is in its’ infancy. Therefore, the purpose of this meta-analysis was two-fo ld. The first aim of this study was to investigate t he impact of stress on the success of ART treatments through a review of the accumulated research. The second purpose of this study was to investigate the effi cacy of psychoeducational interventions in mitigating the impact of stress experi enced by women parti cipating in an ART treatment program. Overview of Method Four primary processes were incorporated into this study: 1. The formulation of the problem, 2. The collection of data and relevant research studies, 3. The evaluation of the data, and

PAGE 218

2044. The analysis and interpretation of the results. Based on the purposes identified for this synthesis, the two hypothesis were tested in this meta-analysis: 1. Increased levels of stress reduc e the likelihood of ART treatment success, and 2. Psychoeducational interventions mitigate the effects of stress experienced by women participati ng in an ART treatment program. Because it was expected t hat the studies to be incl uded in the analysis would differ from one another in study characteri stics and effect size parameter, both hypotheses were analyzed through a random effects model. Furthermore, a random effects model, asserting that the studies to be included in the analysis differ from the possible studies in t he universe as a consequence of sampling procedures, addresses one common criticism of primary resear ch investigating the relationship between stress and infertility treatment outcomes of homogeneous groups represented in the sample s. The conceptualization of this study implies that the st udies included in the analysis are different from one another in ways too complex to capture by the inclusion of simple study characteristics. Studies included in this meta-analysis were located through an exhaustive comprehensive search of the English languag e literature from January 1985 through December 2003. Electronic search es were conducted through a variety of databases including MEDLINE/PubM ed, ERIC, Psychinfo, and Dissertation Abstracts Online. Electronic branching fr om primary studies identified in this

PAGE 219

205search was also conducted to locate any additional studies similar in nature. Following the search of rele vant electronic databases, manual scans of reference lists and other publications was conducted to identify any additional relevant studies. In addition, a sear ch for registered clinical trials being conducted in the field of reproductive endocrinology was per formed. Finally, researchers current and active in this field research were contacted in an effort to locate any additional studies, published as well as unpub lished, left uncovered in the original search. Using this method, a total of 427 published research articles, one conference abstract, and three doctoral disse rtations were located for hypothesis one. Of these articles, a total of 32 published studies, one conference abstract, and two doctoral dissertations were empiri cal investigations of the relationship between stress and ART treatment outcomes. For hypothesis two, the literature search revealed a total of 62 published studi es and articles and one dissertation. Among these articles located, 21 publis hed articles and one doctoral dissertation reported the results of empirical invest igations of the relationship between psychoeducational interventions and stress. Once all studies were located through the search of the literature, this researcher reviewed each article to ensur e all studies meeting specified criteria for inclusion in the meta-analysis were identified. For each hypothesis, each study included in the meta-analysis met a ll of the following in clusion criteria: 1. For hypothesis one, the study mu st have involved situations where women were participating in an ART treatment program and the focus of the study was on the relationshi p between stress and ART treatment

PAGE 220

206outcomes. For hypothesis two, th e study must have involved women participating in a psychoeducational intervention program and an ART treatment program. The focus of the study was on the relationship between psychoeducational inte rventions and stress. 2. The study must have been cond ucted between January 1985 and December 2003 and prospective in design. 3. The study must report outcome m easures of stress or anxiety and treatment outcomes. For hypothes is one, treatment outcome was defined achieving pregnancy or failure to achieve pregnancy. For hypothesis two, treatment outcome was defined as the post-treatment score of the stress measures. 4. Studies must be prospective in design. The authors of studies meeting the abov e inclusion criteria but reporting insufficient data for effect size calculati ons were contacted in an effort to obtain all necessary data and information for inclus ion in the meta-analysis. Studies in which all necessary information for effect size calculations was received were included in the analysis while studies fo r which there remained insufficient data for effect size calculations were excluded. Of the total 35 empirical investigations located through the literature review, 13 (37%) studies met the criteria for inclusion in the meta-analysis for hypothesis one. For hypothesis two, a total of 4 (18%) of the 22 studies located through the literature review met inclusion criteria.

PAGE 221

207 Studies meeting the criteria for inclusion in the meta-analysis were subjected to a structured review of the qua lity of each study. This quality review applied the validity framework develope d by Campbell and his associates, providing a matrix of design and study f eatures (Wortman, 1994). A coding plan was developed to capture all rele vant indicators including: Design Participants Controls/Implementation Protocol Outcomes Statistics, producing a summary score to indicate the quality of each study. In addition to coding the quality of each study, characteri stics of interest in each study were also captured. Information including publication characteristics, ecological characteristics, methodological characte ristics, and results were recorded for each study. Both coding sheets, quality index and study characteristics coding forms, were subjected to a pilot test. Two professional evaluators current in the field of measurement and research were recr uited to complete this pilot test. A total of six articles included in the meta-analysis for hypothesis one and two articles included in the meta-analysis for hypothesis two were randomly selected for the pilot test. Interrater agreement was assessed by comparing the values recorded by each coder for each of the variables of interest. Raters were in agreement if all coders recorded identical values. Results of the pilot test

PAGE 222

208indicate a reasonably high interrater agreem ent rate. On the quality index form, the agreement rate was 92.65% while an agreement rate of 98.00% was found for the study characteristics form, resu lting in an overall agreement rate of 94.09%. Information collected on each st udy was then compiled and organized using a Microsoft Excel XP spreadsheet. Effect sizes were computed for each study. For hypothesis one, the effect size represent ed the magnitude of the influence of the influence of stress on AR T treatment outcomes. For hypothesis two, the effect size represented the magnitude of the influence of psychoeducational interventions on an individual’s level of stress while participating in an ART treatm ent program. The effect si ze metric computed for each analysis was Cohen’s d This effect size metric provided a common scale for comparison of study outcomes. Because of the complexity of the conceptualization of stress in infertility re search, three separate constructs were represented in the analysis: acute stress chronic stress, and depression. For studies reporting results from multiple measures of st ress representing the same construct, Cohen’s d effect sizes were calculated for each measure and then averaged to provide one statistic for each construct measured. In addition to multiple constructs, many studies r eported results at multiple endpoints throughout the study. Effect sizes for st udies providing measures at multiple points during the treatment program were computed separately and compared across studies using univariate analysis pr ocedures. For studies failing to report

PAGE 223

209relevant data and information to comput e an effect size for nonsignificant findings, an effect size of 0 was assumed. Statistical analysis of the resultin g effect sizes from each study was conducted with SAS 8.2. Results culled fr om each study were combined by each endpoint and construct to provide an overall es timate of effect. The overall mean and variance of each effect size was det ermined, weighted by the sample size and study quality, estimating T, the standardized mean difference. Next, a homogeneity test was conducted to det ermine whether the sample mean differences in each meta-analysis were homogenous. Once the overall mean and variance estimates across all studi es was computed, the data was analyzed to identify extreme observations. To identify possible study outliers, the SAMD statistic was calculated. Finally, a fully hierarchical regression model using a mixed effects linear regression model wa s constructed for each hypothesis. For hypothesis one, this model was construc ted utilizing the coefficients for each endpoint (baseline/pre-treatment, follicular phase, oocyte retrieval, embryo transfer, and follicular phase) and the coefficients computed for each construct (acute stress, chronic st ress, and depression) as well as the corresponding null model. For each meta-analysis, two hypot heses were tested. First, the null hypothesis for the regression coefficient q for any q = 0, p as follows: 0:0 q H In addition, the null hypothesis for the random effects variance was tested as follows:

PAGE 224

2100 :2 0 H Finally, a moderator analysis for the variabl es duration of infertility and country in which the study was performed fo r hypothesis one was conducted. Summary of Findings Effect sizes across studies invest igating the relationship between stress and ART treatment outcomes were vari ed, ranging from –0.630 to 1.412, exemplifying the ongoing controversy on this topic. However, the overall mean effect size, 0.2748, demonstr ates that there is a small relationship between stress and ART treatment outcomes indica ting that stress does indeed impact treatment success negatively. While the only construct consistently reported for each endpoint (baseline/pretreatment, follicular phase, oocyte retrieval, embryo transfer, and follicular phase) was acut e stress, results support Menning’s conceptualization of the crisis model app lied to infertility. These findings demonstrate that infertility, to some exten t, is a disruption in normal equilibrium. It is important to note that five st udies included in the meta-analysis for hypothesis one were identified as outliers. A sensitivity analysis showed that the inclusion of these studies in subse quent analyses did affect the results and implications. Therefore, a statistica l analysis was conducted and reported for analyses including the outliers as well as ex cluding outliers. Statistical analysis investigating the relationship betw een stress and ART treatment outcomes revealed that neither the time (baseline/ pre-treatment, follic ular phase, oocyte retrieval, embryo transfer, and luteal phase) nor construct (acute or chronic

PAGE 225

211stress) significantly contributed to the model regardless of whether the analysis included or excluded identified outliers. Therefore, a nu ll model, pooling all effect sizes, was constructed. While the anal ysis including study outliers for the null model indicated that stress did not have a statistically significant effect on ART treatment outcomes, analysis excluding out liers revealed that stress does have a small effect on ART treatment outcomes, indicating that higher levels of stress have a negative effect on ART treatment outcomes. An analysis of the variance estimates for the null model excluding out liers reveals that both the intercept estimate (0.05271) and residual estimate (0.6279) are st atistically significant indicating that the effect sizes pooled wi thin each study varied greatly, while there is very little variability among t he studies. A moderat or analysis was also conducted for the variables duration of in fertility and country in which the study was conducted. The results of this analysis demonstrated that, with study outliers excluded, neither the variable durat ion of infertility nor the variable country in which the study was conduct ed acted as a moderating variable. The Q -statistic for results culled for hypot hesis two revealed that there is a moderately positive effect ( d = 0.48052) of psychoeducational interventions in mitigating the effects of stress for women participat ing in an ART treatment regime. The SAMD statistic identified tw o studies as extreme observations. Therefore, the HLM regre ssion analyses including as well as excluding outliers were conducted and reported. Excluding outliers, the HLM regression analysis revealed that none of the coefficients co mputed were statistically significant. However, an analysis of the variance estimates revealed that the residual

PAGE 226

212variance estimate (0.7393) was statistically significant. Based on these results, the null model, pooling effect sizes, was c onstructed. This analysis reveals that while psychoeducational interventions dem onstrate a small positive effect on mitigating stress experienced by wom en participating in an ART treatment program, this coefficient was not statisti cally significant at the .05 level. However, the residual variance estimate (1.1090) was statistically significant, revealing a large amount of variability among the effect sizes pooled within each study. Limitations of Study One limitation to this study was the i nability to include m any of the primary studies that have been conducted to date due to insufficient amounts of data reported, specifically am ong the studies located for hypothesis one. This limitation manifested its’ significa nce when conducting the HLM regression analysis. Ideally, this analysis would in clude the interaction effects of time (baseline-pre-treatment, follicular phase, oocyte retrieval, embryo transfer, and luteal phase) and construct (acute st ress, chronic stress, and depression). However, due to the limitations in t he available data, this analysis was not possible. In addition to the exclusion of studies due to insufficient data, one study included in the meta-analysis fo r hypothesis one reported an insufficient amount of data to compute effe ct sizes for statistically nonsignificant results. In these instances, the value of the effect size was 0, providing a conservative estimate of these results.

PAGE 227

213For the analysis investigating t he efficacy of psychoeducational interventions on stress experienced by women participating an ART treatment program, the analysis as we ll as any generalizations that could be made was greatly limited because there was a total of only four studies meeting criteria for inclusion. Due to such a small number of studies included in the analysis, variables that may be operating as moderat ors could not be investigated in terms of their impact on findings. Because the investigation of psychoeducational interventions on the stress experienc ed by women participating in an ART treatment program is in its infancy, genera lizations from this study should only be made to the four studies in cluded in this analysis. Another limitation to this study incl uded the possibility of publication bias for studies investigating the effects of stress on ART treatment outcomes. As demonstrated in the funnel plot created, it appears that studies with smaller sample sizes reporting a negative effect, indicating that participants experiencing greater levels of stress were more likely to become pregnant following an ART treatment program, were not published. Th is could be a result of researchers, expecting stress to decrease the likeliho od of ART treatment success, attributing unexpected results to sampling error or to small sample sizes and therefore, failing to submit this research for publication. Implications Based on the empirical evidence pr oduced by this study, it appears that stress has a negative association with ART treatment outcomes. The value of the average effect size of 0.2012 indicates a small relationship. However, it is

PAGE 228

214important to consider the 95% confidence in terval that illustrates the possibility of this value to range between a very sma ll relationship of 0.03007 and a moderate relationship of 0.37240. These results s uggest that a purely biotechnological approach may not be successful in treati ng infertility for some women. In addition, in this analysis, it was found t hat the effects were consisitent across time and constructs measured. However, the residual variance was statistically significant, indicating that among t he effect sizes pooled within studies, significant variability exists. These results suggest the need for continued research aimed at describing theoreti cal models explaining this relationship between stress and ART treatment outcomes. Specifically, sufficient amounts of information allowing a research synthes ist to construct a regression model investigating the interaction effects between time (basel ine/pre-treatment, follicular phase, oocyte retrieval, embryo transfer, and luteal phase) and construct (acute stress, chronic stre ss, and depression) would be valuable. Although the results investigati ng the effects of psychoeducational interventions on the stress experienc ed by women participating in an ART treatement program were statistically insignificant, it is important to recognize the large amount of sampling error, as dem onstrated by the 95% confidence interval ranging from –0.00670 to 0.62090. In additi on, because the number of studies included in this analysis was very small, the power of the analysis was limited. Therefore, programs addr essing the stress experienced by some women participating in an ART treat ment program may not onl y be an ethical provision that reproduction endocrinologists provi de to their patients, but may also be an

PAGE 229

215additional variable to consider when the initial evaluation regarding the appropriateness of ART as a tr eatment option for a couple experiencing infertility is completed. Implications for Further Research Although this study has established that a statistically significant relationship exists between stress and AR T treatment outcomes, many questions still exist. The variance estimates of the analysis indicate that while effect sizes did not differ greatly across studies, ther e is a large amount of variability among the effect sizes within each study. The differences among these studies as well as the impact of other possible moderat ing variables such as the level of education, the number of previous ART treatment attempts and etiology of infertility remain unknown. Therefore, sci entists should continue research on this topic, incorporating an d reporting information regarding possible moderating variables and investigating explanatory models of these findings. Furthermore, it is critical that all data gathered through t hese empirical investigations be reported and made available, allowing research syn thesis studies to include all data accumulated. Finally, research synthes ists conducting future analysis of the relationship between stress and ART tr eatment outcomes may consider disaggregating the data further by the c onstruct measured. Specifically, while this study examined the effect s for the following constructs: Acute Stress, Chronic Stress and Depression,

PAGE 230

216researchers should consider disaggregati ng the data further by physiological measures (such as heart rate, etc.) and psychometric measures (such as the State-Trait Anxiety In ventory, etc.) of the stress response. Studies investigating the effects of psychoeducational interventions on the stress experienced by women participat ing in an ART treatment program continue to be important for the advancem ent of effective treatment programs. Furthermore, studies investigating the e ffects of a variety of psychoeducational interventions are needed. Studies investig ating the effects of psychoeducational interventions should include, but are not limited to, the following widely accepted formats for treating p sychological distress: Individual psychotherapy interventions, Cognitive behavioral approaches, Counseling programs (i ndividual and group), and Patient education programs and support groups. Studies implemented through a variety of designs such as pre-treatment counselling as well as interventions initiated before tr eatment begins and continuing through the conclu sion of the ART treatment regime would provide additional information regardi ng educational formats most effective in mitigating stress. Information valuable to the inte rpretation of these findings includes the frequency of the intervention, the time l ength of each session, the duration of the program, and format. Fi nally, women participating in a psychoeducational intervention may be more proactive about treatment options as well as less secretive about their infertility, building more substantial social supports and

PAGE 231

217networks. Studies incor porating varying levels of social support or other variables, such as locus of control (exter nal versus internal), which may confound results may help to identify any additional moderating variables. Finally, studies in this field of research should strive to triangulate methods, incorporating the valuable qualitative information published to date. An implication for further research re sulting from this study includes the investigation into a methodological iss ue. Specifically, the formulas used to calculate the confidence intervals in the Proc Mixed command in SAS. Conventional methods used for computing confidence intervals have been developed and are successfully used with univ ariate statistics. However, when applied to multivariate or mixed model s, these conventional methods for computing the confidence interval do not appear to be very effective. Specifically, using the Proc Mixed command in SAS, the computations for confidence intervals for the varian ce components appear to only include the value of 0 if the variance estimate is 0 it self. Therefore, t here would never be a nonzero variance component with a confidence interval including 0. Because the test of significance and the confidenc e intervals are computed differently, conflicting results may arise when the vari ance components are very small. As the use of mixed models and multivariate level statistics becomes more popular among researchers, applied researc hers should address this interesting methodological issue, providing alte rnative methods for constructing the confidence intervals in these situations.

PAGE 232

218Closing Remarks While to some, infertility represents a social state of childlessness, to many women and couples experiencing in fertility, it can be a devastating experience, causing great levels of st ress and turmoil. As the numbers of couples seeking treatment for infertility rises, the issues surrounding infertility represent a growing health problem While technological advancements continue to provide couples experiencing infertility with additional options, physicians should begin to consider and provide couples assistance with the complex issues and emotional reactions to infertility. Research aimed at the advancement of understanding the comple x relationship between stress and infertility, specifically ART treatment outcomes, should continue, providing physicians and couples with invaluable info rmation regarding the physiological as well as the emotional experience of infertility and infertility treatment.

