USF Libraries
USF Digital Collections

The association of maternal pregnancy complications and sudden infant death syndrome

MISSING IMAGE

Material Information

Title:
The association of maternal pregnancy complications and sudden infant death syndrome
Physical Description:
Book
Language:
English
Creator:
Myers, Patricia D
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Pregnancy Complications   ( mesh )
Sudden Infant Death   ( mesh )
fetal hypoxia
sids
maternal risk factors
prenatal outcomes
infant
Dissertations, Academic -- Public Health -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Sudden Infant Death Syndrome (SIDS) is the third leading cause of infant mortality between birth and the first year of life in the United States. Along with the identification of various maternal risk factors, the role of fetal hypoxia has been hypothesized to be one of many causal factors associated with SIDS. The purpose of this study was to develop a profile of the SIDS infant and assess whether six pregnancy complications consistent with fetal hypoxia were associated with the increased outcome of SIDS. The secondary data analysis of Florida linked birth to death certificate data specific to Hillsborough County and Duval County were analyzed retrospectively for the period of time between 1998 and 2000. Of the 86, 342 births, 69 SIDS cases were identified, 34 in Hillsborough County and 35 in Duval County.A majority of the infants were White males with an average age of death of 80 days. The Chi-Square test for Independence with Cramer's V, odds ratios and 95% confidence intervals were calculated to determine if an association existed between pregnancy complications, specific maternal risk factors and SIDS. Eclampsia was the only statistically significant prenatal complication found in this cohort (OR=4.67: 95% CI 1.49, 14.57). Maternal tobacco use (OR= 3.13: 95% CI 1.83, 5.36) and late initiation into prenatal care were also found to be significant in the SIDS cases, with the greatest risk occuring in women who did not receive prenatal care (OR=4.37: 95% CI 1.38, 13.89). These findings will assist with the development of a profile of infants who are at greater risk of dying of SIDS in Hillsborough County and Duval County as well as contribute to what is currently known about the association between fetal hypoxia and SIDS.
Thesis:
Thesis (M.S.P.H.)--University of South Florida, 2003.
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 Patricia D. Myers.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 62 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 - 001416930
oclc - 52833044
notis - AJJ4782
usfldc doi - E14-SFE0000068
usfldc handle - e14.68
System ID:
SFS0024764:00001


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 001416930
003 fts
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 030729s2003 flu sbm s000|0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000068
035
(OCoLC)52833044
9
AJJ4782
b SE
SFE0000068
040
FHM
c FHM
049
FHME
090
RA425
096
WQ 240
1 100
Myers, Patricia D.
4 245
The association of maternal pregnancy complications and sudden infant death syndrome
h [electronic resource] /
by Patricia D. Myers.
260
[Tampa, Fla.] :
University of South Florida,
2003.
502
Thesis (M.S.P.H.)--University of South Florida, 2003.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 62 pages.
520
ABSTRACT: Sudden Infant Death Syndrome (SIDS) is the third leading cause of infant mortality between birth and the first year of life in the United States. Along with the identification of various maternal risk factors, the role of fetal hypoxia has been hypothesized to be one of many causal factors associated with SIDS. The purpose of this study was to develop a profile of the SIDS infant and assess whether six pregnancy complications consistent with fetal hypoxia were associated with the increased outcome of SIDS. The secondary data analysis of Florida linked birth to death certificate data specific to Hillsborough County and Duval County were analyzed retrospectively for the period of time between 1998 and 2000. Of the 86, 342 births, 69 SIDS cases were identified, 34 in Hillsborough County and 35 in Duval County.A majority of the infants were White males with an average age of death of 80 days. The Chi-Square test for Independence with Cramer's V, odds ratios and 95% confidence intervals were calculated to determine if an association existed between pregnancy complications, specific maternal risk factors and SIDS. Eclampsia was the only statistically significant prenatal complication found in this cohort (OR=4.67: 95% CI 1.49, 14.57). Maternal tobacco use (OR= 3.13: 95% CI 1.83, 5.36) and late initiation into prenatal care were also found to be significant in the SIDS cases, with the greatest risk occuring in women who did not receive prenatal care (OR=4.37: 95% CI 1.38, 13.89). These findings will assist with the development of a profile of infants who are at greater risk of dying of SIDS in Hillsborough County and Duval County as well as contribute to what is currently known about the association between fetal hypoxia and SIDS.
590
Adviser: Perrin, Karen M.
653
fetal hypoxia.
sids.
maternal risk factors.
prenatal outcomes.
infant.
0 690
Dissertations, Academic
z USF
x Public Health
Masters.
2 650
Pregnancy Complications.
Sudden Infant Death.
773
t USF Electronic Theses and Dissertations.
856
u http://digital.lib.usf.edu/?e14.68



PAGE 1

THE ASSOCIATION OF MATERNAL PREGNANCY COMPLICATIONS AND SUDDEN INFANT DEATH SYNDROME by PATRICIA D. MYERS A thesis submitted in partial fullfillment of the requirements for the degree of Master of Science in Public Health Department of Family and Community Health College of Public Health University of South Florida Major Professor: Karen M. Perrin, Ph.D. Elizabeth Gulitz, Ph.D. Jeffery Kromrey, Ph.D. Date of Approval: May 23, 2003 Keywords: SIDS, maternal risk factors, fetal hypoxia, prenatal outcomes, infant Copyright 2003, Patricia D. Myers

PAGE 2

Acknowledgments The success in completing this document could not have been accomplished without the support of a number of indivi duals of whom I owe gratitude and recognition. First, I would like to thank my advisor and the major professo r of my thesis committee, Dr. Kay Perrin for recognizing my potential and acknowledging my need to be challenged and to Dr. Betty Gulitz and Dr. Jeff Kromery for taking the time from their busy schedules to meet with me on numerous occasions to organize my proposal and interpret my findings. I am truly indebted to Leisa Stanley of the Healthy Start Coalition of Hillsborough County for allowing me the opportunity to work unde r her supervision as an intern this year. I value her patience and ongoing guida nce through the beginn ing stages of my thesis. A special thanks to John Harris with the Healthy Start Co alition of Hillsbourgh County and Dr. David Darr for their knowledge and tolerance in assisting me with the challenges I faced with SAS and SPSS. I am appreciative for the assistance of the Florida Department of Health, specifically Karen Freeman and Dan Thompson, as well as Ca rol Brady and Dawn Clarke of the Northeast Florida Healthy Start Coaltion for helping me to secure the Duval County data for this research project. Finally, my deepest gratitude goes to Dr. Mark Scholl for always taking the time even when there was not a lot of time to give, for st anding beside me, giving me feedback and being my rationale side and to Bethany Bell for shar ing her own graduate school experiences with me as well as provide her ongoing s upport throughout the process. I would not have completed it without you both. I am also grateful for my frie nds and family in Atlanta for respecting my choice to complete my MSPH and pr oviding encouragement along the way.