PAGE 233

219REFERENCES Abma, J., Chandra, A., Mosher, W., Peterson, L., Piccinino, L. (1997). Fertility, family planning, and women’s heal th: New data from the 1995 National Survey of Family Growth. National Center for Health Statistics. Vital Health Stat 23 (19). Abarbanel, A.R & Bach, G. (1959). Group psychotherapy fo r the infertile couple. International Journal of Fertility 4, 151. Abbey, A., Halman, L. J., Andrews, F. M. (1992). Psychosocial, treatment, and demographic predictors of the stress associat ed with infertility. Fertility and Sterility, 57(1), 122-128. Allen, P.I.M., Batty, K.A., D odd, C.A.S, Herbert, J., Hugh, C.J., Moore, G.F., et al. (1985). Dissociation between em otional and endocrine responses preceding an academic examination in male medical students. Journal of Endocrinology, 107 (2), 163-170. American Society for Reproductive Medici ne. (2000-2003). Frequently asked questions about infertility. Retr ieved October 6, 2003, from http://www.asrm.org/Patients/faqs.html. American Society for Reproductive Medici ne. (2001). Patient’s fact sheet: Intracytoplasmic sperm injection (ICSI) Retrieved October 6, 2003, from http://www.asrm.org/Patients/ FactSheets/ICSI-Fact.pdf. Anderson, J. & Alesi, R. (1977). Infertility counseling. In Gabor T. Kovacs (Ed.), The subfertility handbook: a clinician’s guide Cambridge, United Kingdom: The Cambridge University Press. Andrews, F.M., Abbey, A., & Halman, L.J. (1992). Is fert ility-problem stress different? The dynamics of stress in fertile and infertile couples. Fertility Sterility, 57(6), 1247-1253. Andrews, L. & Hendricks, X. (1986). Legal and moral status of IVF/ET. In Christopher M. Fredericks, J.D. Paul sen, & Alan H. DeCherney. (Eds.), Foundations of in vitro fertilization Washington, DC: Taylor & Francis Inc. Appley, M. H. & Trumbull, R. (1977). On the concept of psychological stress. In Alan Monat & Richard S. Lazarus (Eds.), Stress and coping an anthology New York, NY: Columbia University Press. Arthur, W., Bennett, W., & Huffcutt, A.I. (2001). Conducting meta-analysis using SAS: multivariate applications Mahwah, N.J.: Erlbaum Associates.

PAGE 234

220Astor, J. & Pawson, M. (1986). The value of p sychometric testing in the investigation of infertility. Journal of Psychosomatic Obstetrics and Gynecology, 5, 107-111. Averill, J. R. (1983). Studies on anger and aggression: implications for theories of emotion. American Psychologist, 38 (11), 1145-1160. Baird, D. D. & Wilcox, A. J. (1985). Cigarette smok ing associated with delayed conception. JAMA, 253 (20), 2979 – 2983. Ball, S. (1995). Anxiety and test performance. In C.D. Spielberger & P.R. Vagg (Eds.), Test anxiety: Theory, a ssessment and treatment. Washington, D.C.: Taylor & Francis Group. Balmaceda, J. P., Manzur, A., & Asch, R. H. (1995). Ga mete intrafallopian transfer. In Edward E. Walla ch & Howard A. Zacur (Eds.), Reproductive medicine and surgery St. Louis, Missouri: Mosby-Year Book. Beck, A. T. & Em ery, G. (1985). Anxiety disorders and phobias New York, NY: Basic Books. Beck, A. T. Ward, D., Mendelson, M., Mock, J., & Erbaugh. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 53-63. Bell J. S. (1981). Psychological probl ems among patients att ending an infertility clinic. Journal of Psychosomatic Research 25 (1), 1-3. Berg, B. J. & Wilson, J. F. (1990). Psychiatric morbidity in the infertile population: a reconceptualization. Journal of Behavioral Medicine, 14 (1), 11-25 Berg, B. J. & Wilson, J. F. (1991). Psychological functioning across stages of treatment for infertility. Fertility Sterility, 53(4), 654-661. Berger, D. M. (1977). The role of the psychiatrist in a reproductive biology clinic Fertility Sterility 28 (2), 141-145. Bernstein, J., Mattox, J., & Kellner, R. (1988). Psychologi cal status of previously infertile couples after a successful pregnancy. JOGNN, 17(6), 404-408 Blake, P., Fry, R., & Pesjack, M. (1984). Self-assessment and behavior change manual New York, NY: Random House. Blenner, J. L. (1990). Passage th rough infertility: A stage theory. Imagethe Journal of Nursing Scholarship, 22 (3), 153-158. Boivin, J., Takefman, J. E., Tulandi, T. & Brender, W. ( 1995). Reactions to infertility based on extent of treatment failure. Fertility Sterility, 63(4), 801807. Boivin, J. & Takefman, J. E. (1995). Stress level across stages of in vitro fertilization in subsequently pregnant and nonpregnant women. Fertility and Sterility, 64 (4), 802-810.

PAGE 235

221Bresnick, E. & Taymor, M. L. (1979). The role of counselling in infertility. Fertility and Sterility, 32, 154-156. Breznitz, S. & Goldberger, L. (1993). Stress research at crossroads. In Goldberger, Leo & Breznitz, Shlomo (Eds.), Handbook of stress theoret ical and clinical aspects, second edition New York: The Free Press. Cahill, S. & Suchy S. (1981). The infertility resources handbook Kew, Vic.: The Citizen’s Welfare Service for Victoria. Cannon, W. B. (1939). The wisdom of the body New York, NY: W.W. Norton & Co. Centers for Disease Control and Prevention. (1997). Fertility, family planning, and women’s health: New data from the 1995 national survey of family growth Atlanta: Author. Centers for Disease Control and Prevention. (2002). Assisted reproduction technologies success rates Atlanta: Author. Center for the Study of Stress and Adaptation. (2003). Retriev ed December 3, 2003, from http://1sweb.1a.asu.edu/orchin ik/Orchinik_Lab_Web/stress_researsch_ center.html. Chamoun, D., McClamrock, H. D., & Adashi, E. Y. (1997). Ovulation initiation with clomiphene citrate. In Ma chelle M. Seibel (Ed.), Infertility: A comprehensive text, second edition Stamford, CT: Appleton and Lange. Chandra, A. Infertility. Division of Vital Statistics, National Centers for Health Statistics, Centers for Dis ease Control and Prevention Retrieved on August 6, 2003, from http://www.cd c.gov/nccdphp/drh/datoact/pdf/rhow4. pdf. Collins, J. A. (1997). Unexplained infertility. In M. Seibel (Ed.), Infertility a comprehensive text, second edition Stamford, CT: Appleton & Lange. Collins, J. A. & Rowe, T. C. (1989). Age of the fe male partner is a prognostic factor in prolonged unexplained infe rtility: A multicenter study. Fertility Sterility, 52(1), 15-20. Connolly, K. J., Edelmann, R. J., Cook e, I. D., & Robson, J. (1993). An evaluation of counseling for couples undergoing treatment for in-vitro fertilization. Journal of Psychosomatic Research, 36(5), 459-468. Cooper, H. M. (1982). Sci entific guidelines for conduc ting integrative research reviews. Review of Educational Research, 52, 291-302. Cooper, S. (1979). Female infertility: Its effect on self-esteem, body image, locus of control, and behaviour Unpublished PhD dissertation, Boston University School of Education.

PAGE 236

222Cooper, S. (1993). Paradise lost: sexual function and infertility. In Machelle M. Seibel, Ann A. Kiessling, Judith Be rnstein, & Susan R. Levin (Eds.), Technology and infertility: clinical, p sychosocial, legal, and ethical aspects New York, NY. Springer-Verlag New York Inc. Creach-Le Mer M. N.; Stoleru, S. G., Cornet; Zerah, S. ; Fermanian, J.; Bimbard, S.; & Spira, A. (1999). Women’s anxiety is a predi ctor of the implantation step of in vitro fertilization. Psychosomatic Medicine 61, 91. Csemiczky, G., Landgren, B., & Collins, A. (2000). The influence of stress and state anxiety on the outcome of IVF-treatment: Psychological and endocrinological assessment of Swedish women ent ering IVF-treatment. Acta Obstet Gynecol Scand, 79 (2), 113-118. Damewood, M. D. (1995) In vitro fertilization and assisted reproductive technologies. In Edward E. Wa llach & Howard A. Zacur (Eds.), Reproductive medicine and surgery St. Louis, Missouri: Mosby-Year Book. Daniluk, J. C. (1988). Infertility: intrapersonal and interpersonal impact. Fertility & Sterility 49 (6), 982-990. Demyttenaere, K., Nijs, P., Ev ers-Kiebooms, G., & Koninckx P. R. (1989). The effect of a specific emotional stressor upon prolactin, cortisone and testosterone concentrations in women va ries with their trait anxiety level. Fertility & Sterility, 52 (6), 942-948. Demyttenaere, K., Nijs, P., Evers-Kiebooms, G., & Konin ckx, P. R. (1991). Coping, ineffectiveness of coping and the psychoendocrinological stress responses during in-vitro fertilization. Journal of Psychosomatic Research, 35 (2/3), 231-243. Demyttenaere, K., Nijs, P., Evers-Kiebooms, G., & Konin ckx, P. R. (1992). Coping and the ineffectiveness of coping influence the outcome of in vitro fertilization through stress responses. Psychoneuroendocrinology, 17 (6), 655-665. Demyttenaere, K., Bonte, L., Gheldo f, M., Vervaeke, M., Meuleman, C., Vanderschuerem, D., & D’Hooghe, T. (1998). Coping style and depression level influence outcome in in vitro fertilization. Fertility and Sterility, 69 (6), 1026-1033. Dennerstein L. & Morse C. (1985). Psychological issues in IVF. Clinics in Obstetrics and Gynaecology 12 (4), 835-46. Derogatis, L. R. (1975). The Affects Balance Scale. Baltimore: Clinical Psychometric Research. Derogatis, L. R. (1987). The Derogatis Stress Profile (DSP): Quantification of psychological stress. In G. Fava & T. Wise (Eds.), Advances in psychosomatic medicine Basel: Karger.

PAGE 237

223Derogatis, L. R. & Coons, H. L. (1993). Self-report measures of stress. In Leo Goldberger and Shlomo Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects second edition New York, NY: The Free Press. Derogatis, L. R., Lipman, R.S., & Covi, L. (1973). SCL-90: An outpatient psychiatric rating scale: Preliminary report. Psychopharmacology Bulletin 9, 13-27. Deutsch, H. C. (1945). The psychology of women volume 2 New York, NY: Grune & Stratton. Dickersin, K. (1994). Research registers. In Harris Cooper and Larry V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Dise-Lewis, J. E. (1988). The life events and coping in ventory: An assessment of stress in children. Psychosomatic Medicine 50 (5), 484-489. Dlugi, A. M., Mersol-Barg, M. S., & Seibel, M. (1997) Gamete intrafallopian transfer. In Machelle M. Seibel (Ed.), Infertility a comprehensive text second edition Stamford, CT: Appleton and Lange. Dohrenwend, B. S. & Dohrenwend, B. P. (1980). Wh at is a stressful event? In Hans Selye (Ed.), Selye’s guide to stre ss research volume I New York, NY: Van Nostrand Reinhold Company. Dohrenwend, B. P., Raphael, K. G., Schwar tz, S., Stueve, A. & Skodol, A. (1993). The structured event probe and narrative rating method for measuring stressful life events. In Leo Goldberger & Shlomo Breznitz (Eds.), Handbook of stress: theoretical and clinical aspects: Second edition New York, NY: The Free Press. Domar, A. D. & Seibel, M. M. (1997). Emotional aspects of infertility. In Machelle M. Seibel (Ed.), Infertility: a comprehensive text, second edition Stamford, CT: Appleton and Lange. Domar, A. D., Seibel, M. M., & Benson, H. (1990). The mi nd/body program for infertility: a new behavioral treatment approach for women with infertility. Health Psychology, 19(6), 568-575. Domar, A. D., Broome, A., Zuttermeister P. C., Seibel, M., & Friedman, R. (1992). The prevalence and predict ability of depression in infertile women. Fertility and Sterility, 58(6), 1158-1163. Domar, Alice D., Zuttermeister, Patricia C., Seibel, Machelle, & Benson, Herbert. (1992). Psychological improvement in infertile women after behavioral treatment: a replication. Fertility & Sterility, 58(1), 144-147. Domar, A. D., Clapp, D., Slawsby, E., Kesse l, B., & Orav, J. (2000). The impact of group psychological interventions on distress in infertile women. Health Psychology, 19(6), 568-575.

PAGE 238

224Downey J., Yingling S., McKinney M., Hu sami N., Jewelewicz R., Maidman J. (1989). Mood disorders, psychiatric symptoms, and distress in women presenting for infertility evaluation. Fertility & Sterility, 52 (3), 425-432. Duffy, E. (1941). An ex planation of “emotional” pheno mena without the use of the concept of “emotion”. Journal of General Psychology, 25, 283-293. Edelmann, R. J. & Connolly, K. J. (1986). Psychological aspects of infertility. British Journal of Medical Psychology, 59, 209-219. Edwards, R. G. & Gardner, R. L. (1968). Choosing se x before birth. New Scientist, 38, 218-220. Ellertsen, B. Johnses, T. B., & Ursin, H. (1978). Relationship between hormonal responses to activation and coping. In H. Ursin, E. Baade, & S. Levine (Eds.), Psychobiology of stress – a study of coping men New York: Academic Press. Elliott, G. R. & Eisdorfer, C. (1982). Stress and human health: Analysis and implications of research. New York, NY: Springer Publishing Company. Ellis, A. (1977). Psychot herapy and the value of a human being. In A. Ellis & R. Grieger (Eds.), Handbook of rational-emotive therapy New York, NY: Springer. Facchinetti, F., Matteo, M. L., Artini, G. P., Volpe, A., & Gena zzani, A. R. (1997). An increased vulnerability to stress is associated with a poor outcome of in vitro fertilization-embryo transfer treatment. Fertility Sterility, 67(2), 309314. Flinter, F. A. (2001). Prei mplantation genetic diagnosis. British Medical Journal, 322 (7293), 1008-1009. Folkman, S. & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239. Folkman, S. & Lazarus, R. S. (1985). If it changes it mu st be a process: A study of emotions and coping. Journal of Personality and Social Psychology 48, 150170. Fowles, D. C., Christie, M. J., Edelberg, R., Grings, W. W., Lykken, D. T. & Venables, P. (1981). Publicatio n recommendations for electrodermal measurements. Psychophysiology 18, 232-239. Freeman E. W., Garcia C. R., Rickels K. (1983). Behavioral and emotional factors: comparisons of anovulatory infertile women with fertile and other infertile women. Fertility Sterility 40 (2), 195-201. Freeman E. W., Rickels K., Tausig J., Boxe r A., Mastroianni L. Jr., & Tureck R. W. (1987). Emotional and psychosoc ial factors in follow-up of women after IVF-ET treatment. A pilot investigation. Acta Obstet Gynecol Scand, 66 (6), 517-521.

PAGE 239

225Gallinelli, A., Roncaglia, R., Matteo, M. L., Ciaccio, I., Volpe, A., & Facchinetti, F. (2001). Immunological changes and st ress are associated with different implantation rates in patients undergoi ng in vitro fertilization-embryo transfer. Fertility Sterility 76(1), 85-91. Garner, C. H., Kelly, M., & Arnold, E. S. (1984). Psychological profile of IVF patients. Fertility & Sterility, 41, 57S. Goldstein, M., Rosenwaks, Z., & Davey, L. F. (2003). Micromanipulation techniques offer new hope for couples with male factor infertility. Presented at InterNational Council on Infertility Information Dissemination. Granvold, D. K. (1994). Concepts and methods of cognitive treatment. In Donald K. Granvold (Ed.), Cognitive and behavioral treatment Pacific Grove, CA: Brooks/Cole Publishing Company. Greenberg, J. S. (1990). Comprehensive stress management Dubuque, IA: William C. Brown. Greil, A. L. (1991). Not yet pregnant: Infertile couples in contemporary America. New Brunswick: Rutgers University Press. Greil, A. L. (1997). Infe rtility and psychological distress: A critical review of the literature. Social Science Medicine, 45 (11), 1679-1704. Handyside, A. H, Pattinson, J. K. Penketh, R. J., Delhan ty, J. D., Winston, R. M., Tuddenham, E. G. (1989). Biopsy of human preimplantation embryos and sexing by DNA amplification. Lancet, 1 (8634), 347-349. Harlow, C. R., Fahy, U. M., Talbot, W. M., Wa rdle, P. G., & Hull, M. G. R. (1996). Stress and stress-related hormones during in-vitro fertilization treatment. Human Reproduction, 11 (2), 274-279. Harper, J. C. (2001). Introduction. In Harper, Delhanty, Handyside (Eds.), Preimplantation genetic diagnosis London, UK: John Wiley & Sons. Hart, V. A. (2002). Infertilit y and the role of psychotherapy. Issues in Mental Health Nursing, 23, 31-41. Hathaway, S. R. & McKinley, J. C. (1940). A multiphasi c personality schedule (Minnesota). Part I: Constr uction of the schedule. J ournal of psychology 10, 249-254. Hedges, L. V. (1987). How hard is hard science, how soft is soft science? American Psychology 42, 443-455. Hedges, L. V. (1994). Statistical consi derations. In Harris Cooper and Larry V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Helgeson, V. S. & Cohen, S. (1996). Social support and adjustment to cancer: Reconciling descriptive, correlati onal, and intervention research. Health Psychology, 15 (2), 135-148.

PAGE 240

226Helsa, J. S. (1995). Homologous artificial insemination. In Edward E. Wallach & Howard A. Zacur (Eds.), Reproductive medicine and surgery. St. Louis, Missouri: Mosby-Year Book. Hoffert, S. P. (1997). Me ta-analysis gaining status in science and policymaking. The Scientist 11 (18), 1-5. Honea-Fleming, P. (1986). Psychosoc ial Components in Obstetric/Gynecologic conditions with a special consideration of infertility. The Alabama Journal of Medical Sciences, 23 (1), 27-35. Horowitz, M. J., Krupnick, J. Kaltreider, N., Wilner, N ., Leong, A., & Marmer, C. (1981). Initial psychological response to parental death. Archives of General Psychiatry 38, 316-323. Howe, G., Westhoff, C., Vesse y, M., & Yeates, D. (1985). Effects of age, cigarette smoking, and other factor s on fertility: Findings in a large prospective study. British Medical Journal, 290, 1697. Hughes, E. G. (1992). Meta-analysis and the cr itical appraisal of in fertility literature. Fertility and Sterility, 57 (2), 275-277. Hull, M. G.R, Glazener, C. M. A., Kelly, N. J., Conway, D. I., Foster, P. A., Hinton, R. A., Coulson, C., Lambert, P.A., Watt, E. M., & Desai, K. M. (1985). Population study of causes, treatment, and outcomes of infertility. British Medical Journal, 291, 1693-1697. Hunt, M. (1997). How science takes stock: T he story of meta-analysis. New York, NY: Russell Sage Foundation. Hunter, John E. & Schmidt, Frank L. (1990). Methods of meta-analysis: Correcting error and bias in research findings. Newbury Park, CA: Sage Publications. Hunter, John E. & Schmidt, Fr ank L. (1994). Correcting for sources of artificial variation across studies. In Harris Cooper and Larry V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Imeson, M. & McMurray, A. (1996). Couples’ experiences of infertility: A phenomenological study. Journal of Advanced Nursing, 24, 1014-1022. Jansen, R. P. S. (1993). Relative in fertility: Modeling cl inical paradoxes. Fertility and Sterility, 59, 1041-1045. Jenkins, C. D., Rosenman, R. H., & Fri edman, M. (1967). Development of an objective psychological test for t he determination of the coronary-prone behavior pattern in employed men. Journal of Chronic Diseases 20, 371 – 379.

PAGE 241

227Johnson, L., Neaves, W. B., Barnard, J. J., Keillore, G. E., Brown, S. W., & Yanagimachi, R. (1999). A compar ative morphological study of human germ cells in vitro or in situ within seminiferous tubules. Biology of Reproduction, 61, 927-934. Karahasanglu, A., Barglow, P., & Growe, G. (1972). Psychol ogical aspects of infertility. Journal of Reproductive Medicine, 9, 241. Katkin, E. S., Dermit, S. & Wine, S. K.F. (1993). Psychophysiological assessment of stress. In Leo Gol dberger & Shlomo Breznitz (Eds.), Handbook of stress: theoretical and clinical aspects: Second edition New York, NY: The Free Press. Kee, B. S.; Jung, B. J. & Lee, S. H. (2000). A study on psychological strain in IVF patients. Journal of Assisted Reproduction and Genetics 17(8), 445448. Klonoff-Cohen, H., Chu, E., Natarajan, L, & Sieber, W. (2 001). A prospective study of stress among women undergoing in vitro fertilization or gamete intrafallopian transfer. Fertility Sterility, 76(4), 675-687. Kovacs, G. T. & Vollenhoven, B. (1997). Cervical factor, unexplained subfertility and artificial insemination with husband sperm. In Gab Kovacs (Ed.), The subfertility handbook: A clinician’s guide New York, NY: Cambridge University Press. Laufer, N., Simon, A., Hu rwitz, A., & Glatstein, I. Z. (1997). In vitro fertilization. In Machelle M. Seibel (Ed.), Infertility a comprehensive text second edition Stamford, CT: Appleton and Lange. Lazarus, R.S. (1966). Psychological stress and the coping process New York, NY. McGraw-Hill. Lazarus, R. S. (1993). Coping theory and research: Past, present and future. Psychosomatic Medicine, 55 (3), 234-237. Lazarus, R. S, Cohen, J. B., Folkman, S., Kanner, A., & Schaef er, C. (1980). Psychological stress and adaptation: So me unresolved issues. In Hans Selye (Ed.), Selye’s guide to stress research New York, NY: Van Nostrand Reinhold Company. Linn, M. W. (1985). A Global Assessm ent of Recent Stress (GARS) scale. International Journal of Psychiatry in Medicine 15, 47-59. Lovely, L. P., Meyer, W. T., E kstrom, R. D., & Golden, R. N. (2003). Effects of stress on pregnancy outcome among women undergoing assisted reproduction procedures. Southern Medical Journal, 96 (6), 548-551. Lunenfeld, B. & Lunenfeld, E. (1997). Ovulation i nduction with human menopausal gonadotropin. In Machelle M. Seibel (Ed.), Infertility: A comprehensive text, second edition Stamford, CT: Appleton and Lange.

PAGE 242

228MacLachlan, V. (1997). The results of assisted reproductive technology. In Gab Kovacs (Ed.), The subfertility handbook: A clinician’s guide New York, NY: Cambridge University Press. Mahlstedt, P. P. (1985) The psychological component of infertility. Fertility and Sterility, 43 (3), 335-346. Mahlstedt P. P., Macduff S., Bernstein J. (1987). Emotional factors and the in vitro fertilization and embryo transfer process. Journal of In Vitro Fertilization Embryo Transfer: IVF 4 (4), 232-6. Mahlstedt, P. P. & Wood, S. M. (1995). Beyond conception: the psychological dilemmas of infertility and assisted repr oductive technology. In Edward E. Wallch & Howard A. Zacur (Eds.), Reproductive medicine and surgery St. Louis, Missouri: Mosby-Year Book, Inc. Mahoney, M. J. (1974). Cognition and beh avior modification. Cambridge, MA: Ballinger. Mandle, C. L., Jacobs, S., Arcari, P. M. & Domar, A. (1996). The efficacy of relation response interventions with adult patients: A review of the literature. Journal of Cardiovascular Nursing, 10 (3), 4-26. March, C. M. (1986). Ovulation induction. Ala. Journal of Med. Science, 23 (1), 31-35. Mason, J. W. (1975). Psychologic stress and endocrine function. In E.J Sachar (Ed.), Topics in Psychoendocrinology New York: Grune & Stratton. McCubbin, H. I., Larse, A. & Olson, D. H. (1981). Family Crisis Oriented Personal Evaluation Scales (F-COPES). St. Paul, MN: Family Social Sciences. McCubbin, H. I., McCubbin, M. A., Patterson, J. M., Cauble, A. E., Wilson, L. R. & Warwick, W. (1983). CHIP – Copi ng health inventory for parents: An assessment of parental coping patterns in the care of the chronically ill child. Journal of Marriage and the Family 45, 359 – 370. McCubbin, H. I., Patterson, J. M. & Wilson, L. (1981). The Family Inventory of Life Events and Changes (FILE) St. Paul, MN: Family Social Sciences. McEwen, B. S. & Mendelson, S. (1993). Effects of st ress on the neurochemistry and morphology of the brain: Counter-regulation versus damage. In Goldberger, Leo & Breznitz, Shlomo (Eds.), Handbook of stress theoretical and clinical aspects, second edition New York: The Free Press. McGrath, J. E. (1970). Social and psychological factors in stress New York, NY: McGraw-Hill, Inc. McGuire, L. S. (1975). The role of counseling in infertility. Postgrad Med, 57, 173.

PAGE 243

229McNair, D. M., Lorr, M. & Droppleman, L. F. (1971). Profile of mood states San Diego, CA: Educational and I ndustrial Testing Service. McNaughton-Cassill, M. E., Bostwick, J. M., Vanscoy, S. E., Arthur, N. J., Hickman, T. N., Robinson, R. D. et al. (2000). Development of brief stress management support groups fo r couples undergoing in vitro fertilization treatment. Fertility Sterility, 74(1), 87-93. Menning, B. E. (1980). The emot ional needs of infertile couples. Fertility and Sterility, 34(4), 313-319. Menning, B. E. (1982). The psyc hological impact of infertility. Nursing Clinics of North America, 17(1), 155 -163. Merari, D, Feldberg, D., Elizur, A., Goldman, J., & Modan, B. (1992). Psychological and hormonal changes in the course of in vitro fertilization. J Assist Reproduction Genet 9, 161-9. Miall, C. E. (1985). Perceptions of informal sanctioning and the stigma of involuntary childlessness. Deviant Behavior 6, 383. Millon, T., Green, C., & Meagher, R. (1982). Handbook of clinical health psychology. In Theodore Millon, Catherine Green, & Robert Meagher (Eds.), Handbook of clinical health psychology New York, NY: Plenum. Monat, A. & Lazarus, R. S. (1977). Stress and Coping: An Anthology New York, NY: Columbia University Press. Mosher, W. D. & Pratt, W. F. (1991). Fecundity and infertility in the United States: incidence and trends. Fertility and Sterility, 56(2), 192-193. National Center for Health Statistics. (2001). National survey of family growth (NSFG) survey description Retrieved March 11, 2003, from http://www.cdc.gov/nchs/about/m ajor/nsfg/nsfgback.htm. National Woman’s Health Informati on Center (NWHIC). (2001). Endocrinology and female infertility, record P-6401 Retrieved March 11, 2003, from http://216.205.53.178/endo/pubrelations/ patientInfo/infertility.htm. Newton, C. R., Hearn, M. T., & Yuzpe, A. A. (1990). Psyc hological assessment and follow-up after in vitro fertilization: assessing the impact of failure. Fertility Sterility, 54(5), 879-886. Nuojua-Huttunen, S., Tuomivaara, L., J untunen, K., Tomas, C., Kauppila, A., & Martikainen, H. (1995). Intrafollicular inseminat ion for the treatment of infertility. Human Reproduction, 10, 91-93. Nuojua-Huttunen, S., Tuomivaara, L., Junt unen, K., Tomas, C., Kauppila, A., & Martikainen, H. (1997) Comparison of fallopian tube sperm perfusion with intrauterine insemination in the treatment of infertility. Fertility and Sterility, 67, 939-942.

PAGE 244

230Nuojua-Huttunen, S., Tomas, C., Bloig, R ., Tuomivaara, L., & Martikainen, H. (1999). Intrauterine insemi nation (IUI) treatment in subfertility: an analysis of factors affecting outcome. Human Reproduction, 14, 698-703. Office of Technology Assessment. (1988). I nfertility: medical and social choices. Washington, DC: U.S. Go vernment Printing Office. Pearlin, L. I. (1993). Environmental and social sour ces. In Leo Goldberger & Shlomo Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects New York, NY: The Free Press. Pearlin, L. I., Lieberman, M. A., Menaghan, E. G. & Mull an, J. T. (1981). The stress process. Journal of Health and Social Behavior 22, 337-356. Pearson, L. H. (1992). Nursing Times, 88(1), 36-38. Peipert, J. F. & Bracken, M. B. (1997). Systematic re views of medical evidence: the use of meta-analysis in obstetrics and gynecology. Obstetrics and Gynecology, 89(4), 628-633. Pernoll, M. L. (2001). Benson & pernoll’s handbook of obstetrics and gynecology New York, NY: McGraw-Hill Medical Publishing Division. Phipps, W. R. (1996). The fu ture of infertility services. Fertility & Sterility, 66 (2), 202-204. Phipps, W. R., Cramer, D. W., Schiff, I., Stillman, R., Al brecht, B., Gibson, M. et al. (1987). The association betwe en smoking and female infertility as influenced by cause of the infertility. Fertility Sterility, 48 (3), 377. Piggott, T. D. (1994). Methods for hand ling missing data in research synthesis. In Harris Cooper and Larry V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Piotrowski, C. & Lubin, B. (1990). Assessment practices of health psychologist: Survey of APA division 38 Clinicians. Professional Psychology 21, 99106. Radloff, L. S. (1977). The CES-D Sc ale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1, 385 – 401. RESOLVE: The National Infert ility Association. (1998-2003). Health insurance coverage of infertility treatment Retrieved October 6, 2003, from http://www.resolve.org. Rosch, P. l. J. (2003). International congress on stress: How it all began Retrieved December 3, 2003, from http://www.stress.org/cong.htm. Rosenthal, M. C. (1994). The fugitive literature. In Harris Cooper and Larry V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation.

PAGE 245

231Rosenthal, R. (1978). How often are our numbers wrong? American Psychologist 33, 1005-1008. Rutherford, R. N.; Klemer, R. H., Banks, A. L., & C oburn, W. A. (1966). Psychogenic infertility from the male viewpoint. Pacific Med Surg 7, 434439. Ryan, M. A. (2001). Ethics and economics of assisted reproduction: The cost of longing Washington, DC: Georget own University Press. Salvatore, P., Garibaldi, S ., Offindani, A., Coppola, F., Amore, M., & Maggini, C. (2001). Psychopathology, personality, and marital relationship in patients undergoing in vitro fertilization procedures. Fertility Sterility, 75(6), 11191125. Sandelowski, M. & Pollock, C. (1986). Women’s experiences of infertility. Journal of Nursing Scholarship, 18(4), 140-144. Sanders, K. A. & Bruce, N. W. (1997). A prospective study of psychological stress and infertility in women. Human Reproduction, 12 (10), 2324-2329. Sanders, K. A. & Bruce, N. W. (1999). Psychosoc ial stress and treatment outcome following assisted reproductive technology. Human Reproduction, 14 (6), 1656-1662. Sarason, I. G., Johnson, J. H. & Siegel, J. M. Assessing the impact of life changes. In I.G. Sarason & C.D. Spielberger (Eds.), Stress and anxiety, volume 6 New York, NY: Wiley. Savage, J. (1989). Mourning unlived lives. A psychological study of childbearing loss Illinois: Chiron Publications. Schenker, J. G., Meirow, D., & Schenker, E. (1997). Stress and human reproduction. European Journal of Obstetrics & Gynecology and Reproductive Biology, 45, 1-8. Schindler, L. & Vollmer, M. (1984). Cognitive perspecti ves in behavioral marital therapy: Some proposals for bridging theor y, research, and practice. In K. Halweg & N.S. Jacobson (Eds.), Marital interaction: Analysis and modification New York, NY: Guilford Press. Schlegel, P. N. & Girardi, S. K. (1997). In vitro fe rtilization for male factor infertility. The Journal of Clinical Endoc rinology & Metabolism, 82(3), 709716. Schneider, M. G. (2000). The role of psychosocial fact ors in stress in infertile couples Unpublished doctoral dissertation, University of South Florida, Florida. Schoener, C. J. & Krysa, L. W. (1996). The comfort and discomfort of infertility. Journal of Gynecological Neonatal Nursing, 25 (2), 167-172.

PAGE 246

232Seibel, Machelle M. (1993). Medical evaluation and treatment of the infertile couple. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility clinical, psychosocial, legal, and ethical aspects New York: SpringerVerlag. Seibel, M. M. (1997). Ovul ation induction with follicle-stimu lating hormone. In M. M. Seibel (Ed.), Infertility: A comprehensive text, second edition Stamford, CT: Appleton and Lange. Seibel M. M., Seibel S. G. & Zilberstein M. (1994). Control of human sex ratios. Gender distribution--not sex selection. Human Reproduction 9 (4), 569 – 570. Seibel, M. M. & Taymor, M. L. (1982). Emotional aspects of infertility. Fertility and Sterility, 37(2), 137-145. Selye, H. (1936). A syndrome produced by diverse nocuous agents. Journal of Neuropsychiatry in Cli nical Neuroscience, 10(2), 230-231. Selye, H. (1956). The stress of life. New York: McGraw-Hill. Selye, H. (1977). Selections from the stre ss of life. In A. Monat & R. S. Lazarus (Eds.), S tress and coping an anthology New York, NY: Columbia University Press. Selye, H. (1983). The stress concept: Past present, and future. In C. L. Cooper (Ed.), Stress research: Issues for the eighties Chichester: Wiley. Selye, H. (1994). History of the stress concept. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress theoretical and clinical aspects, second edition New York: The Free Press. Shepherd, J. (1992). Stre ss management and infertility. Aust N Z J Obstet Gynaecol, 32(4), 353-356. Sherrod, R. A. (1995). A ma le perspective on infertility. MCN The American Journal of Maternal Child Nursing, 20 (5), 269-275. Skinner v Oklahoma. (1942). 316 US 535, 541. Slade, P., Emery, J., & Li eberman, B. A. (1997). A prospective longitudinal study of emotions and relationships in in-vitro fertilization treatment. Human Reproduction 12 (1), 183-190. Smeenk, J. M. J., Verhaak, C. M., Eugster, A., Minnen, A., Zielhuis, G.A. & Braat, D. D. M. (2001). The effect of anxiety and depression on the outcome of in-vitro fertilization. Human Reproduction, 16(7), 1420-1423. Sofikitis, N., Miyagawa, I., Sharlip, I. Hellstrom, W., Mekr as, G., & Mastelou Yonago, E. (2003). Human pregnanc ies achieved by intra-ooplasmic injections of round spermatid (RS) nucle i isolated from testicular tissue of azoospermic men. Paper presented at the American Urologic Association Conference, San Francisco, CA.