PAGE 3

i Table of Contents List of Tables iii Abstract iv Chapter One. Introduction 1 Statement of the Problem 3 Purpose of the Study 4 Definition of Terminology 5 Abruptio Placentae 5 Anemia 6 Eclampsia 6 Fetal Hypoxia 6 Placenta Previa 7 Pregnancy-Induced Hypertension (PIH) 7 Uterine Bleeding 7 Hypotheses 7 Hypothesis I: Anemia 7 Hypothesis II: Pregnancy-Induced Hypertension 8 Hypothesis III: Eclampsia 8 Hypothesis IV: Uterine Bleeding 8 Hypothesis V: Placenta Previa 8 Hypothesis VI: Abruptio Placentae 8 Hypothesis VII: Late Initiati on into Prenatal Care 8 Hypothesis VIII: Smoking 8 Hypothesis IX: Smoking as a Confounder 8 Chapter Two. Review of Literature 9 Current Theory of the Etiology of SIDS 9 Role of Hypoxia in the Outcome of SIDS 10 Uterine Bleeding/Placenta Previa/Abr uptio Placentae 11 Preeclampsia/Eclampsia 12 Risk Factors Related to Pregnancy Complications 13 Maternal Tobacco Use during Pregnancy 13 Anemia 15 Late Initiation of Prenatal Care 15 Contributions to Current Literature 17 Chapter Three. Methods 18 Subjects 19 Measures 20

PAGE 4

ii Data Collection 21 Data Analysis 21 Protection of Huma n Subjects 22 Chapter Four. Results 23 Descriptive Statistics of Sample 23 Analyses of Hypotheses I-VI 25 Analysis of Hypothesis VII 26 Analysis of Hypothesis VIII 27 Analysis of Hypothesis IX 27 Chapter Five. Discussion 28 The Profile of the SIDS Infant 28 Summary of Maternal Pregnancy Comp lications 30 Hypotheses of Association between Pregnanc y Complications and SIDS 32 Hypotheses of Maternal Tobacco Use 33 Hypothesis of Late Initiation into Pr enatal Care 35 Limitations of Study 35 Implications 36 Recommendations for Future Study 37 References 39 Bibliography 44 Appendices 52 Appendix A: Modified De scription of Data Fields 53 Appendix B: University of South Florida IRB Exemption Certification 55

PAGE 5

iii List of Tables Table 1 2000 Demographic Profile by County 45 Table 2 2000 Vital Statistics of Resident Live Births and Deaths by County 45 Table 3 Frequency of Pregnancy Complicatio n by County and Year 46 Table 4 Distribution of Infants w ho died of SIDS by County between 1998-2000 47 Table 5 Maternal anemia during pregnancy a nd the risk of SIDS 48 Table 6 Maternal hypertension asso ciated with pregnancy and the risk of SIDS 48 Table 7 Maternal eclampsia during pregnancy and the risk of SIDS 48 Table 8 Maternal uterine bleedin g during pregnancy and the risk of SIDS 49 Table 9 Maternal placenta prev ia during pregnancy and the risk of SIDS 49 Table 10 Maternal abruptio pl acentae during pregnancy and the ri sk of SIDS 49 Table 11 Maternal tobacco use during pregnanc y and the risk of SIDS 50 Table 12 Late initiation into prenatal care and the risk of SIDS 50 Table 13 Lack of prenatal care during pregna ncy and the risk of SIDS 50 Table 14 Pregnancy complicatons and th e risk of SIDS while controlling for smoking 51

PAGE 6

iv The Association of Maternal Pregnancy Complications and Sudden Infant Death Syndrome Patricia D. Myers ABSTRACT Sudden Infant Death Syndrome (SIDS) is the third leading cause of infant mortality between birth and the first year of life in the United States. Along with the identication of various maternal risk fact ors, the role of fe tal hypoxia has been hypothesized to be one of many causal factors a ssociated with SIDS. The purpose of this study was to develop a profile of the SIDS infant and assess whether six pregnancy complications consistent with fetal hypoxia were associated with the increased outcome of SIDS. The secondary data analysis of Florida linked bi rth to death certificate data specific to Hillsborough County and Duval Count y were analyzed retrospectively for the period of time between 1998 and 2000. Of the 86, 342 births, 69 SIDS cases were identified, 34 in Hillsborough County and 35 in Duval County. A majority of the infants were White males with an average age of d eath of 80 days. The Chi-Square test for Independence with Cramer’s V, odds ratios and 95% confidence inte rvals were calculated to determine if an association existed between pregnancy complications, specific maternal risk factors and SIDS. Eclampsia wa s the only statistically significant prenatal complication found in this cohort (OR=4.67: 95% CI 1.49, 14.57). Maternal tobacco use (OR= 3.13: 95% CI 1.83, 5.36) and late initiatio n into prenatal care were also found to

PAGE 7

v be significant in the SIDS cases, with the gr eatest risk occuring in women who did not receive prenatal care (OR=4.37: 95% CI 1.38, 13.89). These findings will assist with the development of a profile of infants who are at greater risk of dying of SIDS in Hillsborough County and Duval County as well as contribute to what is currently known about the association between fetal hypoxia and SIDS.

PAGE 8

1 Chapter One Introduction Sudden Infant Death Syndrome (SIDS) is one of the leading causes of death of infants between one and 12 months of age (Na tional Institute of Child Health and Human Development [NICHD], 2000), accounting for appr oximately 30% of all deaths in this age group (Beers & Berkow, 1999). While the exact cause of SIDS is unknown, organizations such as the American SIDS Institute (2002) have supported collaborative multidisciplinary research efforts with the aim of identifying potential risk factors associated with SIDS. In addition, the SI DS Alliance (2001) has partnered with the National Institute of Child Health and Human Development (NICHD) to develop research and education programs targeted at determining the cause of SIDS deaths as well as providing a national support center for SIDS families. The findings of clinical and epidemiological research have resulted in a decrease of approximately 2000 deaths per year, yet SIDS continues to claim the lives of almost 3000 infant s in the United States each year (National Center for H ealth Statistics [NCHS], 2001). In 1974, federal legislation passed the Sudden Infant Death Syndrome Act (PL 93-270) appointing the NICHD with the stat utory responsibility of conducting SIDS research (Beggs, Bucks, Corwin, Dailey, Eagl estaff, Fifer & Jacobson, 2001). Since that time, reports have been produced revealing immense progress made in the understanding of SIDS. In 1999, United States Congress reque sted the development of a 5-year plan

PAGE 9

2 with an objective to reduce the number of deaths caused by SIDS. As a result, the present strategic proposal was formulated w ith the following resear ch initiatives: 1. The continuation of research on the cau se(s) of SIDS, the development of abnormalities that increase vulnerability and result in high-risk infants, identification of genetic markers that may predict SIDS and determine whether SIDS is part of a larger family of nervous system disorders; 2. The analysis of epidemiological a nd physiological data to improve our understanding of environmental and other risk factors; 3. Community-based studies addressing health disparities resulting in fetal, infant and early childhood deaths; and 4. Improving risk reduction th rough the "Back to Sleep" campaign with continued research, monitoring, and outreach in at -risk communities (Sudden Infant Death Syndrome Alliance, 2001). Recommendations made by the American Academy of Pediatrics (AAP) in 1992 are credited for the initial reduction of SIDS cases nationwide. In response to these guidelines, the NICHD, in partnership with th e U.S. Public Health Service, the AAP, the SIDS Alliance, and the Association of SIDS and Infant Mortality Programs implemented the nationwide ”Back to Sleep” campaign to increase awareness of the importance of infants being placed in the supine position wh en asleep. Prior to the launch of this campaign, it had been declared that infants who slept in the prone position were at greater risk for SIDS. As the sleep position change d through public awareness, the incidence of SIDS declined by 38% between 1992 and 1998 (Sudden Infant Death Syndrome (SIDS) Alliance, 2001). Nevertheless, the inciden ce of new cases continue despite the steady decline from 70% in 1992 to only 21% in 1997 of babies sleeping on their stomachs (United States Department of Health and Human Services [DHHS], 1999). Research