PAGE 247

233Speroff, L., Glass, R. H., & Kase, N. G. (1994). Clinical gynecologic endocrinology and infertility: Fifth edition Baltimore, MA: Williams and Wilkins. Spielberger, C. D. & Vagg, P. R. (1995). Test anxiety : A transactional process. In C.D. Spielberger & P.R. Vagg (Eds.), Test anxiety: Theory, assessment and treatment Washington, D.C.: Taylor & Francis Group. Spielberger, C. D., Gorsuch, R. C. & Lushene, R. E. (1970). Manual for the State-Trait Anxiety Inventory. Pao Alto, CA. C onsulting Psychologists Press. Spira, A. (1986). Epidemio logy of human reproduction. Human Reproduction, 1, 111-115. Stein, M. & Miller A. H. (1993). Stress, the hypothalam ic-pituitary-adrenal axis, and immune function. Advances in Experimental Medicine and Biology, 335, 1-5. Steirteghem, A. C, Liebaers, I., & Devr oey, P. (1997). A ssisted fertilization techniques. In M. M. Seibel (Ed.), Infertility a comprehensive text second edition Stamford, CT: Appleton and Lange. Stephen, E. H. (1996). Pr ojections of impaired fecundity among women in the United States: 1995-2020. Fertility and Sterility, 66(2), 205-209. Stephen, E. H., & Chandra, A. (1998). Updated projecti ons of infertility in the United States: 1995-2025. Fertility and Sterility, 70(1), 30-34. Stock, W. A. (1994). S ystematic coding for research synthesis. In H. Cooper and L. V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Strauss, B., Appelt, H., Bohnet, H. G., & Ulrich, D. (1992). Relationship between psychological characteristics and treat ment outcome in female patients from an infertility clinic. Journal of Psychosomatic Obstetrics and Gynaecology, 13, 121-133. Strelau, J. (1995). Te mperament and stress: temper ament as a moderator of stressors, emotional states, coping, and co sts. In C. D. Spielberger, I. G. Sarason, J. M. T. Beb ner, E. Greenglass, P. La ungani, & A. M. O’Rourke (Eds.), Stress and emotion: anxiety, anger, and curiosity, volume 15 Bristol, PA: Taylor & Francis. Talbot, J. & Lawrence, M. (1997). In-vit ro fertilization: indi cations, stimulation and clinical techniques. In G. Kovacs (Ed.), The subfertility handbook: A clinician’s guide New York, NY: Cambridge University Press. Templeton, A., Fraser, C., & Thompson, B. (1991). Infert ility – epidemiology and referral practice. Human Reproduction, 6, 1391-1394. The American Institute of Stress. (2003). Introduction. Retrieved December 3, 2003, from http://www.stress.org.

PAGE 248

234The Stress and Anxiety Resear ch Society. (2003). Introduction. Retrieved December 3, 2003, from http://star-society.org. Thiering, P., Beaurepaire, J., Jones, M., Saunders, D., & Tennan t, C. (1993). Mood state as a predictor of tr eatment outcome after in vitro fertilization/embryo transfer technology (IVF/ET). Journal of Psychosomatic Research, 37 (5), 481-491. Thoits, P. A. (1983). Dim ensions of life events that influence psychological distress: An evaluation and synthesis of the literature. In H. B. Kaplan (Ed.), P sychosocial stress: Trends in theory and research Orlando, FL: Academic Press, Inc. Thonneau, P., Marchard, S., Tallee, A., Fe rial, M., Ducot, B., Lansac, J., Lopes, P., Tabaste, J., & Spira, A. ( 1991). Incidence and main courses of infertility in a resident population of three French regions (1988-1989). Human Reproduction, 6, 811-816. Tietze, C. (1956). Statistical contribut ions to the study of human fertility. Fertility and Sterility, 7, 88-95. Tietze, C. (1968). Fertility after the discontinuation of in trauterine and oral contraception. International Journal of Fertility, 13, 385-389. Tsaltas, J. (1997). Introduc tion. In G. Kovacs (Ed.), The subfertility handbook: A clinician’s guide New York, NY: Cambridge University Press. Vaernes R, Ursin H, Darragh A, & Lam be R. (1982). Endocrine response patterns and psychological correlates. Journal of Psychosomatic Research 26 (2), 123-31. Veevers, J. (1980). Childless by choice. Butterworth, Toronto. Wasser, S. K. (1999). St ress and reproductive failure: an evolutionary approach with applications to premature labor. Am J Obstet Gynecol., 180(1 Pt 3), S272-274. Wasser, S. K., Sewall, G., & Soules, M. R. (1993). Psychosocial stress as a cause of infertility. Fertility and Sterility, 59 (3), 685-689. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negat ive affect: The PANAS Scales. Journal of Personality and Social Psychology 54, 1063 – 1070. Weinberger, D.A., Schwartz, G.E., & Davids on, R.J. (1979). Low-anxious, highanxious, and repressive coping styles: psychometric patterns and behavioral and physiological responses to stress. Journal of Abnormal Psychology, 88 (4), 369-380. White, H. D. (1994). Sc ientific communication and liter ature retrieval. In H. Cooper and L. V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation.

PAGE 249

235Whiteford, L. M. & Gonzalez, L. (1995). Stigma: The hidden burdens of infertility. Social Science Medicine, 40(1), 27-36. Wilson, J. F., & Kopitzke E. J. (2002). Stress and infertility. Current Womens Health Reports, 2(3), 194-199. Wood, C. (1997). The role of gamete intrafallopian trans fer. In G. Kovacs (Ed.), The subfertility handbook: A clinician’s guide New York, NY: Cambridge University Press. Wortman, P. M. (1994). Judging resear ch quality. In H. Cooper and L. V. Hedges (Eds.), The handbook of research synthesis New York, NY: Russell Sage Foundation. Wright J., Allard M., Lec ours A., & Sabourin S. (1989) Psychosocial distress and infertility: a review of controlled research. International Journal of Fertility 34 (2), 126-42. Zilberstein M., & Seibel M. M. (1994). Fertilizat ion and implantation. Current Opinion in Obstetrics and Gynecology 6 (2), 184-189.

PAGE 250

236BIBLIOGRAPHY Andersen, Y. C., Westergaard, L. G., Teisner, B., Byskov, A. G., Ziebe, S., Helledie, et al. (1992). Changes i nduced in serum protein profiles by ovarian stimulation during in vitro fe rtilization – embryo transfer treatment: a comparison between concept ion and non-conception cycles. Human Reproduction, 7(5), 585-591. Anonymous. (1991). Pain of childlessness. BMJ, Jun 1, 302(6788), 1345. Ardenti, R., Campari, C., Agazzi, L., & LaSala, G.B. (1999). Anxiety and perceptive functioning of infertile wo men during in-vitro fertilization: Exploratory survey of an Italian sample. Human Reproduction, 14 (12), 3126-3132. Arthur, Winfred Jr., Bennett, Winston Jr., & Huffcutt, Allen I. (2001). Conducting meta-analysis using SAS. Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers. Benazon N., Wright J., & Sabourin S. ( 1992). Stress, sexual satisfaction, and marital adjustment in infertile couples. Journal of Sex and Marital Therapy 18 (4), 273-284. Bernstein, J. (1993). Psychol ogical issues in infertility. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility: clinical, psychosocial, legal, and ethical aspects New York, NY: SpringerVerlag New York Inc. Boivin, J. (1997). Is there too much emphasis on psychosocial counseling for infertile patients? Journal of Assisted Reproduction and Genetics 14(4), 184 – 186. Brand, H. J. (1982). Psychological stress and infert ility. Part 2: Psychometric test data. British Journal of Medical Psychology, 55 (Pt. 4), 385-388. Brand, H. J., Roos, S. S., & van der Merw e, A. B. (1982). Psychological stress and infertility. Part 1: Psychophy siological reaction patterns. British Journal of Medical Psychology, 55 (Pt. 4), 379-384. Bringhenti, F., Martinelli, F., Ardenti, R., & LaSala, G.B (1997). Psychological adjustment of infertile women enter ing IVF treatment: differentiating aspects and influencing factors. Acta Obstet Gynecol Scand., 76(5), 431437.

PAGE 251

237Burke, L. (1993). The laser and infertility: Pr esent and future. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility: Clinical, psychosocial, legal, and ethical aspects New York, NY: Springer-Verlag. Chiba, H., Mori, I., Morioka, Y., Kashiwakur a, M., Nadaoka, T., Saito, H., et al. (1997). Stress of female infertility: Relations to length of treatment. Gynecology Obstetrics Investigation, 43, 171-177. Clapp, D. (1985). Emotional responses to infertility. Nursing interventions. Journal of Gynecological Neonatal Nursing, 14 (6 Suppl), 32s – 35s. Cohen, B. & Seibel, M. M. (1993). A hi storical perspective of obstetrics and gynecology: A backdrop for reproductive tec hnology. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility: Clinical, psychosocial, legal, and ethical aspects. New York, NY: Springer-Verlag. Epidemiology Data Service Center. ( 1998). In the spotli ght … the national survey of family growth. Datum, 4(4), 1-4. Ferin, M. (1999). Stress and the reproductive cycle. The Journal of Clinical Endocrinology & Metabolism, 84 (6), 1768-1774. Foulot, H. (1993). Hysteroscopy in infertility. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technol ogy and infertility: Clinical, psychosocial, legal, and ethical aspects. New York, NY: Springer-Verlag. Hawkins, T. (2002). Don’t turn to assisted reproduction too quickly warns US expert. National Institute of Enviro nmental Health Sciences. Retrieved August 12, 2003, from http://www. niehs.nih.gov/oc/news/dunson.htm. Hines, R. (2003). Stress and infertility. Southern Medical Journal, 96 (6), 533534. Hirsch, A. M. & Hirsch, S. M. (1995). The long-te rm psychosocial effects of infertility. JOGNN, 517-522. Holmes, T. H. & Rahe, R. H. (1967). The social readjustment rating scale. Journal of psychosomatic research 11, 213-218. Hunt, R. (1993). Operative laparoscopy: Pa tient selection and instrumentation. In Seibel, Machelle M., Kiessl ing, Ann A., Bernstein, Judith, & Levin, Susan R. (Eds.), Technology and infertilit y: Clinical, psychosoc ial, legal, and ethical aspects New York, NY: Springer-Verlag. Hunter, John E., Schmidt, Frank L., & Jacks on, Gregg B. (1982). Meta-Analysis: Cumulating research findings across studies. Beverly Hills, CA: Sage Publications. Institute of Medicine National Academy of Sciences. (1982). Conceptual issues in stress research. In G. R. Elliot, & C. Eisdorfer (Eds.), Stress and human health analysis and implications of research New York, NY. Springer Publishing Company.

PAGE 252

238Jordan, C., & Revenson, T. A. (1999) Gender differences in coping with infertility: A meta-analysis. Journal of Behavioral Medicine, 22 (4), 341 – 358. Kemeter, P. (1988). Studies on psychologic al implications of infertility – effects of emotional stress on fertilization and impl antation in in vitro fertilization. Human Reproduction, 3(3), 341-352. Mason J. W. (1975). A historical view of the stress field. Journal of Human Stress, 1 (2), 22-36. Mazure, & Greenfeld. (1989). Psychological studies of in vitro fertilization – embryo transfer participants. Journal of in vitro fertilization – embryo transfer, 6, 242-256. McNaughton-Cassill, M. E., Bostwick, J. M., Vanscoy, S. E., Arthur, N. J., Hickman, T. N., Robinson, R. D., Neal, G. S., & Bostwick, M. (2002). Efficacy of brief couples support gr oups developed to manage the stress of in vitro fertilization treatment. Mayo Clinic Proc, 77, 1060-1066. Microsoft Encarta Online Encyclopedia. (2003). Stress (psychology) Retrieved September 23, 2003, fr om http://enc arta.msn.com./ Milne, B. J. (1988). Couples’ experi ences with in vitro fertilization. JOGNN 347-352. Mori, E., Nadaoka, T., Morioka, Y., & Sait o, H. (1997). An xiety of infertile women undergoing IVF-ET: Relati on to the grief process. Gynecologic and Obstetric Investigation, 44, 157-162. Mosher, W. D., & Pratt, W. F. (1990). Fecundity and infertility in the United States, 1965-1988. Vital and Health Statistics of the National Center for Health Statistics, 192, 1-12. National Center for Health Statistics. (2002). American women are waiting to begin families Retrieved March 11, 2003, from http://www.cdc.gov/nchs/rel eases/02news/ameriwomen.htm. Negro-Vilar, A. (1993). Stress and other environmental factors affecting fertility in men and women: Overview. Environmental Health Perspectives, 101(Suppl. 2), 59-64. Newton, C. R., Sherrard, W., & Glav ac, I. (1999). The Fertility Problem Inventory: measuring perceived infertility-related stress. Fertility Sterility, 72(1), 54-62. Richards, C. J., Fallick, M. L., & Seibel, M. M. (1993). Game te intrafallopian transfer and ultrasound-guided transcervical fallopian tube canalization. In M. M. Seibel, A. A. Kiessling, J. Bernst ein, & S. R. Levin (Eds.), Technology and infertility: Clinical, psychosocial, l egal, and ethical aspects. New York, NY: SpringerVerlag.

PAGE 253

239Sandler, B. (1968). Emotional stress and infertility. Journal of Psychosomatic Research, 12, 51-59. Schenker, J. G., Meirow, D., & Schenker, E. (1992). Stress and human reproduction. European Journal of Obstetrics & Gynecology and Reproductive Biology, 45, 1-8. Selye, H. (1977). Selections for the st ress of life. In A. Monat, & R. Lazarus (Eds.), Stress and Coping: An Anthology New York, NY: Columbia University Press. Sharma, V., Allgar, V., & Rajkhowa, M. (2002). Factors influencing the cumulative conception rate and discont inuation of in vitro fertilization treatment for infertility. Fertility and Sterility, 78 (1), 40-46. Shepherd, J. (1992). Stre ss management and infertility. Aust NZ Journal of Obstetrics and Gynaecology, 32 (4), 353-356. Seibel M. M., Zilberstein M., & Kearnan M. (1995). In-vitro fertilisation and health care coverage. Lancet 345 (8941), 66. Silber, S. (1993). Surgical advances for ma le infertility. In M. M. Seibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility: Clinical, psychosocial, legal, and ethical aspects. New York, NY: Springer-Verlag. Singer, A. M. & Porter, D. H. (1993). Transcervical fall opian tube catheterization for diagnosis and treatment of fema le infertility. In M. M. Se ibel, A. A. Kiessling, J. Bernstein, & S. R. Levin (Eds.), Technology and infertility: clinical, psychosocial, legal, and ethical aspects New York, NY: Springer-Verlag. Terzioglu, F. (2001). In vestigation into effectiveness of counseling on assisted reproduction techniques in Turkey. Journal of Psychosomatic Obstetrics and Gynecology 22, 133-141. Vartiainen, H. (1990). E ffects of Psychosocial factor s, especially work-related stress, on fertility and pregnancy. Acta Obstet Gynecol Scand, 69, 677678. Verhaak, C. M., Smeenk, J. M., Eugster, A. van Minnen, A., Kr emer, J. A. M., & Kraaimaat, F. W. (2001). Stress and marital satisfaction among women before and after their first cycle of in vitro fertilization and intracytoplasmic sperm injection. Fertility & Sterility, 76(3), 525-531. Vital and Health Statistics. (1997). Fertility, fam ily planning, and women’s health: New data from the 1995 national survey of family growth. U.S. Department of Health and Human Serv ices, Centers for Disease Control and Prevention, National Cent er for Health Statistics, 1-101 Wang, M.C. & Buschm an, B.J. (1999). Integrating results through meta-analytic review using SAS software. Cary, N.C.: SAS Institute Inc. Wilchins, S. A. (1974). Use of group “rap sessions” in the adjunctive treatment of five infertile females. Journal of Med Soc NJ 71, 951.

PAGE 254

240Wright, J., Duchesne, C., S abourin, S., Bissonnette, F., B enoit, J., & Girard, Y. (1991). Psychosocial distress and in fertility: men and women respond differently. Fertility Sterility, 55(1), 100-108. Wright, V. C., Schieve, L. A., Reynolds, M. A. & J eng, G. (2003). Assisted reproductive technology surveillance – United States, 2000. Morbidity and Mortality Weekly Report, 52 (SS09), 1-16. Zion, A. B. (1988). Resour ces for infertile couples. JOGNN, 255-258.

PAGE 255

241 APPENDICES

PAGE 256

242Appendix A: Electronic Da tabases Employed in Computerized Data Search 1. MEDLINE/PubMed which is the U.S. National Library of Medicine’s premier bibliographic database that contains ov er 12 million references to journal articles in life sciences wit h a concentration on biomedicine. 2. ClinicalTrials.gov which provides current information about clinical research studies for both federally and privately, funded trials. 3. The Educational Resource s Information Center (ERIC) which includes Current Index to Journals in Education (CIJE), indexing journal articles of interest to professionals in educati on, and Resources in Education (RIE) identifying unpublished educat ional reports and projec ts of significance. 4. Psychinfo which is the online version of Psychological Abstracts providing information on published and unpublished work in psychology and related disciplines. 5. Dissertation Abstracts Online, which is the on-line version of Dissertation Abstracts International c ontaining all of the disse rtations, accepted from accredited American instituti ons in all subject areas.