PAGE 10

3 investigating other causes beyond environmenta l stressors is necessary for the further reduction of cases. Sudden Infant Death Syndrome (SIDS) is defined as “the sudden and unexpected death of an infant under the age of one year that remains unexplai ned despite a thorough investigation, which includes an autopsy, examination of the death scene, and review of the clinical history” (Willinger, James &, Catz, 1991). The current statute 406.11 of the state of Florida requires th e Florida Medical Examiner s Commission to have an established protocol for the i nvestigation of sudden and unexpla ined deaths of infants in order to make a final diagnosis of Sudde n Infant Death Syndrome. The Florida Administrative Code 11G-2.0031: SIDS Autopsy Protocol (1996) states that the medical examiner must inspect and document the en vironment where the infant was discovered, and complete a clinical history review of prenatal, delivery, and postnatal medical information. The specific details of the mandatory autopsy include a thorough examination of the status of the internal and external features, a complete skeletal survey and photographs of the body, and th e collection of histological slides, bacterial and viral cultures and fluids and tissues for toxicology te sts. The diagnosis of SIDS is made only after the death is classified as natural and c ongenital, infectious, environmental, or other unnatural causes of death in cluding neglect or abuse are excluded (SIDS Autopsy Protocol of 1996). Statement of Problem Although the number of SIDS deaths in the state of Florida has gradually decreased, from 145 cases in 1996 to 96 cases in 2001, since the SIDS Autopsy Protocol went into effect, 710 infants have died of SIDS during this period of time (Florida

PAGE 11

4 Department of Health, 1996-2001). The risk factor s associated with SIDS are likely to be multi-factorial, involving a combination of anatomical, physiological, environmental, and social factors (Hoffman & Hillman, 1992). In recent years, the relationship between prenatal complications and the risk factors associated with SIDS have not been well studied despite the contention th at it is a greater challenge to assess postneonatal risk factors that it is to assess maternal ri sk factors (Hoffman & Hillman, 1992). The epidemiological research has focused on alte ring environmental risks in the postneonatal period by changing maternal behavior, instead of defining the underlyi ng risk factors that occur before birth. While environmental fact ors, such as maternal tobacco use during pregnancy, are known to lead to poor health outcomes of the infant, the correlation to such consequences as intrauterine fe tal hypoxia and anatomic and physiological abnormalities deserve greater attention from th e epidemiological perspective since they have been associated with both adverse outcomes in pregnancy as well as SIDS. Purpose of Study The purpose of this study is to develop a profile of the SIDS infant as well as explore the relationship between prenatal complications asso ciated with fetal hypoxia and the outcome of SIDS specific to the p opulations of Hillsborough County, Florida and Duval County, Florida between 1998 and 2000. Since the “Back to Sleep” Campaign, a majority of the research examining the asso ciation between prenat al factors and SIDS have consisted of retrospectiv e studies, with the attention shifted towards the general hypothesis that an abnormality exists in infa nts who die of SIDS that originates during fetal development. Initially, this study will de scribe the demographics of the SIDS infant in addition to disclosing the association with the maternal risk factors of smoking and late

PAGE 12

5 initiation into prenatal care Next, the events during pr egnancy known to cause fetal hypoxia will be analyzed to clarify what, if a ny, association they ha ve with the outcome of SIDS in these populations. The research linking pregnancy complications with SIDS concentrates on how the adverse events affect fetal growth and devel opment that result in a vulnerable infant. Since only a few publica tions have addressed the consequences of pregnancy-related events in the last decade, further research is n eeded to predict which babies are most vulnerable so that strategies can be developed to prevent SIDS deaths (Sudden Infant Death Syndrome Alliance, 2001). It is hypothesized that ad verse events that occur du ring pregnancy influence the health of the infant that may result in the outcome of SIDS. The likelihood of some sort of adverse event during the prenatal period is frequent enough to at least address the possibility that there could be a positive correlation to SIDS. Factors that warrant investigation include an unde rstanding of the mother’s underlying diagnoses during pregnancy and how it compromises the normal growth and development of the fetus. Therefore, further attention should be focuse d on what takes place during this critical period of time that increases the potential for such a serious outcome. Definition of Terminology This research study discusses a vast am ount of terminology related to pregnancy complications that may not be familiar to those outside the medical field, therefore, definitions of the variable terms are provided in this section. Abruptio Placentae. (Beers & Berkow, 1999) The pr emature separation of a normally implanted placenta from the uterus due to unknown etiology. Abruptio placentae develops in 0.4 to 3.5% of all de liveries. It is associated with various

PAGE 13

6 hypertensive, cardiovascular, and rheumatoid di seases and especially with use of cocaine in any form. Symptoms include hemorrhage, pain that becomes constant, albuminuria, anemia and contractions as the pl acenta tears from the uterine wall. Anemia. (Beers & Berkow, 1999) Anemia dur ing pregnancy is defined as hemoglobin (Hb) level of < 10 g/dL. However, any patient with an Hb level < 11 to 11.5 g/dL at onset of pregnancy must be treated as anemic, because the hemodilution that occurs during pregnancy reduces the Hb level to the anemic range. Anemia occurs in as many as 80% of gravid populati ons. Iron deficiency is responsible for 95% of cases of anemia during pregnancy. The deficiency is usually due to inadequate dietary intake (especially in teenage girls), to a previous pregnancy, or to the normal loss of iron in blood with menses (which approximates the amount normally ingested each month, so iron stores are never built up). Eclampsia. (Thomas, 1993) The onset of seizures and a coma between the 20th week of pregnancy and the end of the first week postpartum. It develops in 1 of 200 patients with preeclampsia, or elevated bl ood pressure during pregnancy, and is usually fatal if untreated. Fetal Hypoxia. (Washington, 2003) Deficiency of oxygen exchange between the mother and fetus that may be the result of a number of maternal ri sk factors including maternal infection, pregnancy-induced or ch ronic hypertension, Rh se nsitization, chronic substance abuse, asthma, seizures. Feta l hypoxia is associated with long term complications, such as cerebr al palsy, meconium aspirati on syndrome, seizures and encephalopathy if left untreated.

PAGE 14

7 Placenta previa. (Beers & Berkow, 1999) Implantation of the placenta over or near the internal os of the cervix. The pl acenta may cover the internal os completely (total previa) or partially (partial previa), or it may encroach on the internal os (low-lying placenta or marginal previa). Placenta previa occurs in 1 of 200 deliveries, usually in multiparas (women who have had more than one pregnancy), in patients who have had a cesarean section, or in patient s with uterine abnormalities (e.g., fibroids) that inhibit normal implantation. Pregnancy-Induced Hypertension (PIH). (Beers & Berkow, 1999) The development of hypertension, with a increase in systolic blood pr essure of 30 mm Hg and/or diastolic of 15 mm Hg over baseline, proteinuria and edema during pregnancy. The two categories of PIH consist of preec lampsia and eclampsia, with preeclampsia being the progressive disease that leads to the more severe state of eclampsia. The result of hypertension is the reduction of blood flow to the placenta, which can then lead to fetal hypoxia and intrauterine growth retardation. Uterine Bleeding. (Beers & Berkow, 1999) Bleeding from the uterus. Complications of pregnancy are the most common organic causes of abnormal bleeding in women of reproductive age and nearly ha lf of patients with uterine bleeding and symptoms of pregnancy or a confirmed early pregnancy spontaneous ly abort the fetus. Hypotheses Hypothesis I: Anemia. Women with anemia during pregnancy are more likely to have an infant die of SIDS when compared to women without anemia during pregnancy.

PAGE 15

8 Hypothesis II: Pregnancy -Induced Hypertension. Women with pregnancyinduced hypertension during pregnancy are more likely to have an infant die of SIDS when compared to women without pregna ncy-induced hypertension during pregnancy. Hypothesis III: Eclampsia. Women with eclampsia dur ing pregnancy are more likely to have an infant die of SIDS when compared to women without eclampsia during pregnancy. Hypothesis IV: Uterine Bleeding. Women with uterine bleeding during pregnancy are more likely to have an infa nt die of SIDS when compared to women without uterine bleeding during pregnancy. Hypothesis V: Placenta Previa. Women with placenta previa during pregnancy are more likely to have an infant die of SI DS when compared to women without placenta previa during pregnancy. Hypothesis VI: Abruptio Placentae. Women with abruptio placentae during pregnancy are more likely to have an infa nt die of SIDS when compared to women without abruptio placentae during pregnancy. Hypothesis VII: Late Initiation into Prenatal Care. Women who initiate prenatal care late or not at all during pregnancy are more likely to have an infant die of SIDS when compared to women who do initiate prenatal care early in the pregnancy. Hypothesis VIII: Smoking. Women who smoke during pregnancy have a greater risk of having an infant die of SIDS than a woman who does not smoke during pregnancy. Hypothesis IX: Smoking as a Confounder. Smoking during pregnancy is a confounder between any of the pregnancy complications and the outcome of SIDS.

PAGE 16

9 Chapter Two Review of Literature The preventive strategies created by Be ggs et al. (2001) es tablish the foundation of the following review of literature. According to this strategic plan: “The more we know about how risk fact ors contribute to SIDS pathogenesis, the more confident we can be that an infant wi ll not succumb to SIDS…The early months are a time of rapid growth and developmen t, with changing nut ritional and sleep requirements. To prepare for these change s, the baby needs a healthy start during the prenatal period. What are the requirements for normal development? What factors increase the likelihood th at the developmental process will go awry? How can we reduce that risk? Our knowledge of human development in fetal lif e and early infancy is quite limited. Although research on SIDS risk fact ors has identified pren atal and postnatal environmental characteristics and care practic es that are critical to healthy outcomes, more research is needed” (p. 25). Current Theory of the Etiology of SIDS The premise directing innovative SIDS resear ch today is based on the Triple Risk Model of SIDS, which implies that causality may be multi-factorial. According to this hypothesis, “SIDS results when the following th ree events take place simultaneously: 1) an underlying vulnerability in the infant; 2) a critical development period in state-related homeostatic control; and 3) an exogenous stressor(s) that exacerbate the infant’s underlying vulnerability” (Filiano & Kinney, 1994, p. 195). It is suggested by Filiano and Kinney (1994) that the vulnerability of ba bies who die of SIDS, which occurs in a critical period of development, is a re sult of an adverse condition during fetal development that prevents the normal respons e to environmental stressors. An objective

PAGE 17

10 of the SIDS research sponsored by the NICHD is to clarify the role of fetal development in the etiology of SIDS by investigating the e ffect prenatal factor s have on the infant. A multidisciplinary approach to the triple risk hypothesis has been developed to determine what is unique in the cases of SI DS when compared to a healthy infant that causes the underlying vu lnerability. Primarily, epidemio logical studies support that the concept of the vulnerable infant is substantiate d by the risk factors th at develop as a result of a suboptimal fetal environment, incl uding maternal smoking, maternal anemia, infection and drug use during pregnancy (F iliano & Kinney, 1994). Neuropathological research suggests that the vulnerability may result from an abnormality in brain development, originating during gest ation, which causes a dysfunction in cardioventilation and arousal (Kinney et al., 1992). In a prospective study, Naeye & Blanc (1976) found that the like lihood of brain stem dysfunc tion is further evidenced by future SIDS victims that experience resp iratory, feeding, and temperature regulation problems, and have an abnormal response to tactile stimulation. Role of Hypoxia in the Outcome of SIDS The concept of the underlying vulnerability of the infant is addressed in findings that lend support to the role of fetal hypoxia in SIDS (Buc k et al., 1989). The support of the hypoxia hypothesis is validated by post-mortem studies of infants who died of SIDS that reported abnormal tissue findings consis tent with chronic hypoxia (Naeye, 1980). In utero hypoxia is thought to have a permanent affect on the cen tral nervous system of the fetus, particularly the brain stem which c ontrols cardiorespiratory function (Filiano & Kinney, 1995).

PAGE 18

11 Uterine Bleeding/Placental Previa/Abruptio Placentae The assessment of antenatal factors known to result in fetal hypoxia was collected from birth certificates and compared between the death certificates of SIDS infants and two sets of controls to dete rmine if variation existed (Buc k et al., 1989). The choice of ten antenatal variables was based on condi tions that caused placental dysfunction and jeopardized a favorable uterine environment fo r the fetus. When compared to controls who died of other causes, Buck et al. (1989) found uterine bleeding not associated with placental complications was a significant risk factor of SIDS infants (OR=5.44, 95%=1.39, 21.35). The live control comparison observed a number of significant risk factors: placenta previa (OR= 6.84), ecl ampsia (OR= 5.52), and abruptio placentae (OR=4.58). These findings were consistent with prio r epidemiological research that account for an increased frequency of SIDS when associated with a history of pregnancy complications. Specifically, the pregnancy re lated events related to placenta previa, placenta abruption, and abnormal uterine bleedi ng (Standfast, Jereb &, Janerich, 1980). In the comparison of infants who died of SI DS to infants who died of other causes, Standfast et al. (1980) repor ted that 58% of the SIDS mothers had abnormal uterine bleeding when compared to only 17% in the cohort mothers. Furthermore, a similar association was found in the repo rt of placenta previa and ab ruptio placentae in mothers whose infants later died of SIDS. Events that occur during pregnancy that compromise the fetal environment can have a permanent effect on the infant. The adverse conditions involving the placenta, defined as placenta previa and placenta abrup tion are examples of such an event. Since

PAGE 19

12 the placenta provides the major source of blood, oxygen, and nourishment to the fetus, any anomalies can lead to detrimental outcomes in the fetus (Beers & Berkow, 2000). The decrease in surface area of the placenta reduces oxygen and blood exchange between the maternal and fetal circulatory system and deficiency in oxygen transport to the fetus is believed to impinge on the cardiac, resp iratory and nervous sy stems involved in the etiology of SIDS (Li & Wi, 1999). In a population-based case-control study using California linked birth and d eath certificate data, approximately 2100 SIDS cases were matched to controls by birth year to determine whether placental abnormalities were associated with SIDS. After potential conf ounders were controlled for using logistic regression, an association was found between these placental anomalies and a twofold increase in the risk of SIDS (OR=2.1 95% CI 1.3-3.1). Preeclampsia/Eclampsia The maternal exposure to preeclampsia/ eclampsia during pregnancy is another known complication that leads to similar results of fetal hypoxia that may make infants more susceptible to SIDS. Preeclampsia is a result of pregnancy induced hypertension that can lead to a more severe form of ecl ampsia that in time can reduce the blood flow and oxygen supply from the mother to the pla centa. A strong association exists between preeclampsia and fetal growth retardation as a result of the lack of nutrition to the fetus (Voto et al., 1999). In addition, preeclampsia leads to a reduction in placental size and can cause abruptio placentae, or separation from the uterine wall, which can lead to fetal hypoxia. In a comparison of SIDS infants to matc hed live controls, Li & Wi conducted a study to ascertain whether preeclampsia/eclamps ia during pregnancy increased the risk of