PAGE 257

243Appendix B: Quality Review Form Table 3 Evaluation of the quality of individual studies Category Questions Response I. Design a) Is the design described? b) Is the design appropriate to the study questions? c) Is the design prospective? d) Are there clear inclusion and exclusion criteria for participants? e) Is there a description of each measure used and why it was chosen? f) Are experimental methods, such as treatment schedules, clearly defined? Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A II. Participants a) Did the subjects meet the inclus ion / exclusion criteria? b) Are demographics for all subject groups reported including: i. Age ii. Country or Race/Ethnicity iii. Economic Status iv. Employment Status v. Occupational Status vi. Education Level vii. Etiology of Infertility for infertile women viii. Classification of Infertility for infertile women ix. Duration of Infertility for infertile women x. Type of ART treatment received by infertile women xi. ART treatment History for infertile women Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Continued on the next page

PAGE 258

244Appendix B: (Continued) Table 3 (Continued) IV. Controls / Implementation a) If there are parallel controls, are they comparable to the experimental group? b) In a multi-group study, were the groups comparable at baseline for prognostic factors? Yes No N/A Yes No N/A V. Protocol a) Were the treatment regimens followed? b) Was the attrition rate low? c) Was the implementatio n of the treatment / intervention protocol clearly described including: i. Type of Treatment / Intervention ii. Duration of Intervention iii. Length of Time for each Intervention iv. Frequency of Intervention Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A Yes No N/A VI. Outcomes a) Are the outcome s clearly defined, including methods of measurement? b) Do the outcome meas ures answer the study questions? Yes No N/A Yes No N/A VIII. Statistics a) Are the an alytic methods clearly described b) Are the analytic methods appropriate for the data and study design? c) Are the summary statisti cs needed for the calculation of effect size in the paper or available from the investigator? d) Are non-significant statistics reported? Yes No N/A Yes No N/A Yes No N/A Yes No N/A

PAGE 259

245Appendix C: Study Character istics Important to Coding I. Publication Characteristics A. Title B. Year of Study C. Authors II. Ecological Characteristics A. Age B. Ethnicity C. Socio-Economic Status D. Etiology, Classification, and Duration of Infertility E. Type of ART Treatment and ART Treatment History F. Psychoeducational Intervention, Du ration and Frequency of Intervention III. Methodological Characteristics A. Sample Size B. Sampling Method D. Measure(s) of Stress or Anxiety E. Measure(s) of ART Success (pregnancy / live birth rate) F. Research Design IV. Results A. Group Means and Standard Deviations B. Significance Level C. Other: F statistic, Chi-square, Correlation, etc.

PAGE 260

246Appendix D: St udy Coding Sheet Study Coding Form Meta-Analysis Coding Part I: Increased leve ls of stress will r educe the likelihood of ART treatment success. STUDY TITLE: I. Qualifying the study: For each study, answer the following ques tions as either “yes” or “no”. 1. Does the study involve women participating in an ART treatment program? 2. Does the study focus on the relationship between stress and ART treatment outcomes? 3. Was the study conducted bet ween January 1985 and December 2003? 4. Does the study employ a prospective design? 5. Does the study report outcome measur es of stress or anxiety as well as ART treatment outcomes? If the answer to each of the above questions is yes, the study qualifies for inclusion in the meta-analysis.

PAGE 261

247Appendix D: (Continued) Study Coding Form Meta-Analysis Coding Part II: Psychoeduc ational interventions provided to patients receiving infertility treatment will mitigate the effects of stress during Assisted Reproductive Technology (ART) treatment STUDY TITLE: I. Qualifying the study: For each study, answer the following ques tions as either “yes” or “no”. 1. Does the study involve women participating in a psychoeducational intervention program and an ART treatment program? 2. Does the study focus on the re lationship between the psychoeducational intervention and stress? 3. Was the study conducted bet ween January 1985 and December 2003? 4. Does the study employ a prospective design? 5. Does the study report outcome measur es of stress by ca tegory (treatment group versus control group if applicable)? If the answer to each of the above questi ons is yes, the study qualifies for inclusion in the meta-analysis.

PAGE 262

248Appendix D: (Continued) II. Coding the study: Please provide the following information about the study. If the requested information is not available in the study documentation, leave it blank. Some questions require you to choose from a se t of options. Circl e all that apply. A. Publication Characteristics 1. Title of the study: 2. Year of Publication: 3. Authors: B. Ecological Characteristics 1. Age of Female Participants: Mean: Range: 2. Country: 3. Race: a. White N: %: b. Black N: %: c. Hispanic N: %: d. Asian / Pacific Islander N: %: e. American Indian N: %: f. Other N: %:

PAGE 263

249Appendix D: (Continued) 4. Economic Status By Category: a. Low N: %: b. Low Middle N: %: c. Middle N: %: d. Middle N: %: e. Upper Middle N: %: f. High N: %: By Range: a. < 21,000 N: %: b. $21,000 $40,000 N: %: c. $41,000 $60,000 N: %: d. > $60,000 N: %: 5. Employment Status a. Yes N: %: i. Part-Time N: %: ii. Full-Time N: %: b. No N: %: c. Unknown N: %: 6. Occupational Status a. Labor N: %: b. Secretarial N: %: c. Professional N: %:

PAGE 264

250Appendix D: (Continued) 7. Yrs Education (Fem ales): Mean: Range: a. Elementary / Intermediate N: %: b. Secondary N: %: c. Secondary Graduate N: %: d. Post-Secondary N: %: e. Higher Education Graduate N: %: 8. Etiology of Infertility a. Female Factor N: %: i. Peritoneal Factors N: %: ii. Endocrine Disorders N: %: iii. Mechanical Factors N: %: iv. Idiopathic N: %: b. Male Factor N: %: c. Combination N: %: 9. Classification of Infertility a. Primary N: %: b. Secondary N: %: 10. Duration of Infert ility: Mean: Range: a. 1 – 3 year N: %: b. 4 – 6 years N: %: c. 7 – 9 years N: %: d. > 9 years N: %:

PAGE 265

251Appendix D: (Continued) 11. Type of ART Treatment (Please Choose) a. Gamete Intrafallopian Transfer (GIFT) b. Zygote Intrafallopian Transfer (ZIFT) c. In Vitro Fertilization (IVF) d. Intra-Cytoplasmic Sperm Injection (ICSI) 12. Previous ART Treatment Attempts: Mean: Range: Reported by Number: a. 0 N: %: b. 1 N: %: c. 2 N: %: d. 3 N: %: e. 4 or > N: %: Reported by Ranges: a. 0 N: %: b. 1 2 N: %: c. 2 3 N: %: d. 3 4 N: %: e. > 4 N: %: 13. Psychoeducational Intervention (Please choose) a. Counseling b. Support Group c. Group Behavioral / Cognitive Behavioral Format d. Other:

PAGE 266

252Appendix D: (Continued) 14. Duration of Psychoeducational Intervention (Please choose) a. Daily for durati on of ART treatment b. 1 – 3 sessions during ART treatment c. 6 weeks during ART treatment d. 8 weeks during ART treatment e. 10 weeks during ART treatment f. Other: 15. Length of Psychoeducational Intervention (Please choose) a. 1 hour b. 1.5 hours c. 2 hours d. Other: 16. Frequency of Psychoeducational Intervention (Please choose) a. Daily b. Weekly c. Bi-Weekly d. Other:

PAGE 267

253Appendix D: (Continued) C. Methodological Characteristics 1. Measure used for assessment a. Anxiety Measures b. Depression Measures c. Mood State Measures d. Infertility Stress Measures e. Physiological Stress Response Measures 2. Time of Measure a. Anxiety Measures b. Depression Measures c. Mood State Measures d. Infertility Stress Measures e. Physiological Stress Response Measures

PAGE 268

254Appendix D: (Continued) 3. Sample Size of groups: Total N: a. 1 N: %: b. 2 N: %: c. 3 N: %: d. 4 N: %: 4. Sampling Method a. Random b. Convenience 5. Research Design a. Prospective b. Retrospective D. Results 1. Significance levels for findings: a. Measure: Time: Results: Time: Results: Time: Results: b. Measure: Time: Results: Time: Results: Time: Results:

PAGE 269

255Appendix D: (Continued) c. Measure: Time: Results: Time: Results: Time: Results: Time: Results: d. Measure: Time: Results: Time: Results: Time: Results: Time: Results: e. Measure: Time: Results: Time: Results: Time: Results: Time: Results: f. Measure: Time: Results: Time: Results: Time: Results: Time: Results:

PAGE 270

256Appendix D: (Continued) 2. Other results, such as t-tests, F statisti cs, correlations, etc. a. Measure: Time: Results: Time: Results: Time: Results: Time: Results: b. Measure: Time: Results: Time: Results: Time: Results: Time: Results: c. Measure: Time: Results: Time: Results: Time: Results: Time: Results: d. Measure: Time: Results: Time: Results: Time: Results: Time: Results:

PAGE 271

257Appendix D: (Continued) e. Measure: Time: Results: Time: Results: Time: Results: Time: Results: f. Measure: Time: Results: Time: Results: Time: Results: Time: Results: 3. Means and Standard Deviations Time: Time: Time: Group: Group: Group: Group: Group: Group: Biologic Measure X SD X SD X SD X SD X SD X SD SBP DBP HR Cortisol Prolactin Other: Other: Other:

PAGE 272

258Appendix D: (Continued) Time: Time: Time: Group: Group: Group: Group: Group: Group: Standard Measure Subscale X SD X SD X SD X SD X SD X SD PC Vital BDI Total I CPQ II IQ MMQ PANAS 1 2 3 4 5 6 POMS Total State STAI Trait I II III IV WOC V Other: Other:

PAGE 273

259Appendix E: Coding Manual I. Publication Characteristics A. Title – List the title of the Study B. Year of Publication – List the publication year C. Authors – List the author’s last names II. Ecological Characteristics A. Age – List the Mean age and range of ages for the females B. Country – List the country of where the study was performed C. Race – List the total number (N) and the percentage (%) for each category D. Economic Status – Record the total number (N) and the percentage (%) for either the ca tegory or by range E. Employment Status – Record the total number (N) and percentage (%) of the female participant’s employment during the study. For those who were employed, record the total num ber (N) and percentage (%) for either part-time or full-time. F. Occupational Status – Record the total number (N) and percentage (%) for each type of occupation G. Yrs Education (Females) – Record the mean and range of the number of years of education for the female parti cipants. If categorized according to level, record the total number (N ) and percentage (%) for the female participants by level.

PAGE 274

260Appendix E: (Continued) H. Etiology of Infertility – Record the reason for infertility. Female factors may be categorized by peritoneal (such as endometriosis or tubal factors), endocrine (such as anovulation), mechanica l (such as cervical or uterine factors), or idiopathic (unexplained / unknown). I. Classification of Infertility – Reco rd the total number (N) and percentage (%) of women participating in the study identified as ei ther primary or secondary infertility. (W omen who report one or more parity / children are classified as secondary. Women who report zero parity / children and one or more spontaneous or selective aborti on are classified as secondary. Women who report zero parity / childr en and zero abortions or previous pregnancies are classified as primary.) J. Duration of Infertility – Refers to the length of time from original diagnosis or the length of time trying to concei ve. Record the mean number of years and range of years. If categorized, record the total number (N) and percentage (%) for the appr opriate range of years. K. Type of ART Treatment – Circle all that apply for this study. L. Previous ART Treatment Attempts – Record the mean number and range of the total number of ART treatments that women have undergone prior to the study. If the number is repo rted by number or ranges, record the total number (N) and percentage (% ) for the appropriate category.

PAGE 275

261Appendix E: (Continued) M. Psychoeducational Intervention – Circle a ll that apply that to this study. If the type is not listed as a choice, please fill in the name of the intervention in the “other” blank provided. N. Duration of Psychoeducational In tervention – Circle how long the intervention is implemented during t he ART treatment program. If the duration is not listed as a choice opti on, please fill in the duration of the intervention in the “other” blank provided. O. Length of Psychoeducational Interv ention – Circle how long (length in terms of time) each intervention session la sts. If the approp riate length of time is not provided as a choice opti on, please fill in the length of the intervention in the “other” blank provided. P. Frequency of Intervention – Circl e how often the intervention is implemented. If the appr opriate frequency is not provided as a choice option, please fill in the length of the intervention in the “other” blank provided. III. Methodological Characteristics A. Measure(s) Used for assessment – Please list each measure used according to the construct it purports to represent. B. Time of Measure – Using the fo llowing guide, record the time the assessment was administered according to the construct. If the time is not listed below, please record “9” and list the time of the measure:

PAGE 276

262Appendix E: (Continued) 1. Baseline (beginning of cycle) bef ore administration of injections 2. Early follicular stage after in jections begin, but before oocyte retrieval 3. At the time of oocyte retrieval 4. At the time of embryo transfer 5. At the time of pregnancy test 6. 1 week following pregnancy test 7. 2 weeks following pregnancy test 8. 3 weeks following pregnancy test 9. Other C. Sample Size – Record the total num ber included in the sample. Using the following guide, record the total number (N) and percentage (%) for each group of comparison: 1. Conceived or Treatment / Intervention group 1 2. Failed to Conceive or Tr eatment / Intervention group 2 3. Treatment / Intervention group 3 4. Control Group D. Sampling Method – Choose whet her participants were purely a convenience sample or whether pa rticipants were randomized into treatment groups

PAGE 277

263Appendix E: (Continued) E. Research Design – Circle “prospective” if the study design included a sample that had not yet participated in the ART treatment program at the initiation of the study or “retrospec tive” if the participants completed the measures of stress after the AR T treatment program had already concluded or as a follow-up. IV. Results A. Significance levels for findings – Reco rd the measure used, the time of the measure as described above (1 – 9), and the results (alpha level) of the measure. If the study reports a measure only as “non-significant”, record the result as “NS”. B. Other results – This includes results from ANOVAs (F statistic), t-tests, chi-square, correlations, etc. Record the measure used, the time of the measure as described above (1 – 9), and all the results of the statistics provided. For example, if an AN OVA was performed, in the space provided for results, record the val ue of the F statistic, the degrees of freedom, and the alpha level. C. Means and Standard Deviations – Record information provided in the appropriate space of the table. St andardized assessments are listed first, followed by biological / physiological measures. For each result reported, record the time of the measure as de scribed above (1 – 9), the Group as described above (1 – 4), the mean and standard deviation.

PAGE 278

264Appendix F: Empirical Studi es Located for Meta-Analysis Empirical Studies Meeting Inclus ion Criteria for Hypothesis One 1. Ardenti, Rossella, Cam pari, Cinzia, Agazzi, Lorena, & La Sala, Giovanni Battista. (1999). Perceptive functi oning of infertile women during IVF: Exploratory survey of an Italian sample. Human Reproduction, 14(12), 3126-3132. 2. Boivin, Jacky & Takefman, Janet E. ( 1995). Stress level across stages of in vitro fertilization in subsequently pregnant and nonpregnant women. Fertility and Sterility 64(4), 802-810. 3. Csemiczky, Gyrgy, Landgren, Britt-Mari e, & Collins, Aila (2000). The influence of stress and state anxiety on the outcome of IVF-treatment: psychological and endocrinological assessment of Swedish women entering IVF-treatment. Acta Obstetricia et Gy necologica Scandinavica, 79, 113-118. 4. Demyttenaere, Koen, Bont e, L., Gheldof, M., Verv aeke, M., Meuleman, C., Vanderschuerem, D., & D’Hooghe, T. (1998). Coping style and depression level influence outcome in in vitro fertilization. Fertility and Sterility, 69(6), 1026-1033. 5. Demyttenaere, Koen, Nijs, Piet, Ever s-Kiebooms, Gerry, & Koninckx, Philippe. (1992). Coping and the ineffectiveness of coping influence the outcome of in vitro fertilization through stress responses. Psychoneuroendocrinology, 17(6), 655-665. 6. Facchinetti, Fabio, Volpe, Annibale, Matteo, Maria Lucia, Genazzani, Andrea R., & Artini, G. Paolo. (1997). An increased vulnerability to stress is associated with a poor outcome of in vitro fertilization-embryo transfer treatment. Fertility and Sterility, 67(2), 309-314. 7. Gallinelli, A., Roncaglia, R., Matteo, M. L., Ciaccio, I., Volpe, A., & Facchinetti, F. (2001). Immunological chan ges and stress are associated with different implantation rates in pat ients undergoing in vitro fertilizationembryo transfer. Fertility and Sterility 76(1), 85-91. 8. Kee, Baik Seok, Jung, Byeong Jun, & Lee, Sang Hoon. (2000 ). A study on psychological strain in IVF patients. Journal of Assisted Reproduction and Genetics, 17(8), 445-448. 9. Merari, Dalia, Feldberg, Dov, Eliz ur, Avner, Goldman, Jacob, & Modan, Baruch. (1992). Psychological and ho rmonal changes in the course of in vitro fertilization. Journal of Assisted Reprod uction and Genetics, 9(2), 161-169.