PAGE 20

13 SIDS in infants. The analysis of linked bi rth and death certificate data revealed 49 cases of SIDS whose mother expe rienced pre-eclampsia/eclamp sia during pregnancy. The results showed an association with a 50% increased risk of SIDS in infants when compared to non-SIDS controls (OR=1.5, 95% CI 1.1, 2.0) after controlling for potential confounders (Li & Wi, 2000). Although consistenc y of these findings ar e not statistically significant, Buck et al. (1989) reported a threefold increase in the association between preeclampsia/eclampsia and SIDS cases when compared to controls who died of other causes (OR =3.1 95% CI 0.4, 25.7) or liv e controls (OR=3.4, 95% CI 0.8, 14.0). Risk Factors that Prompt Complications in Pregnancy The identification of risk factors has been an effective means towards determining what is currently known about SIDS, yet it is important to note th at the risk factors associated with SIDS should not be mistaken as the cause of SIDS nor are reliable in predicting the outcome (SID S Network, 1995). The awareness of risk factors is beneficial in modifying behavi or and establishing preventative measures to decrease the outcome of SIDS. While a numbe r of risk factors have been developed as a result of two decades of SIDS research, the focus of this study will remain on those related to maternal behaviors and underlying health problems dur ing pregnancy, specifica lly the findings of pregnant women that include maternal toba cco use, anemia, and late initiation into prenatal care. Maternal Tobacco Use during Pregnancy Smoking during pregnancy has been asso ciated with a number of perinatal complications that have likewise been linked to SIDS. Women who smoke while pregnant have a dose-dependent increased risk for placenta previa, placenta abruptio,

PAGE 21

14 preeclampsia, and premature rupture of membranes (Schuler-Maloney & Lee, 1998). Support from the literature suggests that smoking is related to an increased incidence of placental previa, as much as 5 per 1000 in nonsmokers in comparison to 20 per 1000 in smokers (Andres & Day, 2000). A meta-ana lysis of studies conducted by Ananth, Smulian & Vintzileos (1999) concentrati ng on the association between pregnancy complications and smoking demonstrated a 90% increase in the incidence of abruptio placentae, with the attributable risk ra nging from 15-25%. Furthermore, women who smoke with preeclampsia were at greater risk of the development of abruptio placentae. Mothers who smoke during pregnancy are th ree times more likely to have a SIDS baby, and exposure to passive smoke doubles the risk (SIDS Alliance, 1998). The adverse effects of smoking have been show n to increase the arousal threshold during quiet sleep, which may lead to the weakened ability of the infant to respond to life threatening events (Horne et al., 2002). Furthermore, nicotine has been proven to decrease the blood flow into the placenta along with increase fe tal carbon monoxide and carboxyhemoglobin concentrations resulting in fetal hypoxia (Bulterys et al., 1990). In a prospective study that obtained smoking fre quency and amount from a questionnaire at three different intervals during pregna ncy, approximately 25,000 live infants were followed from birth to the first year to study the smoking habits of women during pregnancy and the later outcome of their infant (Wisborg, Kesmodel, Hendriksen, Olsen & Secher, 2000). The death records and hospita l records were reviewed to ascertain the cause of death as SIDS. Crude findings reveal ed that infants of smokers were three times more likely to die of SIDS compared to infants on non-smokers (OR 3.5 95% CI 1.4, 8.7) with the risk increasing with the number of cigarettes smoked each day (p< .0.05). After

PAGE 22

15 adjustments were made for maternal age, bi rth weight, and gestational age, the risk remained with smokers being 2.9 times more lik ely to have an infant die of SIDS when compared to non-smokers (Wisborg et al., 2000). Anemia An extension of the fetal hypoxia hypothe sis includes the pos sible interaction between maternal tobacco use and anemia dur ing pregnancy and the outcome of SIDS. A nested case-control study of the US Collaborative Perinata l Project cohort by Bulterys, Greenland & Kraus (1990) found a dose-respon se relationship between smoking and low hematocrit that was also greater than multiplicative when both variables were present. Infants were at greatest risk of SIDS if th e mother smoked more than ten cigarettes a day and had a hematocrit less than 30% (OR= 4.0 95% CI 2.1, 7.4), but low hematocrit in nonsmokers was not an increased risk (Bulterys et al., 1990). It is hypothesized that the lack of oxygen supply (caused by maternal an emia) to the fetus is intensified by the introduction of nicotine in the system. This effect, along with the hypothesis that smoking decreases the necessary nutrients required for central nervous system development, may predispose an infant to greater risk of SIDS. Late Initiation of Prenatal Care The health of a pregnant woman has a profound affect on the health of the developing fetus and newborn (Beggs et al., 2001) The goal of prenatal care is not only to monitor the development of the fetus dur ing pregnancy, but to assess for high risk factors that may have a nega tive effect on the pregnancy outcome. While the various potential for risk factors do not endanger the pregnancy to the same extent, it is important that they are identified early in the prenatal period so that appropr iate interventions are

PAGE 23

16 established to ensure the well being of the mo ther and child. The schedule of visits is generally every four weeks during the first two trimeste rs (the first 28 weeks of gestation), then every two w eeks until the 36th week of gest ation, and then every week until birth (Olds, London & Ladewig, 1996). The lack of adequate prenatal care is cons idered to be one of the most indirect significant risk factors associated with SIDS since all of the complications noted in this review can not only be modified and/or pr evented, but if acquired, can be monitored closely under medical supervision. In comparison to other causes of infant mortality, the odds ratio of the delay in prenatal care is hi gher in infants who died of SIDS (OR=2.2 – versusOR=1.6) (Peterson, Van Belle & Chi n, 1979). An inverse correlation was found in the rate of SIDS and the time in which in itiation of prenatal car e began. Standfast et al. (1980) show that the SIDS rate was highest among those with no prenatal care, with the rate of 4.4 per 1000 in infants who received no prenatal care, which then decreased to 1.04 per 1000 if prenatal care wa s initiated in the first th ree months (Table 5). Analysis of over 10 million live births from the National Center for Health Statistics data in the United States between 1995 and 1997 also revealed that a relationship exists between the absence of pr enatal care and postne onatal death in women with high-risk pregnancy conditions (Vintz ileos, Ananth, Smulian, Scorza, & Knuppel, 2002). The rates of postneonatal death increas ed from 2 per 1000 with prenatal care to 7.5 per 1000 in its absence. Additional fi ndings show a 1.8 fold increase among African American women and a 1.59 fold increase in White women, with the association increasing in the following conditions: pr egnancy-induced hypertension, maternal anemia, and abruptio placent ae (Vintzileoa et al., 2002).

PAGE 24

17 Contribution to the Literature The establishment of the tr iple risk hypothesis, speci fically the concept of the vulnerable infant, has provided an agenda for past and future research in terms of the relationship between complicati ons in pregnancy and the outcome of SIDS. The prenatal risks involved with anemia, placental abnorma lities, uterine bleeding, pregnancy induced hypertension, and eclampsia all share a comm on link to the mechanisms involved with fetal hypoxia, and consequently are related to SIDS. The early ini tiation into prenatal care provides the best opportunity to detect high-risk behavi ors, such as smoking, as well as other risk factors associat ed with poor infant outcomes. While a greater part of the SIDS research supports the hypot hesis of the triple risk model, a bulk of the literatur e focus on the exogenous stressors that have an influence on SIDS. The publications on the cause of the vulnerable infant have been limited in recent years, and are assumed to be nonexistent in the populations unde r study. The present study will investigate the women and infa nts of Hillsborough County and Duval County to develop the profile of the SIDS infant, es tablish the likelihood of significant maternal risk factors, and determine if specific condi tions that can cause fetal hypoxia during the prenatal period exist and consequently increase the risk of an infant becoming vulnerable during the first months of life.

PAGE 25

18 Chapter Three Methods In 1991, former Governor Lawton Chiles passed Florida Legislation promoting the Healthy Start program with the intent of reducing infant mo rtality and low birth weight babies in addition to improving health outcomes of pregnant women. In over a decade, more than 30 coalitions have been established statewide consisting of representatives of the comm unity that develop and implement programs to enhance maternal and child health. In order to plan and evaluate the Healthy Start programs in each of the participating coun ties, the Florida Department of Health, the Agency for Health care Administration, the Lawton and Rhea Chile Center for Healthy Mothers and Healthy Babies and the University of Flor ida Perinatal Data Research Center have worked collaboratively to develop the data linking birth and death records to Healthy Start prenatal screens. As illustrated in Tables 1 and 2, the counties of Hillsborough and Duval have been chosen for this research because of the similarities they have in population size, demographic characteristics a nd live birth and deat h rates for 2000 (Unites States Census Bureau, 2000 & Florida Department of Healt h, 2000). This correlational study was based on a retrospective review of the data sets provided by the Healthy Start Coalition of Hillsborough County and the Northeast Florida Healthy Start Coalition in Duval County. The sections of The Birth and Fetal Death reco rds linked to Healthy St art prenatal screens

PAGE 26

19 and Infant Deaths that were of relevance to this study consisted of the birth files and death records of infants in Hillsborough County and Duval County between January 1, 1998 and December 31, 2000. Specifically, the cohort for analysis focused on the maternal complications in pregnancy listed on the birth certificate and the cause of death of the infant recorded on the death certificate. The initial identification of subjects was de termined by the infant’s cause of death disclosed as Sudden Infant Death Syndrome. The codes for this diagnosis, depending on the cohort year, were categor ized as either ICD-9 code 798.0 or ICD-10 code R95. Secondly, women who gave birth during this time who reported any of the six pregnancy complications of interest on the birth certifi cate were selected (See Appendix A). The six pregnancy complications, identified as anemia, pregnancy-induced hypertension, eclampsia, uterine bleeding, placenta previa a nd abruptio placentae, were chosen based on their association with fe tal hypoxia. The cohorts were studied to discover the frequency of SIDS deaths, and in these cases whether or not the mother was diagnosed with any prenatal complications. Additional analysis was performed to compare these SIDS cases to the overall prevalence of specifi ed pregnancy complications in this cohort to determine whether the complications in pr egnancy were more frequent in cases of SIDS deaths when compared to the ov erall prevalence of such events. Subjects The infants with an ICD-9 code of 798.0 or ICD-10 code R95 who had a linked death to birth certificate in Hillsboro ugh County or Duval County between 1998 and 2000 were selected to determine the number of SIDS cases. Fu rther investigation determined if the mothers of these infant s reported any pertinen t medical history or

PAGE 27

20 complications during pregnancy. Subsequently the record of women during this period of time who indicated complicat ions during pregnancy was determined. The final sample size consisted of the 69 infants who died of SIDS, 34 from Hillsborough County and 35 from Duval County, and the to tal number of women who ga ve birth in these counties (n=86,342) between 1998 and 2000, with particul ar attention placed on those women who were reported to have one or more of th e specified complications during pregnancy (n=6249). The sample represented a va riety of social, economic and cultural backgrounds of women in two comparable count ies in the state of Florida who delivered babies during the period of 1998 to 2000. Measures The pregnancy complications were meas ured by a set of defined codes, as illustrated in Appendix A, in the data dict ionary of the document titled The Birth and Fetal Death Records Linked to Healthy Start Prenatal Screens and Infant Deaths (19982000). The codes were based on information obt ained from the birth certificate under the categories of either Medical History Factors for Pregnancy (BMEDFAC) or Complications of Labor and Delivery (BCO MPFAC). The choice of complications during pregnancy was based on those cited in pr evious studies that are associated with fetal hypoxia. The diagnoses include: anemia, eclampsia, hypertension/pregnancyassociated, placenta previa, abruptio placentae, and uterine bleeding. The cause of death as SIDS was determined by the established ICD-9 code 798.0 or ICD-10 code R95 indicated on the deat h certificate. In order to create a demographic profile of infants who died of SIDS in Hillsborough County and Duval County, the frequency of gender (BSEX), race

PAGE 28

21 (BCRACE), and age (DDOD-BDOB), and time of y ear were analyzed. Other risk factors associated with both complications in pregna ncy and SIDS, defined as maternal tobacco use and initiation into prenatal care, were also examined. The measure of tobacco use (BTOBUSE) and quantity (BTOBNUM) was de termined in these populations, which were analyzed further to determine if this variable served as a confounder between each of the pregnancy complications and the out come of SIDS. In addition, the month prenatal care was initiated (BPNCBGN) and th e number of prenatal visits (BPNVIS) was studied to investigate whether the impact of late prenatal care a nd infrequent prenatal care was associated with poor infant outcomes. Data Collection The data was collected from the delive ry cohort of the Birth and Fetal Death Records Linked to Healthy Start Prenatal Sc reens and Infant Deaths for the consecutive years between 1998 and 2000. The data sets for both Hillsborough County and Duval County were analyzed to determine the dem ographic profile, pregnancy complications, maternal risk behaviors, and cause of deat h of these cohorts. Except for the cause of death, the compilation of this information re lied solely on the documentation made by the health care provider and the self-report by the mother at time of delivery, which was then recorded on the birth certificat e. The cause of death was based on findings of the medical examiner that were subsequently document ed with a specific ICD-9 or ICD-10 code on the death certificate filed with the state of Florida. Data Analysis The objective of the analysis was to explore the frequency and demographic profile of infants who died of SIDS and the incidence of prenatal complications in the

PAGE 29

22 women who gave birth between 1998 and 2000 in Hillsborough County and Duval County. Initially, frequencies were obtained for the variables of r ace, age of death, and gender of the infant and the tobacco use, time of initiati on into prenatal care, and pregnancy complications of the mother from Statistical Package for Social Sciences (SPSS). Chi-Square analysis and values of Cramer’s V were performed using Statistical Analysis Software (SAS) Version 8.01 to address each of the hypotheses and determine whether a correlation existed between th e independent variables of pregnancy complications, maternal tobacco use, and hi story of prenatal care and the dependent variable of SIDS. In order to measure th e association between the variables from an epidemiological perspective, a crude odds rati o was determined for each of the pregnancy complications, maternal tobacco use, and ini tiation into prenatal car e and the outcome of SIDS with 95% confidence interval calculatio ns to determine significance of findings. Protection of Human Subjects A feature of secondary data analysis preven ted the necessity of direct contact with the participants in the study sample. The identification of participants was not known by the researcher throughout the investigati on of these established cohorts. A written protocol was submitted to the Institutional Review Board (IRB) at the University of South Florida requesting approval to conduct th e research. A written letter was received by the researcher stating that the project met the federal criteria to qualify as an exempt study because there was not direct involvement with human participants (Appendix B).

PAGE 30

23 Chapter Four Results The present research was a retrospectiv e study based on a secondary data analysis of the Birth and Fetal Death Records Linked to Healthy Start Prenatal Screens and Infant Deaths of Hillsborough County and Duval County for the time between 1998 and 2000. The purpose was to investigate the relationship between complications and risk factors in pregnancy associated with fetal hypoxia and the outcome of SIDS. The first section presents a description of the sample and dem ographic profile of infants who died of SIDS in the chosen counties. The second secti on analyzes the hypothe ses of the study. The first six hypotheses are relate d to the specified complicati ons in pregnancy, defined as anemia, eclampsia, pregnancy induced hypertensi on, placenta previa, ab ruptio placentae, and uterine bleeding and their association with SIDS. The seventh and eighth hypotheses address maternal tobacco use and prenatal care respectively and their association with SIDS. Lastly, the ninth hypothesis examined the role of maternal smoking and its possible confounding relationship between the specified prenatal complications as a group and SIDS. Descriptive Statistics of Sample According to the data files of the Bi rth and Fetal Death Records Linked to Healthy Start Prenatal Screens and Infant D eaths document, the sample consisted of