PAGE 279

265Appendix F: (Continued) 10. Merari, D., Feldberg, D ., Shitrit, A., Elizur, A ., & Modan, B. (1996). Psychosocial characteristics of women undergoing in vitro fertilization: a study of treatment outcome. Israel Journal of Obstetrics and Gynecology, 7(2), 65-72. 11. Sanders, K.A. & Bruce, N.W. (1999). Psycholog ical stress and treatment outcomes following ART. Human Reproducti on, 14(6), 1656-1662. 12. Thiering, P., Beaurepaire, J., Jones, M., Saunders, D. & Tennant, C. (1993). Mood state as a predictor of treatment outcomes after in vitro fertilization / embryo transfer technology (IVF/ET). Journal of Psychosomatic Research, 37(5), 481-491. 13. Verhaak, Christianne M., Smeenk, Jesper M.J., Eugster, Antje, van Minnen, Agnes, Kremer, Jan A.M., & Kraaimaa t, Floris W. (2001). Stress and marital satisfaction among women befor e and after their first cycle of in vitro fertilization and intracyt oplasmic sperm injection. Fertility and Sterility, 76(3), 525-531. Empirical Studies Excluded from Meta-Analysis for Hypothesis One Sample includes women not participating in an ART treatment program 1. Kemeter, Peter. (1988). Studies on p sychosomatic implications of infertility – effects of emotional stress on fertiliz ation and implantation in in-vitro fertilization. Human Reproduction, 3(3), 341-352. 2. Sanders, K.A. & Bruce, N.W. (1997). A prospecti ve study of psychosocial stress and fertility in women. Human Reproduction, 12(10), 2324-2329. 3. Vartianinen, Heikki. (1990). Effects of psychosocial factors, especially workrelated stress, on fertility and pregnancy: A prospective study from the stage of planning to become pregnant. (Doctoral dissertation, Kuopio University, 1989). Acta Obstet. Gynecol. Scand., 69, 677-678. Focus is not on the relationship bet ween stress and ART treatment outcome 4. Andersen, C. Yding, Westergaard, L. G., Teisner, B., Byskov, A.G., Ziebe, S., Helledie, L., Petersen, K., & West ergaard, J.G. (1992) Changes induced in serum protein profiles by ovarian stimulation during in-vitro fertilizationembryo transfer treatment: a co mparison between conception and nonconception cycles. Human Reproduction, 7(5), 585-591. 5. Boivin, Jacky & Takefman, Janet E. ( 1996). Impact of the in -vitro fertilization process on emotional, physical and relational variables. Human Reproduction, 11(4), 903-907.

PAGE 280

266Appendix F: (Continued) 6. Demyttenaere, Koen, Nijs, Piet, Evers-Kiebooms, Ge rry, & Koninckx Phillippe R. (1991). Coping, ine ffectiveness of coping and psychoendocrinological stress responses during in-vitro fertilization. Psychosomatic Research, 35(2/3), 231-243. 7. Kee, B.S., Jung, B.T., & Lee, S.H. (2000). A study on psychological strain in IVF patients. Journal of Assisted Reproduc tion and Genetics, 17(8), 445448. 8. Kowalcek, I., Kasimzade, T., & Huber G. (2003). Expectations for success in fertility treatment involving assisted reproduction. Arch Gynecol Obstet., 268(2), 78-81. 9. Milad, Magdy P., Klock, Susan C., Moses, Scott, & C hatterton, Robert. (1998). Stress and anxiety do not result in pregnancy wastage. Human Reproduction, 13(8), 2296-2300. 10. Phromyothi, V. & Virutamasen, P. (2003). The determinant factors and the anxiety level of infertile couples during the treatment of in vitro fertilization and embryo transfer at Chulalongkorn Hospital. Journal of Med. Assoc. Thai., 86(5), 425-429. 11. Salvatore, Paola, Gariboldi, Sim onetta, Offidani, Ada, Coppola, Francesco, Amore, Mario, & Maggini, Carlo. (2001). Psychopathology, personality, and marital relationship in patient s undergoing in vitro fertilization procedures. Fertility and Sterilit y, 75(6), 1119-1125. 12. Sharma, Vinay, Allgar, Victoria, & Rajk howa, M. (2002). Factors influencing the cumulative conception rate and disc ontinuation of in vitro fertilization treatment for infertility. Fertility and Ster ility, 78(1), 40-46. 13. Stolru, S., Cornet, D., Vaugeois, P., Fermanian, J., Magnin, F., Zerah, S., & Spira, A. (1997). The influence of psychological factors on the outcome of the fertilization step of in vitro fertilization. Journal of Psychosomatic Obstet. And Gynecol., 18, 189-202. 14. Young, P., Martin, C., & Thong, J. (2000). A comparison of psychological functioning in women at different stages of in vitro fert ilization treatment using the mean affect adjective check list. Journal of Assisted Reproduction and Genetics, 17(10), 553-556. Retrospective design 15. Beutel, M, Kupfer, J, Kirchmeyer P., Kehde, S., Khn, F.M., SchroederPrintzen, I., Gips, H., Herrero, H.J.G. & Weidner, W. (1999). Treatmentrelated stresses and depression in couples undergoing assisted reproductive treatment by IVF or ICSI. Andrologia, 31, 27-35.

PAGE 281

267Appendix F: (Continued) Insufficient data reported 16. Chreac’h-Le Mer, M.N., Stoleru, S.G., Cornet, D., Zerah, S., Fermanian, J., Bimbard, S., Spira, A. (1999 ). Women’s anxiety is a predictor of the implantation step of in vitro fertilization Paper presented at the American Psychosomatic Society 57th Annual Scientific Meeting, Vancouver, B.C., Canada. 17. Harlow, C.R., Fahy, U.M., Talbot, W.M. Wardle, P.G., & Hull, M.G.R. (1996). Stress and stress-related hormones during in-vitro fertilization treatment. Human Reproduction, 11(2), 274-279. 18. Klonoff-Cohen, Hillary, Chu, Elaine, Natarajan, Loki, & Sieber, William. (2001). A prospective study of st ress among women undergoing in vitro fertilization or gamete intr afallopian transfer. Fertility and Sterility, 76(4), 675-687. 19. Lovely, Laurie P., Meyer, William R., Ekstrom, David, & Golden, Robert N. (2003). Southern Medical Jour nal, 96(i6), 548-552. 20. Slade, P., Emery, J., & Lieberman, B.A. (1997). A prospective, longitudinal study of emotions and relationships in in-vitro fertilization treatment. Human Reproduction, 12(1), 183-190. 21. Smeenk, J.M.J., Verhaak, C.M., Eugster, A., van Minnen, A., Zielhuis, G.A., & Braat, D.D.M. (2001). The effect of anxiety and depression on the outcome of in-vitro fertilization. Human Reproduction, 16(7), 1420-1423. Paper unobtainable 22. Emery, Josephine Angela. (1993) Psychological aspects of IVF: a prospective study. Doctoral dissertati on, University of Manchester, United Kingdom. Empirical Studies Meeting Inclus ion Criteria for Hypothesis Two 1. Connolloy, Kevin J., E delmann, Robert J., Bartlett, Helen, Cooke, Ian D., Lenton, Elizabeth, & Pike, Sheila. (1993) An evaluation of counselling for couples undergoing treatment for in-vitro fertilization. Human Reproduction, 8(8), 1332-1338. 2. Emery, M., Bran, M.D., Darwiche, J., Oppizzi, L., Jo ris, V., Capel, R., Guex, P., & Germond, M. (2003) Results from a prospective, randomized, controlled study evaluating the acceptabi lity and effects of routine pre-IVF counselling. Human Reproduction, 18(12), 2647-2653. 3. McNaughton-Cassill, Mary Ellen, Bo stwick, J. Michael, Arthur, Nancy J., Robinson, Randal D., & Neal, Gregory S. (2002). Efficacy of brief couples support groups developed to manage the stress of in vitro fertilization treatment. Mayo Clin Proc., 77, 1060-1066.

PAGE 282

268Appendix F: (Continued) 4. Terzioglu, F. (2001). Investigation into effectiveness of counseling on assisted reproductive techniques in Turkey. Journal of Psychosomatic Obstetrics and Gynecology, 22, 133-141. Empirical Studies Excluded from Meta-Analysis for Hypothesis Two Sample includes women not participating in an ART treatment program 1. Domar, Alice D., Zuttermeister, Patr icia C., Seibel, Machelle, & Benson, Herbert. (1992). Psychol ogical improvement in infertile women after behavioral treatment: a replication. Fertility and Sterilit y, 58(1), 144-147. 2. Domar, Alice D., Seibel, Machelle M., & Benson, Herbert. (1990). The Mind/Body Program for in fertility: a new behaviora l treatment approach for women with infertility. Fertility and Sterility, 53(2), 2469-249. 3. Domar, Alice D., Slawsby, Ellen, Ke ssel, Bruce, Clapp, Diane, Oray, John, & Freizinger, Melissa. (2000). T he impact of group psychological interventions on distress in infertile women. Health Psychology, 19(6), 568-575. 4. Domar, Alice D., Clapp, Diane, Slawsb y, Ellen A., Dusek, Jeffery, Kessel, Bruce, & Freizinger, Melissa. (2000). The impact of group psychological interventions on pregnancy rates in infertile women. Fertility and Sterility, 73(4), 805-811. 5. Hsu, Y.L. & Kuo, B.J. (2002). Evaluations of em otional reactions and coping behaviors as well as correlated factor s for infertile couples receiving assisted reproductive technologies. Journal of Nursing Research, 10(4), 291-302. 6. Liswood, Margaret Peggi. (1993). Treat ing the crisis of in fertility: a cognitivebehavioral approach. Doctoral dissert ation, University of Toronto, Canada. 7. Lukse, M.P. (1985). The effect of group counseling on the frequency of grief reported by infertile couples. Journal of Obstetrics and Gynecology Neonatal Nursing, 14(6 Suppl), 67s-70s. 8. McQueeney, Debra A., Stanton, Annette L., & Sigmon, Sandra. (1997). Efficacy of emotion-focused and probl em-focused group therapies for women with fertility problems. Journal of Behavioral Medicine, 20(4), 313331. 9. O’Moore, A.M., O’Moore, R.R., Harri son, R.F., Murphy, G., & Carruthers, M.E. (1983). Psychosomatic aspects in idiopathic infertility: effects of treatment with autoge nic training. Journal of Psychosomatic Research, 27(2), 145-151. 10. Sarrel, Philip M. & DeCherney Alan H. (1985). Psychotherapeutic intervention for treatment of c ouples with secondary infertility. Fertility and Sterility, 43(6), 897-900.

PAGE 283

269Appendix F: (Continued) 11. Stewart, Donna E., Boydell, Katheri ne M., McCarthy, Karolina, Swerdlyk, Susan, Redmond, Carol, & Cohrs, Wilm a. (1992). A prospective study of the effectiveness of brief professional ly-led support groups for infertility patients. International Journal of Psychia try in Medicine, 22(2), 173-182. 12. Tuschen-Caffier, Bruna, Florin, Irme la, Krause, Walter, & Pook, Martin. (1999). Cognitive-behavioral therapy for idiopathic infertile couples. Psychotherapy and Psychosomatics, 68, 15-21. Sample includes only men 13. Lottman, P.E., Hendriks, J.C., Vruggink P.A., & Meuleman, E.J. (1998). The impact of marital satisfaction and p sychological counselling on the outcome of ICI-treatm ent in men with ED. International Journal of Impotence Research, 10(2), 83-87. Focus of the outcome m easure(s) is not stress 14. Bartlam, Bernadette & McLeod, John. (2000). Infertility counselling: the ISSUE experience of setting up a telephone counselling service. Patient Education and Counse ling, 41, 313-321. 15. Freeman, E.W., Bo xer, A.S., Rickels, K., Tureck, T., & Mastroianni, L. Jr. (1985). Psychological evaluation and s upport in a program of in vitro fertilization and embryo transfer. Fertility and Sterilit y, 43(1), 48-53. 16. McNaughton-Cassill, Mary Ellen, Bo stwick, Michael, Vanscoy, Sara E., Arthur, Nancy J., Hickman, Timothy N., Robinson, Randal D., & Neal, Greg S. (2000). Development of br ief stress management support groups for couples undergoing in vitr o fertilization treatment. Fertility and Sterility, 74(1), 87-93. 17. Pengelly, Paul, Inglis, Margaret, & Cudmore, Lynne. ( 1995). Infertility: Couples’ experiences and the use of counselling in treatment centers. Pyschodynamic Counselling, 1, 507-524. Wischmann, T. Stammer, H., Scherg, H., Gerhard, I., & Verres, R. (2001). P sychosocial characteristics of infertile couples: a study by t he “Heidelberg Fertility Consultation Service’. Human Reproduction, 16(8), 1753-1761. Retrospective design 18. Weaver, Susan M., Clifford, Ellen, Ha y, Douglas, & Robinson, John. (1997). Psychosocial adjustment to unsucce ssful IVF and GIFT treatment. Patient Education and Couns eling, 31, 7-18.

PAGE 284

270Appendix G: Conceptual De finitions of Constructs Acute stress, defined as the moment ary state experienc ed during the ART treatment program, was repr esented using the following psychological measures: 1. Stait-Trait Anxiety Inventory (STAI) – state anxiety subtest, 2. Infertility Questionnaire, 3. Daily Infertility Records, and 4. POMS – tension / anxiety subtest. The physiological measures of acute stress included: 1. Cortisol, 2. Prolactin, 3. Systolic Blood Pressure, 4. Diastolic Blood Pressure, 5. Heart Rate, 6. 6-SM, 7. FSH, and 8. Estradiol. Chronic Stress is conceptualized in this study as anxiety or stress persisting over a long duration of time. T he following were included in the analysis as measures of chronic stress: 1. STAI – trait anxiety subtest Depression is conceptualized in this study as negative em otion of extreme sadness.

PAGE 285

271Appendix G: (Continued) The following were included in the anal ysis as measures of depression: 1. Becks Depression Inventory, 2. Center for Epidemiology St udies Depression Scale (CES-d), 3. Zung Depression Inventory, 4. DACL, and 5. POMS – depression / dejection subtest.

PAGE 286

272Appendix H: SAS Code for Combining Effect Sizes TITLE 'COMBINING STANDARDIZED MEAN DIFFERENCES BY STRESS MEASURES AND TIME' ; DATA ORIG; INPUT ID 1 2 CONSTRUCT 4 TIME 6 AGE 8 COUNTRY 10 ETIOLOGY 12 DURATION 14 D 16 20 .3 N1 22 24 N2 26 28 QI 30 32 .2 ; T1 = 0 ; T2 = 0 ; T3 = 0 ; T4 = 0 ; T5 = 0 ; C1 = 0 ; C2 = 0 ; C3 = 0 ; ARRAY CON CONSTRUCT; ARRAY TIM TIME; DO OVER TIM; IF TIME = 1 THEN T1 = 1 ; IF TIME = 2 THEN T2 = 1 ; IF TIME = 3 THEN T3 = 1 ; IF TIME = 4 THEN T4 = 1 ; IF TIME = 5 THEN T5 = 1 ; END ; DO OVER CON; IF CONSTRUCT = 1 THEN C1 = 1 ; IF CONSTRUCT = 2 THEN C2 = 1 ; IF CONSTRUCT = 3 THEN C3 = 1 ; END ; T1C1 = T1*C1; T1C2 = T1*C2; T1C3 = T1*C3; T2C1 = T2*C1; T2C2 = T2*C2; T2C3 = T2*C3; T3C1 = T3*C1; T3C2 = T3*C2; T3C3 = T3*C3; T4C1 = T4*C1; T4C2 = T4*C2; T4C3 = T4*C3; T5C1 = T5*C1; T5C2 = T5*C2; T5C3 = T5*C3; +------------------------------------+ COMPUTE CI FOR EACH STUDY +------------------------------------+; V=((N1+N2)/(N1*n2))+((D** 2 )/( 2 *(N1+N2))); LOWER = D ( 1.96 *SQRT(V)); UPPER = D + ( 1.96 *SQRT(V)); +---------------------------------------------------+ WEIGHT STUDIES BY QUALITY INDEX & SAMPLE SIZE +---------------------------------------------------+;

PAGE 287

273Appendix H: (Continued) W = 1 /V; QISQ = QI** 2 ; QIW = QI*W; QISQW = QISQ*W; QIWD = QI*W*D; DSQ = D** 2 ; WDSQ = W*DSQ; WD = W*D; WSQ = W** 2 ; N = N1 + N2; CARDS ; 01 1 3 0 3 1 0 0.459 011 029 .85 01 2 3 0 3 1 0 0.080 011 029 .85 03 1 1 2 3 0 1 -.071 017 023 .79 03 1 2 2 3 0 1 0.573 017 023 .79 03 1 3 2 3 0 1 0.806 017 023 .79 03 1 4 2 3 0 1 0.643 017 023 .79 03 1 5 2 3 0 1 0.866 017 023 .79 03 2 1 2 3 0 1 0.190 017 023 .79 06 1 1 1 2 0 2 1.412 047 091 .87 06 2 1 1 2 0 2 1.787 047 091 .87 06 3 1 1 2 0 2 0.949 047 091 .87 08 1 1 2 1 1 0 -.630 082 230 .92 08 2 1 2 1 1 0 -.061 082 230 .92 08 3 1 2 1 1 0 -.117 082 230 .92 09 1 1 2 3 1 1 0.199 039 088 .81 09 3 1 2 3 1 1 0.326 039 088 .81 10 1 3 2 3 1 2 -.616 038 162 .98 10 2 3 2 3 1 2 0.034 038 162 .98 11 1 1 1 1 1 0 0.498 032 024 .77 11 2 1 1 1 1 0 0.579 032 024 .77 12 1 1 2 3 2 0 0.865 009 013 .81 12 1 2 2 3 2 0 0.598 009 013 .81 12 1 5 2 3 2 0 1.528 009 013 .81 13 1 1 2 3 1 2 0.241 016 033 .88 13 1 3 2 3 1 2 0.437 016 033 .88 13 2 1 2 3 1 2 0.629 016 033 .88 14 1 2 1 3 1 2 0.473 010 030 .82 14 1 3 1 3 1 2 0.190 010 030 .82 14 1 4 1 3 1 2 0.000 010 030 .82 14 3 2 1 3 1 2 2.723 010 030 .82 20 1 2 0 2 1 0 -.212 023 090 .93 20 1 3 0 2 1 0 -.177 023 090 .93 20 1 4 0 2 1 0 0.065 023 090 .93 20 1 5 0 2 1 0 0.107 023 090 .93 20 2 1 0 2 1 0 0.048 023 090 .93 20 3 2 0 2 1 0 -.328 023 090 .93 20 3 3 0 2 1 0 -.306 023 090 .93 20 3 4 0 2 1 0 0.067 023 090 .93 20 3 5 0 2 1 0 -.007 023 090 .93 25 3 1 1 3 1 1 0.041 023 075 .91 28 1 1 2 2 1 3 0.024 023 090 .87