PAGE 31

24 86, 342 births 45,877 (53%) in Hillsbor ough County and 40,465 (47%) in Duval County, between 1998 and 2000. Of the 86, 342 births the rate of havi ng at least one of the specified pregnancy complications was 72.37 per 1000 births (n=6,249). The classification of the 6,249 pregnancy compli cations, as noted in Table 3, was as following: anemia; 24.13%, pregnancy-i nduced hypertension; 49.62%, eclampsia; 13.20%, uterine bleeding; 2.7%, placenta previa ; and 6.27%, abruptio placentae; 4.00%. On average, 33.63% of the women with one of the six pregnancy complications of interest reported more than one condition on the birth certif icate. Of the women with anemia, 30.90% reported more than one complication, in addition, 30.18% with pregnancy-induced hypertension, 45.23% with eclampsia, 37.87% with uterine bleeding, 43.88% with placenta previa and 26.8% of thos e with abruptio placentae were found to have more than one complication during pregna ncy. The examination of the births that later resulted in SIDS rev ealed that 8.69% of the women had more than one of the pregnancy complications, of these, 4.34% reporting anemia or pregnancy induced hypertension, 1.5% with eclampsia or abrupt io placentae. There were no reports of uterine bleeding or placenta previa among these cases of SIDS. The overall rate of SIDS was 0.80 per 1000 births (n=69). As shown in Table 4, the cases of SIDS consisted of 65.21% male s and 34.78% females, of which 60.86% were White, 37.68% were Black, and 1.45% were categorized as Native American. The average age at death was approximately 80 days. The report of tobacco use during pregnancy was 26.08% among the SIDS cases, with approximately 61.11% of those who smoked admitting to smoking at least a pack a day and 38.89% smoking at least 1 pack per day. Among the SIDS cases, 73.13% of the wo men initiated prenatal care in the first

PAGE 32

25 trimester, 17.91% in the second trimester, and 8.95% at the se venth month or not at all. The highest percent (17.39%) of the prenatal care visits consisted of the recommended number for a healthy pregnancy of 12, yet 40.61% of women received 10 or less prenatal care visits. The crude risk ratios used to measure the association for each of the pregnancy complications and maternal risk factor and the outcome of SIDS are shown in Tables 5 through 13. Of the six complications of inte rest, significance was found in women with eclampsia being 4.7 times more likely to have an infant die of SIDS when compared to women without eclampsia (95% CI 1.67, 16.29) The risk behaviors of maternal smoking and late initiation in to prenatal care were also found to have a positive association with 95% confiden ce in the outcome of SIDS. Further analysis of the initiation into prenatal care showed the highest risk among those who receive no prenatal care (OR=4.37, 95% CI 1.38, 13.89) when compared to those who initiated prenatal care sometime during the pregnancy. Analyses of Hypotheses I-VI Initial analysis was performed to determine if Hillsborough County and Duval County differed in terms of the type of pr egnancy complications. When grouping all the maternal diagnoses together, it was concluded that the two counties we re different in the distribution of each of the six pregnancy co mplications, defined as anemia, eclampsia, pregnancy induced hypertension, placenta pr evia, abruptio placentae, and uterine bleeding (p<. 0001) with a Cramer’s V of 0.4, which suggests a strong measure of association exists. Because th e distribution of the frequenc ies was so different between the two counties, further analysis was performe d to ensure that the county did not serve

PAGE 33

26 as a confounder between each of the pregnancy complications and the outcome of SIDS. Chi square tests were performed separately for each county for the variables of anemia, PIH, and eclampsia. Statistical significan ce was found in Duval eclampsia cases only (Chi Square 8.67, p=0.0032). A Breslow Day Test of Homogeneity was done to determine if the odds ratios were significantly different between the two counties. Because each of the tests were proven to be insignificant, it was concluded that the county in which the SIDS infant was born was not a confounder between the pregnancy complication and the outcome of SIDS, and th e odds ratios illustrated in Tables 5-7 denote the measure of association. The first six hypotheses addressed the research question as to whether an association existed between each of the pr egnancy complications, and the outcome of SIDS. A Chi-Square test was performed on each of the pregnancy complications using SAS, with the strength of the relationship re ported in Cramer’s V values. In order to keep the overall of .05 under control, a modified B onferroni procedure identified as a Holm procedure, was used to test these hypot heses in a family-wise approach. The Holm procedure only declared statistical signi ficance in the relationship between the complication of eclampsia and SIDS (p<. 05), yet Cramer’s V showed the strength of the association was valued at 0.01, which is gene rally considered to be a rather weak association. Analysis of Hypothesis VII An association exists between women who do not receive pren atal care and the outcome of SIDS (Chi-Square 7.45, p<. 05, Cram er’s V=. 01). Further analysis was performed to investigate the relationship be tween early initiation into prenatal care,

PAGE 34

27 defined as months 1-4, and late or no pren atal care. An association exists between women who receive prenatal care at month 5 or later a nd the outcome of SIDS (Chi-Square 11.22, p<. 05, Cramer’s V= .01). Analysis of Hypothesis VIII An association exists between women w ho smoke and the outcome of SIDS (ChiSquare 29.28, p<. 0001), strength of the asso ciation is minimal (Cramer’s V= .02). Analysis of Hypothesis IX It was hypothesized that smoking may act as a potential conf ounder between the pregnancy complications and the outcome of SIDS. Table 14 shows a crude breakdown of the population based on smoking status with the adjusted odds ratio after controlling for smoking. The purpose of using the Coch ran Mantel-Haenszel Statistics was to determine if the group of pregnancy complicat ions was independent of the SIDS cases when adjusting for smoking. Further analysis was performed using SAS to determine the results of the Breslow-Day Test for Homoge neity of the Odd Ratio, which was found to be insignificant at p=. 7301. The null hypothe sis of homogeneity be tween the odds ratios is accepted; therefore, there is no difference in the odds ratios between those who smoke and the nonsmokers. It is concluded that smoking is not a confounder in this study and that a woman with one of the pregnancy co mplications is 2.2 more likely to have an infant die of SIDS when controlling fo r smoking status (95% CI= 1.13-4.31).

PAGE 35

28 Chapter Five Discussion For over three decades, fetal hypoxia has b een hypothesized to be a predisposed factor that increases an infant’s risk of dying from SIDS. In spite of this, the studies that have tested this hypothesis have been limite d. The main purpose of the current study was to present additional information on the prevalence of pregnancy complications associated with fetal hypoxia and the relationship each has on the future outcome of SIDS. An extension of this analysis assisted in the development of a descriptive profile of the SIDS infant unique to the populations of Hillsborough County, Florida and Duval County, Florida. Specifically, the characterist ics of race, gender, age of death, and the inquiry of particular maternal risk factors were revealed to assist with the proper interventions that can be established to re duce the frequency of SI DS in each of these populations. The Profile of the SIDS Infant The demographic profile of the infants who died of SIDS in Hillsborough County and Duval County reveal the highest fre quency among males, with 62.22% being White and 33.33% being Black. It is unclear as to the reason, but these results are similar to past research that have found a 50% increase d risk of male infant s dying of SIDS when compared to females (Hoffman & Hillman, 1992). The analysis of each population shows that 78.78% of SIDS infants are White in Hillsborough County compared to

PAGE 36

29 51.42% in Duval County; a finding that complete ly contradicts the cu rrent research that proclaims that African American infants are nearly two-and-a-half times more likely to die of SIDS than white infants (SIDS Allian ce, 1998). From the perspective that supports the environmental influences of SIDS, it has been hypothesized that the racial differences in these counties may be the result of intense program development and intervention supported by the ‘Back to Sleep Campaign’ that was initiated by the Healthy Start programs prior to 1998. During this time, em phasis may have been placed on the Black population because research had strongly suggested that this is where the greatest effort was needed to decrease the rates of infant mortality (Perrin, Stanley, Myers, BerneckiDejoy, Harris, & Perrin, 2002). The average infant’s age at death is approximately 80 days, with 61.01% of the deaths taking place between the months of October and March. This is a mind provoking finding considering that the co rrelation of SIDS deaths dur ing the winter months are often related to cold climat e, heavier bedding, and heated houses (Hillman & Hoffman, 1992). Such climate is rare in this part of the country, yet the findings are consistent with other publications that examine environmental f actors. The age of the infant at death has led to a number of etiological inferences; however it is important to consider the actual gestational age at birth sin ce the stage of anatomical a nd physiological development may differ (Hoffman & Hillman, 1992). Furthermore, studies have also taken a closer look into racial differences and maternal tob acco use among those who die earlier, yet the efforts made to confirm these results have not been remarkable (Haglund & Cnattingius, 1990; Hoffman & Hillman, 1992).