PAGE 288

274Appendix H: (Continued) 28 2 1 2 2 1 3 0.050 023 090 .87 28 3 1 2 2 1 3 -.332 023 090 .87 ; DATA FIRST; SET ORIG; IF T1C1 = 1 ; PROC PRINT DATA = FIRST; TITLE 'SUMMARY STATISTICS FOR D' ; VAR ID D V N TIME CONSTRUCT LOWER UPPER QI W QISQW QIW QIWD WDSQ WD; PROC MEANS SUM N DATA = FIRST; VAR N W WSQ WDSQ WD; +----------------------------------------------+ WRITE OUT SUMMARY STATISTICS FOR VARIANCES +----------------------------------------------+; OUTPUT OUT = COMBINE SUM = Total_N SW SWSQ SWDSQ SWD MEAN = MNN N = N1; PROC PRINT; VAR SW SWSQ SWDSQ SWD MNN N1; DATA COMBINE; SET COMBINE; +----------------------------------------------+ COMPUTE UNCONDITIONAL VARIANCE COMPONENT +----------------------------------------------+; C = (SW -((SWSQ/SW))); Q = SWDSQ ((SWD** 2 )/SW); ProbQ = 1 probchi(Q, N1 1 ); DF = N1 1 ; BTWNV = (Q (N1 1 ))/C; EQ = C*BTWNV + (N1 1 ); PROC PRINT; VAR C Q DF PROB_Q BTWNV EQ; +----------------------------------------------+ COMPUTE TOTAL WEIGHTED VARIANCE +----------------------------------------------+; DATA TWO; IF _N_ = 1 THEN SET COMBINE; SET FIRST; WW = 1 /(BTWNV + V); WV1 = QISQ*WW; WV2 = QI*WW; WV2B = WV2** 2 ; IWV = WV1 / WV2B; WQIWD = QI*WW*D; PROC PRINT; TITLE 'SUMMARY STATISTICS FOR WEIGHTED D UNDER RANDOM EFFECTS MODEL' ; VAR ID D V LOWER UPPER IWV QI W WW WV1 WV2 WV2B WQIWD C Q BTWNV EQ; PROC MEANS SUM N ; VAR N WW WV1 WV2 WQIWD;

PAGE 289

275Appendix H: (Continued) +----------------------------------------------+ WRITE OUT SUMMARY STATISTICS TO COMBINE +----------------------------------------------+; OUTPUT OUT = RESULTS SUM = Total_N SWW SWV1 SWV2 SWQIWD MEAN = MNN N = N1 N2 N3; DATA REFINAL; SET RESULTS; DPLUS = SWQIWD/SWV2; WWV = SWV1/(SWV2** 2 ); VERROR = ((MNN1 )/(MNN3 ))*( 4 /MNN)*( 1 +((DPLUS** 2 )/ 8 )); UNBIASV = WWV VERROR; IF UNBIASV LT 0 THEN UNBIASV = 0 ; A = 1 +( 0.75 /(MNN3 )); SE = SQRT(WWV); DPLUSC = DPLUS/A; SEC = SE/A; CLL = DPLUSC ( 1.96 *SEC); CUL = DPLUSC + ( 1.96 *SEC); PROC PRINT ; VAR SWV1 SWV2 WWV VERROR UNBIASV MNN A SE DPLUS DPLUSC SEC; PROC PRINT ; TITLE 'RANDOM EFFECTS WEIGHTED AVERAGE D AND VARIANCE' ; VAR DPLUSC SEC CLL CUL; RUN ;

PAGE 290

276Appendix I: SAS Code for Calculati ng SAMD Statistic for Hypothesis One TITLE 'COMBINING STANDARDIZED MEAN DIFFERENCES BY STRESS MEASURES AND TIME' ; DATA ONE; INPUT SID 1 2 CONSTRUCT 4 TIME 6 AGE 8 COUNTRY 10 ETIOLOGY 12 DURATION 14 D 16 20 .3 N1 22 24 N2 26 28 QI 30 32 .2 ; V = ((N1+N2)/(N1*N2)+((D** 2 )/( 2 *N1+N2))); W = 1 /V; N = N1 + N2; DN = D*W; CARDS ; 01 1 3 0 3 1 0 0.459 011 029 .85 01 2 3 0 3 1 0 0.080 011 029 .85 03 1 1 2 3 0 1 -.071 017 023 .79 03 1 2 2 3 0 1 0.573 017 023 .79 03 1 3 2 3 0 1 0.806 017 023 .79 03 1 4 2 3 0 1 0.643 017 023 .79 03 1 5 2 3 0 1 0.866 017 023 .79 03 2 1 2 3 0 1 0.190 017 023 .79 06 1 1 1 2 0 2 1.412 047 091 .87 06 2 1 1 2 0 2 1.787 047 091 .87 06 3 1 1 2 0 2 0.949 047 091 .87 08 1 1 2 1 1 0 -.630 082 230 .92 08 2 1 2 1 1 0 -.061 082 230 .92 08 3 1 2 1 1 0 -.117 082 230 .92 09 1 1 2 3 1 1 0.199 039 088 .81 09 3 1 2 3 1 1 0.326 039 088 .81 10 1 3 2 3 1 2 -.616 038 162 .98 10 2 3 2 3 1 2 0.034 038 162 .98 11 1 1 1 1 1 0 0.498 032 024 .77 11 2 1 1 1 1 0 0.579 032 024 .77 12 1 1 2 3 2 0 0.865 009 013 .81 12 1 2 2 3 2 0 0.598 009 013 .81 12 1 5 2 3 2 0 1.528 009 013 .81 13 1 1 2 3 1 2 0.241 016 033 .88 13 1 3 2 3 1 2 0.437 016 033 .88 13 2 1 2 3 1 2 0.629 016 033 .88 14 1 2 1 3 1 2 0.473 010 030 .82 14 1 3 1 3 1 2 0.190 010 030 .82 14 1 4 1 3 1 2 0.000 010 030 .82 14 3 2 1 3 1 2 2.723 010 030 .82 20 1 2 0 2 1 0 -.212 023 090 .93 20 1 3 0 2 1 0 -.177 023 090 .93 20 1 4 0 2 1 0 0.065 023 090 .93 20 1 5 0 2 1 0 0.107 023 090 .93 20 2 1 0 2 1 0 0.048 023 090 .93 20 3 2 0 2 1 0 -.328 023 090 .93 20 3 3 0 2 1 0 -.306 023 090 .93 20 3 4 0 2 1 0 0.067 023 090 .93 20 3 5 0 2 1 0 -.007 023 090 .93 25 3 1 1 3 1 1 0.041 023 075 .91 28 1 1 2 2 1 3 0.024 023 090 .87

PAGE 291

277Appendix I: (Continued) 28 2 1 2 2 1 3 0.050 023 090 .87 28 3 1 2 2 1 3 -.332 023 090 .87 ; PROC SORT; BY TIME CONSTRUCT; DATA TC; SET ONE; IF TIME EQ 1 AND CONSTRUCT EQ 1 ; PROC MEANS MEAN MAXDEC = 2 ; VAR D; PROC MEANS SUM MAXDEC = 2 ; VAR DN; PROC MEANS MEAN SUM N MAXDEC = 2 ; VAR N; +----------------------------------------------+ USING THE RESULTS FROM ABOVE OUTPUT, INPUT THE DATA FOR COMPUTATIONS TO FOLLOW +----------------------------------------------+; DATA TWO; SET TC; MEAND = .32; TOTALD = 15.22; MEAN_N = 107.13 ; TOTALN = 857; K = 8 ; +----------------------------------------------+ COMPUTATIONS EXCLUDING INDIVIDUAL STUDIES +----------------------------------------------+; TOTAL_NW = TOTAL_N N; MEANDW = (TOTALD DN)/TOTALNW; VARI = ( 4 *(N-1 )*(1 +((MEANDW** 2 )/8 )))/(N*(N3 )); VAR_D = ( 4 *(MEAN_N1 )*(1 +((MEAN_DW** 2 )/8 )))/(MEAN_N*(MEAN_N3 )*K); SAMD_I = (D-MEAN_DW)/(SQRT(VAR_I+VAR_D)); SAMD = ABS(SAMDI); SAMDR = ROUND (SAMD, .01); +----------------------------------------------+ RANK ORDERS THE SAMD FOR EACH STUDY +----------------------------------------------+; PROC RANK DESCENDING OUT = TEMP; VAR SAMDR; RANKS RANK_ID; PROC SORT; BY RANKID; DATA THREE; SET TEMP; FILE OUTPUT; PUT RANK_ID SAMDR; PROC PRINT; VAR SID D SAMDR RANKID; PROC MEANS MEAN N STD MIN MAX RANGE; VAR SAMDR; RUN;

PAGE 292

278Appendix J: Scree Plot Identify ing Outliers for Hypothesis One Figure 2. Scree plot for studies reporting the e ffects of acute stress measured at baseline/pre-treatment on ART treatment outcomes 7.77 5.44 1.88 1.73 0.97 0.75 0.27 0.02 0 1 2 3 4 5 6 7 8 9 10 012345678910Rank IDSAMD Figure 3. Scree plot for studies reporting the e ffects of chronic stress measured at baseline/pre-treatment on ART treatment outcomes 9.53 1.91 1.86 1.24 0.39 0.110.1 0 1 2 3 4 5 6 7 8 9 10 012345678910Rank IDSAMD

PAGE 293

279Appendix J: (Continued) Figure 4. Scree plot for studies reporting t he effects of depression measured at baseline/pre-treatment on ART treatment outcomes 5.14 1.84 1.59 1.32 005 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD Figure 5. Scree plot for studies reporting the effects of acute stress measured during the follicular phase on ART treatment outcomes 1.58 1.39 1.27 1.22 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD

PAGE 294

280 Appendix J: (Continued) Figure 6. Scree plot for studies reporting the effects of chronic stress measured during the follicular phase on ART treatment outcomes 7.63 2.26 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD Figure 7. Scree plot for studies reporting t he effects of acute stress measured during oocyte retrieval on ART treatment outcomes 3.91 2.47 1.5 1.42 0.82 0.6 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD

PAGE 295

281Appendix J: (Continued) Figure 8. Scree plot for studies reporting the e ffects of chronic stress measured during oocyte retrieval on ART treatment outcomes 1.6 0.66 0.31 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD Figure 9. Scree plot for studies reporting the effe cts of acute stress measured at embryo transfer on ART treatment outcomes 1.78 0.13 003 0 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD

PAGE 296

282Appendix J: (Continued) Figure 10. Scree plot for studies reporting the effe cts of acute stress measured during the luteal phase on ART treatment outcomes 1.53 0.87 0110 1 2 3 4 5 6 7 8 9 10 012345678910 Rank IDSAMD

PAGE 297

283 Appendix K: SAS Code for HLM Regression Analysis for Hypothesis One TITLE1 'REGRESSION ANALYSIS INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 2 TI 4 9 .3 VI 11 17 .5 T1 19 T2 21 T3 23 T4 25 T5 27 C1 29 C2 31 C3 33 ; WI = 1 / VI; CARDS ; 03 -0.071 0.10236 1 0 0 0 0 1 0 0 06 1.412 0.03949 1 0 0 0 0 1 0 0 08 -0.630 0.01718 1 0 0 0 0 1 0 0 09 0.199 0.03716 1 0 0 0 0 1 0 0 11 0.498 0.07513 1 0 0 0 0 1 0 0 12 0.865 0.20504 1 0 0 0 0 1 0 0 13 0.241 0.09340 1 0 0 0 0 1 0 0 28 0.024 0.05459 1 0 0 0 0 1 0 0 03 0.190 0.10275 1 0 0 0 0 0 1 0 06 1.787 0.04384 1 0 0 0 0 0 1 0 08 -0.061 0.01655 1 0 0 0 0 0 1 0 11 0.579 0.07591 1 0 0 0 0 0 1 0 13 0.629 0.09684 1 0 0 0 0 0 1 0 20 0.048 0.05460 1 0 0 0 0 0 1 0 28 0.050 0.05460 1 0 0 0 0 0 1 0 06 0.949 0.03553 1 0 0 0 0 0 0 1 08 -0.117 0.01657 1 0 0 0 0 0 0 1 09 0.326 0.03742 1 0 0 0 0 0 0 1 25 0.041 0.05682 1 0 0 0 0 0 0 1 28 -0.332 0.05508 1 0 0 0 0 0 0 1 03 0.573 0.10641 0 1 0 0 0 1 0 0 12 0.598 0.19616 0 1 0 0 0 1 0 0 14 0.473 0.13613 0 1 0 0 0 1 0 0 20 -0.212 0.05479 0 1 0 0 0 1 0 0 14 2.723 0.22602 0 1 0 0 0 0 0 1 20 -0.328 0.05507 0 1 0 0 0 0 0 1 01 0.459 0.12803 0 0 1 0 0 1 0 0 03 0.806 0.11042 0 0 1 0 0 1 0 0 10 -0.616 0.03344 0 0 1 0 0 1 0 0 13 0.437 0.09475 0 0 1 0 0 1 0 0 14 0.190 0.13378 0 0 1 0 0 1 0 0 20 -0.177 0.05473 0 0 1 0 0 1 0 0 01 0.080 0.12547 0 0 1 0 0 0 1 0 10 0.034 0.03249 0 0 1 0 0 0 1 0 20 -0.306 0.05500 0 0 1 0 0 0 0 1 03 0.643 0.10747 0 0 0 1 0 1 0 0 14 0.000 0.13333 0 0 0 1 0 1 0 0 20 0.065 0.05461 0 0 0 1 0 1 0 0 20 0.067 0.05461 0 0 0 1 0 0 0 1 03 0.866 0.11168 0 0 0 0 1 1 0 0 12 1.528 0.24110 0 0 0 0 1 1 0 0 20 0.107 0.05464 0 0 0 0 1 1 0 0 20 -0.007 0.05459 0 0 0 0 1 0 0 1;

PAGE 298

284 Appendix K: (Continued) PROC MIXED COVTEST CL ; CLASS SID; MODEL TI = T1 T2 T3 T4 T5 C1 C2 C3 / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 299

285Appendix L: SAS Code for Null Mode l HLM Analysis for Hypothesis One TITLE1 'REGRESSION ANALYSIS INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 2 TI 4 9 .3 VI 11 17 .5 T1 19 T2 21 T3 23 T4 25 T5 27 C1 29 C2 31 C3 33 ; WI = 1 / VI; CARDS ; 03 -0.071 0.10236 1 0 0 0 0 1 0 0 06 1.412 0.03949 1 0 0 0 0 1 0 0 08 -0.630 0.01718 1 0 0 0 0 1 0 0 09 0.199 0.03716 1 0 0 0 0 1 0 0 11 0.498 0.07513 1 0 0 0 0 1 0 0 12 0.865 0.20504 1 0 0 0 0 1 0 0 13 0.241 0.09340 1 0 0 0 0 1 0 0 28 0.024 0.05459 1 0 0 0 0 1 0 0 03 0.190 0.10275 1 0 0 0 0 0 1 0 06 1.787 0.04384 1 0 0 0 0 0 1 0 08 -0.061 0.01655 1 0 0 0 0 0 1 0 11 0.579 0.07591 1 0 0 0 0 0 1 0 13 0.629 0.09684 1 0 0 0 0 0 1 0 20 0.048 0.05460 1 0 0 0 0 0 1 0 28 0.050 0.05460 1 0 0 0 0 0 1 0 06 0.949 0.03553 1 0 0 0 0 0 0 1 08 -0.117 0.01657 1 0 0 0 0 0 0 1 09 0.326 0.03742 1 0 0 0 0 0 0 1 25 0.041 0.05682 1 0 0 0 0 0 0 1 28 -0.332 0.05508 1 0 0 0 0 0 0 1 03 0.573 0.10641 0 1 0 0 0 1 0 0 12 0.598 0.19616 0 1 0 0 0 1 0 0 14 0.473 0.13613 0 1 0 0 0 1 0 0 20 -0.212 0.05479 0 1 0 0 0 1 0 0 14 2.723 0.22602 0 1 0 0 0 0 0 1 20 -0.328 0.05507 0 1 0 0 0 0 0 1 01 0.459 0.12803 0 0 1 0 0 1 0 0 03 0.806 0.11042 0 0 1 0 0 1 0 0 10 -0.616 0.03344 0 0 1 0 0 1 0 0 13 0.437 0.09475 0 0 1 0 0 1 0 0 14 0.190 0.13378 0 0 1 0 0 1 0 0 20 -0.177 0.05473 0 0 1 0 0 1 0 0 01 0.080 0.12547 0 0 1 0 0 0 1 0 10 0.034 0.03249 0 0 1 0 0 0 1 0 20 -0.306 0.05500 0 0 1 0 0 0 0 1 03 0.643 0.10747 0 0 0 1 0 1 0 0 14 0.000 0.13333 0 0 0 1 0 1 0 0 20 0.065 0.05461 0 0 0 1 0 1 0 0 20 0.067 0.05461 0 0 0 1 0 0 0 1 03 0.866 0.11168 0 0 0 0 1 1 0 0 12 1.528 0.24110 0 0 0 0 1 1 0 0 20 0.107 0.05464 0 0 0 0 1 1 0 0 20 -0.007 0.05459 0 0 0 0 1 0 0 1 ; PROC MIXED COVTEST CL ; CLASS SID;

PAGE 300

286Appendix L: (Continued) MODEL TI = / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 301

287Appendix M: SAS Code for Moderato r Analysis for Hypothesis One – Duration of Infertility TITLE1 'REGRESSION MODERATOR ANALYSIS INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 2 TI 4 9 .3 VI 11 17 .5 DURATION 19 21 .1 NA 23 SWA 25 AUSI 27 SEA 29 EURO 31 ; WI = 1 / VI; CARDS ; 03 -0.071 0.10236 4.4 1 0 0 0 0 06 1.412 0.03949 5.5 0 0 0 1 0 08 -0.630 0.01718 0 0 1 0 0 09 0.199 0.03716 3.7 0 0 0 0 1 11 0.498 0.07513 0 0 1 0 0 12 0.865 0.20504 0 0 0 0 1 13 0.241 0.09340 6.3 0 0 0 0 1 28 0.024 0.05459 6.9 0 1 0 0 0 03 0.190 0.10275 6.1 0 0 0 0 1 06 1.787 0.04384 5.5 0 0 0 1 0 08 -0.061 0.01655 0 0 1 0 0 11 0.579 0.07591 0 0 1 0 0 13 0.629 0.09684 6.3 0 0 0 0 1 20 0.048 0.05460 0 1 0 0 0 28 0.050 0.05460 6.9 0 1 0 0 0 06 0.949 0.03553 5.5 0 0 0 1 0 08 -0.117 0.01657 0 0 1 0 0 09 0.326 0.03742 3.7 0 0 0 0 1 25 0.041 0.05682 4.1 0 0 0 0 1 28 -0.332 0.05508 6.9 0 1 0 0 0 03 0.573 0.10641 4.4 1 0 0 0 0 12 0.598 0.19616 0 0 0 0 1 14 0.473 0.13613 6.1 0 0 0 0 1 20 -0.212 0.05479 0 1 0 0 0 14 2.723 0.22602 6.1 0 0 0 0 1 20 -0.328 0.05507 0 1 0 0 0 01 0.459 0.12803 0 0 0 0 1 03 0.806 0.11042 4.4 1 0 0 0 0 10 -0.616 0.03344 6.1 0 0 0 0 1 13 0.437 0.09475 6.3 0 0 0 0 1 14 0.190 0.13378 6.1 0 0 0 0 1 20 -0.177 0.05473 0 1 0 0 0 01 0.080 0.12547 0 0 0 0 1 10 0.034 0.03249 6.1 0 0 0 0 1 20 -0.306 0.05500 0 1 0 0 0 03 0.643 0.10747 4.4 1 0 0 0 0 14 0.000 0.13333 6.1 0 0 0 0 1 20 0.065 0.05461 0 1 0 0 0 20 0.067 0.05461 0 1 0 0 0 03 0.866 0.11168 4.4 1 0 0 0 0 12 1.528 0.24110 0 0 0 0 1 20 0.107 0.05464 0 1 0 0 0 20 -0.007 0.05459 0 1 0 0 0;

PAGE 302

288Appendix M: (Continued) PROC MIXED COVTEST CL ; CLASS SID; MODEL TI = DURATION / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 303

289Appendix N: SAS Code for Modera tor Analysis for Hypothesis One – Country of Study Origin TITLE1 'REGRESSION MODERATOR ANALYSIS INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 2 TI 4 9 .3 VI 11 17 .5 DURATION 19 21 .1 NA 23 SWA 25 AUSI 27 SEA 29 EURO 31 ; WI = 1 / VI; CARDS ; 03 -0.071 0.10236 4.4 1 0 0 0 0 06 1.412 0.03949 5.5 0 0 0 1 0 08 -0.630 0.01718 0 0 1 0 0 09 0.199 0.03716 3.7 0 0 0 0 1 11 0.498 0.07513 0 0 1 0 0 12 0.865 0.20504 0 0 0 0 1 13 0.241 0.09340 6.3 0 0 0 0 1 28 0.024 0.05459 6.9 0 1 0 0 0 03 0.190 0.10275 6.1 0 0 0 0 1 06 1.787 0.04384 5.5 0 0 0 1 0 08 -0.061 0.01655 0 0 1 0 0 11 0.579 0.07591 0 0 1 0 0 13 0.629 0.09684 6.3 0 0 0 0 1 20 0.048 0.05460 0 1 0 0 0 28 0.050 0.05460 6.9 0 1 0 0 0 06 0.949 0.03553 5.5 0 0 0 1 0 08 -0.117 0.01657 0 0 1 0 0 09 0.326 0.03742 3.7 0 0 0 0 1 25 0.041 0.05682 4.1 0 0 0 0 1 28 -0.332 0.05508 6.9 0 1 0 0 0 03 0.573 0.10641 4.4 1 0 0 0 0 12 0.598 0.19616 0 0 0 0 1 14 0.473 0.13613 6.1 0 0 0 0 1 20 -0.212 0.05479 0 1 0 0 0 14 2.723 0.22602 6.1 0 0 0 0 1 20 -0.328 0.05507 0 1 0 0 0 01 0.459 0.12803 0 0 0 0 1 03 0.806 0.11042 4.4 1 0 0 0 0 10 -0.616 0.03344 6.1 0 0 0 0 1 13 0.437 0.09475 6.3 0 0 0 0 1 14 0.190 0.13378 6.1 0 0 0 0 1 20 -0.177 0.05473 0 1 0 0 0 01 0.080 0.12547 0 0 0 0 1 10 0.034 0.03249 6.1 0 0 0 0 1 20 -0.306 0.05500 0 1 0 0 0 03 0.643 0.10747 4.4 1 0 0 0 0 14 0.000 0.13333 6.1 0 0 0 0 1 20 0.065 0.05461 0 1 0 0 0 20 0.067 0.05461 0 1 0 0 0 03 0.866 0.11168 4.4 1 0 0 0 0 12 1.528 0.24110 0 0 0 0 1 20 0.107 0.05464 0 1 0 0 0 20 -0.007 0.05459 0 1 0 0 0;

PAGE 304

290Appendix N: (Continued) PROC MIXED COVTEST CL ; CLASS SID; MODEL TI = NA SWA AUSI SEA EURO / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 305

291Appendix O: SAS Code for Combini ng Effect Sizes for Hypothesis Two TITLE 'COMBINING STANDARDIZED MEAN DIFFERENCES BY CONSTRUCT' ; DATA ORIG; INPUT ID 1 CONSTRUCT 3 D 5 11 N1 13 14 N2 16 17 QI 19 21 .2 ; +------------------------------------+ COMPUTE CI FOR EACH STUDY +------------------------------------+; V=((N1+N2)/(N1*N2))+((D** 2 )/( 2 (N1+N2))); LOWER = D ( 1.96 *SQRT(V)); UPPER = D + ( 1.96 *SQRT(V)); +---------------------------------------------------+ WEIGHT STUDIES BY QUALITY INDEX & SAMPLE SIZE +---------------------------------------------------+; W = 1 /V; QISQ = QI** 2 ; QIW = QI*W; QISQW = QISQ*W; QIWD = QI*W*D; DSQ = D** 2 ; WDSQ = W*DSQ; WD = W*D; WSQ = W** 2 ; N = N1 + N2; CARDS ; 1 1 0.37830 37 45 .92 1 3 1.16280 37 45 .92 2 1 0.31820 24 49 .62 2 1 0.25820 24 49 .62 2 2 -0.2818 24 49 .62 2 2 0.18540 24 49 .62 2 3 0.45640 24 49 .62 2 3 0.20060 24 49 .62 3 1 0.79600 26 19 .91 3 3 0.09640 26 19 .91 4 1 0.59420 30 30 .77 4 3 1.29320 30 30 .76 ; DATA FIRST; SET ORIG; IF CONSTRUCT = 1 ; PROC PRINT DATA = FIRST; TITLE 'SUMMARY STATISTICS FOR D' ; VAR ID D V N CONSTRUCT LOWER UPPER QI W QISQW QIW QIWD WDSQ WD; PROC MEANS SUM N DATA = FIRST; VAR N W WSQ WDSQ WD; +----------------------------------------------+ WRITE OUT SUMMARY STATISTICS FOR VARIANCES +----------------------------------------------+; OUTPUT OUT = COMBINE SUM = Total_N SW SWSQ SWDSQ SWD MEAN = MNN

PAGE 306

292Appendix O: (Continued) N = N1; PROC PRINT; VAR SW SWSQ SWDSQ SWD MNN N1; DATA COMBINE; SET COMBINE; +----------------------------------------------+ COMPUTE UNCONDITIONAL VARIANCE COMPONENT +----------------------------------------------+; C = (SW -((SWSQ/SW))); Q = SWDSQ ((SWD** 2 )/SW); Karen: This computes the probability associated with Q; ProbQ = 1 probchi(Q, N1 1 ); DF = N1 1 ; BTWNV = (Q (N1 1 ))/C; EQ = C*BTWNV + (N1 1 ); PROC PRINT; VAR C Q DF PROBQ BTWNV EQ; +----------------------------------------------+ COMPUTE TOTAL WEIGHTED VARIANCE +----------------------------------------------+; DATA TWO; IF _N_ = 1 THEN SET COMBINE; SET FIRST; WW = 1 /(BTWNV + V); WV1 = QISQ*WW; WV2 = QI*WW; WV2B = WV2** 2 ; IWV = WV1 / WV2B; WQIWD = QI*WW*D; PROC PRINT; TITLE 'SUMMARY STATISTICS FOR WEIGHTED D UNDER RANDOM EFFECTS MODEL' ; VAR ID D V LOWER UPPER IWV QI W WW WV1 WV2 WV2B WQIWD C Q BTWNV EQ; PROC MEANS SUM N ; VAR N WW WV1 WV2 WQIWD; +----------------------------------------------+ WRITE OUT SUMMARY STATISTICS TO COMBINE +----------------------------------------------+; OUTPUT OUT = RESULTS SUM = Total_N SWW SWV1 SWV2 SWQIWD MEAN = MNN N = N1 N2 N3; DATA REFINAL; SET RESULTS; DPLUS = SWQIWD/SWV2; WWV = SWV1/(SWV2** 2 ); VERROR = ((MNN1 )/(MNN3 ))*(4 /MNN)*( 1 +((DPLUS** 2 )/8 )); UNBIASV = WWV VERROR; IF UNBIASV LT 0 THEN UNBIASV = 0 ; A = 1 +(0.75/(MNN3 )); SE = SQRT(WWV); DPLUSC = DPLUS/A; SEC = SE/A;

PAGE 307

293Appendix O: (Continued) CLL = DPLUSC ( 1.96 *SEC); CUL = DPLUSC + ( 1.96 *SEC); PROC PRINT ; VAR SWV1 SWV2 WWV VERROR UNBIASV MNN A SE DPLUS DPLUSC SEC; PROC PRINT ; TITLE 'RANDOM EFFECTS WEIGHTED AVERAGE D AND VARIANCE' ; VAR DPLUSC SEC CLL CUL; RUN ;

PAGE 308

294Appendix P: SAS Code for Computi ng SAMD Statistic for Hypothesis Two DATA ONE; INPUT ID 1 CONSTRUCT 3 D 5 11 N1 1314 N2 1617 QI 1921 .2; V = ((N1 +N2)/(N1*N2)+((D** 2 )/(2 *N1+N2))); W = 1 /V; N = N1 + N2; DN = D*W; CARDS; 1 1 0.37830 37 45 .92 1 3 1.16280 37 45 .92 2 1 0.31820 24 49 .62 2 1 0.25820 24 49 .62 2 2 -0.2818 24 49 .62 2 2 0.18540 24 49 .62 2 3 0.45640 24 49 .62 2 3 0.20060 24 49 .62 3 1 0.79600 26 19 .91 3 3 0.09640 26 19 .91 4 1 0.59420 30 30 .77 4 3 1.29320 30 30 .76 ; PROC SORT; BY CONSTRUCT; DATA TC; SET ONE; IF CONSTRUCT EQ 3 ; PROC MEANS MEAN MAXDEC = 2 ; VAR D; PROC MEANS SUM MAXDEC = 2 ; VAR DN; PROC MEANS MEAN SUM N MAXDEC = 2 ; VAR N; DATA TWO; SET TC; MEAND = .64; TOTALD = 45.73; MEAN_N = 66.6; TOTALN = 333; K = 5 ; +----------------------------------------------+ COMPUTATIONS EXCLUDING INDIVIDUAL STUDIES +----------------------------------------------+; TOTALNW = TOTALN N; MEANDW = (TOTALD DN)/TOTALNW; VARI = ( 4 *(N-1 )*(1 +((MEANDW** 2 )/8 )))/(N*(N3 )); VAR_D = ( 4 *(MEAN_N1 )*(1 +((MEAN_DW** 2 )/8 )))/(MEAN_N*(MEAN_N3 )*K); SAMD_I = (D-MEAN_DW)/(SQRT(VAR_I+VAR_D)); SAMD = ABS(SAMDI); SAMDR = ROUND (SAMD, .01);

PAGE 309

295Appendix P: (Continued) +----------------------------------------------+ RANK ORDERS THE SAMD FOR EACH STUDY +----------------------------------------------+; PROC RANK DESCENDING OUT = TEMP; VAR SAMDR; RANKS RANK_ID; PROC SORT; BY RANKID; DATA THREE; SET TEMP; FILE OUTPUT; PUT RANKID SAMDR; PROC PRINT; VAR ID D SAMDR RANKID; PROC MEANS MEAN N STD MIN MAX RANGE; VAR SAMDR; RUN ;

PAGE 310

296Appendix Q: Scree Plot Identify ing Outliers for Hypothesis Two Figure 13. Scree plot for studies reporting the effe cts of psychoeducational interventions on acute stress experienced duri ng ART treatment regimens 2.18 1.77 1.11 0.8 0.56 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0123456Rank ID Figure 14. Scree plot for studies reporting the effe cts of psychoeducational interventions on chronic stress experienced durin g ART treatment regimens 1.11 0.85 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0123456 Rank IDSAMD

PAGE 311

297Appendix Q: (Continued) Figure 15. Scree plot for studies reporting the effe cts of psychoeducational interventions on depression experienced during ART treatment regimens 4.23 4.14 1.17 0.18 0.140 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 0123456Rank IDSAMD

PAGE 312

298Appendix R: SAS Code for HLM Regression Analysis for Hypothesis Two TITLE1 'HYPOTHESIS TWO REGRESSION ANALYSIS INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 TI 3 9 .3 VI 11 18 .5 C1 20 C2 22 C3 24 ; WI = 1 /VI; CARDS ; 1 0.37830 0.050122 1 0 0 2 0.31820 0.062768 1 0 0 2 0.25820 0.062531 1 0 0 2 -0.2818 0.062619 0 1 0 2 0.18540 0.062310 0 1 0 2 0.45640 0.063502 0 0 1 2 0.20060 0.062350 0 0 1 3 0.79600 0.098133 1 0 0 3 0.09640 0.091196 0 0 1 4 0.59420 0.069609 1 0 0 1 1.16280 0.057494 0 0 1 4 1.29320 0.080603 0 0 1 ; PROC MIXED COVTEST CL ; CLASS SID; MODEL TI = C1 C2 C3 / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 313

299Appendix S: SAS Code for Null Model HLM Regression Analysis for Hypothesis Two TITLE1 'HYPOTHESIS TWO REGRESSION ANALYSIS NULL MODEL INCLUDING OUTLIERS' ; DATA ONE; INPUT SID 1 TI 3 9 .3 VI 11 18 .5 C1 20 C2 22 C3 24 ; WI = 1 /VI; CARDS ; 1 0.37830 0.050122 1 0 0 2 0.31820 0.062768 1 0 0 2 0.25820 0.062531 1 0 0 2 -0.2818 0.062619 0 1 0 2 0.18540 0.062310 0 1 0 2 0.45640 0.063502 0 0 1 2 0.20060 0.062350 0 0 1 3 0.79600 0.098133 1 0 0 3 0.09640 0.091196 0 0 1 4 0.59420 0.069609 1 0 0 1 1.16280 0.057494 0 0 1 4 1.29320 0.080603 0 0 1 ; PROC MIXED COVTEST CL ; CLASS SID; MODEL TI = / SOLUTION DDFM = BW NOTEST CL ; WEIGHT WI; RANDOM INTERCEPT / SUB = SID; RUN ;

PAGE 314

ABOUT THE AUTHOR Karen Rose Mumford has spent her pr ofessional career as an educator. Beginning in 1992, she began her career as a teacher, working with children with severe emotional disturbances and later wit h children with autism. As a teacher, Karen is known for her unique teaching strategies and techniques. Recognized as an educational leader, Karen acc epted an administrative position as Coordinator of Evaluation in 2000. In this position, Karen provides data and information related to program effe ctiveness, improving the educational opportunities and achievem ents of students. Karen received her bachelor’s degr ee in special education and her master’s degree in Educat ional Leadership. After marrying, Karen received her doctoral degree from the University of S outh Florida in Educational Measurement and Research. Along with her co-authors, she was awarded the “Distinguished Paper of the Florida Educat ional Research Association” in 2001. Karen plans to continue conducting research in health rela ted issues and teach at the university level.