PAGE 37

30 Summary of Maternal Pregnancy Complications Of the 86,342 births in Hillsborough C ounty and Duval County between 1998 and 2000, 6,249 reported having at least one of the pr egnancy complications associated with fetal hypoxia and 2,102 had more than one. Overall, the incide nce of placental abnormalities in these two counties were consis tent with the current literature which states the occurrence of placenta previa is approximately 1 in 200 births and abruptio placentae occurs in 0.4-3.5% of births (Beers & Berkow, 1999). Considering the significance of uterine bleeding in other stud ies and the likelihood that it is associated with a placental abnormality, it is surprising that the frequenc y of this diagnosis was so low. Standfast et al. (1980), report that abnormal uterine bleeding “stands out as being strongly associated with SIDS at 58%”, yet the sample size of SIDS mothers was considerably smaller than the current study (p. 1064). Furthermore, abnormal uterine bleeding was the only significant risk factor reported by Buck et al. (1989) when the comparison was made between SIDS infants and non-SIDS controls. The findings of the current study show th at uterine bleeding only occurred in 0.19% of women who reported one of the comp lications in pregnancy associated with fetal hypoxia and none were reported among the SI DS cases. Further scrutiny of the data dictionary that was developed as part of the document titled The Birth and Fetal Death records Linked to Healthy Start Prenatal Sc reens and Infant Deaths revealed a possible explanation for these findings. The diagnosis of uterine bleeding was measured under the variable of Medical History Factors for Pregnancy and the diagnoses of placenta previa and abruptio placentae were categorized as a Complication of Labor and Delivery. As a result, it could not have been assumed th at a connection could be made between the

PAGE 38

31 complaint of uterine bleeding and the occurr ence of one of the placental abnormalities. Additionally, the code for “other excessive bleed ing” that was located in the same section of Complications of Labor and Delivery may ha ve been used for the uterine bleeding that occurs with placenta previa and abruptio placent ae, but the lack of definition for this code made it impossible to make such an assumption for analysis purposes. The mothers of Duval County had signifi cantly higher rates of anemia (17.11%-vs 7.03%) and eclampsia (12.02%-versus-1.25%) wh en compared to Hillsborough County. The complication most reported by the moth ers of Hillsborough County was pregnancyinduced hypertension (PIH= 34.84%). Taking in to consideration th e severity of the condition and the fact that only 1 in 200 pa tients with preeclamps ia will eventually develop eclampsia, it is intere sting to note the findings of the relationship between PIH and eclampsia in this study. In Duval County, 921 women reported having PIH during their pregnancy, 751 reported eclampsia, a nd 11% of those with PIH reported having eclampsia. Whereas in Hillsborough Count y, 2180 women experienced PIH and only 78 reported the diagnosis of eclampsia. The ra te of those with PIH in Hillsborough County who also reported eclampsia was approxima tely 1.46%. The possible explanation for such a discrepancy may be a result of se lf-report bias on the birth certificate. It is speculated that the education received by the obstetric patient during prenatal care may not have provided a clear differen tiation between the risk of hypertension and how it may result in more severe problems such as preeclampsia or eclampsia. Furthermore, the terminology describing hype rtension, preeclampsia and eclampsia may have been used interchangeably, and with th e only choice on the bi rth certificate being eclampsia, this option may have triggered r ecall of such terminol ogy. Because of the

PAGE 39

32 potential of this possibility, it may be necessary to use caution in the upcoming interpretation of the SIDS outcomes. It is recognized that the mo st accurate method for preventing this discrepancy would be to vali date the information documented on the birth certificate against medical records of the pr enatal care visits, yet such a measure is beyond the scope of the study since strict measur es to protect patient confidentiality have made it difficult to obtain such information. Hypotheses of Association between Pregnancy Complications and SIDS The statistical approach in this st udy included the Chi-Square Test of Independence and the Cramer’s V value. Th e six pregnancy complications associated with fetal hypoxia that were the focus of the study include d anemia, eclampsia, pregnancy induced hypertension, uterine bl eeding, placenta previa and abruptio placentae. Generally speaking, it is difficu lt to compare the findings of this study to others that investigate similar associations between these pregnancy complications and SIDS because the complexity of their anal yses allowed for more optimal control of potential confounders by the use of multivariate analyses. With that being said, the preliminary findings of these populations are not in agreement with other studies. The frequency of uterine bleeding and placenta pr evia are absent in the cases of SIDS and statistical significance is found only in the complication of eclampsia with the magnitude of the strength being minimal. Although th e use of the Holm’s test was conducted to keep the level at .05 for the entire group of pr egnancy complications, it is questionable as to the validity of these findings consider ing the dilemma mentioned previously of the probable report bias and confusion of te rminology between hypertension and the more severe diagnoses of preeclampsia and eclampsia. Because of the possibility of

PAGE 40

33 misdiagnosis, an analysis was performed beyond the original calculation by grouping eclampsia and pregnancy-induced hypertension together as one category to determine if infants were more likely to die of SIDS. On ce they were combined, it is worthy to note that significance found in eclampsia alone was eliminated, and together the findings were not significant (p value=0.09). Additional research is needed to determine the accuracy of the self-report on the birth certificate. It is recommended that a more in depth examination of medical records be carried out to verify the extent of the severity of hypertension in these counties as well as de termine the prevalence of uterine bleeding that occurs simultaneously with one of the placental abnormalities. Hypotheses of Maternal Tobacco Use The selection of the risk factors related to maternal tobacco use was chosen for analysis because it is related to the adverse events that occur duri ng pregnancy as well as the outcome of SIDS. The toxic effect of ni cotine and other materi als in cigarettes are known to decrease the blood flow from the pl acenta to the fetus, and smoking increases the risk of adverse events such as hy pertension and placental abnormalities (SchulerMaloney & Lee, 1998). It also augments the pr obability that these conditions will cause some degree of fetal hypoxia. On a national level, it is estimated th at approximately 12% of all women report smoking during pregnancy (Centers for Dis ease Control & Prevention, [CDCP], 2001). In an analysis of similar data used in this study, it has been estimated that infant deaths would decline by 12.52% if women in Florida received adequate pren atal care and did not smoke during pregnancy (Thompson, Simmo ns & Graham, 2002). According to the current study, 10% of the women who gave birth between 1998 and 2000 in Hillsborough

PAGE 41

34 County and Duval County reported smoking on the birth certificate. The women who reported smoking during pregnancy were 3 times more likely to have an infant die of SIDS when compared to nonsmokers (95 % CI =1.83, 5.36). This is an unadjusted risk that does not take factors such as age, socio economic status, race, or education level into consideration. However, the smoking status of the mother was obtained before the outcome of SIDS was known, it is impossibl e that misclassification of smoking or outcome has occurred in this analysis. The only concern is that the self-report of smoking during pregnancy may be underestim ated, yet this would not decrease the present strength of the association, it would only make it more pronounced. Interestingly, the number of smokers in this cohort (n=8,723), 6.6% experienced at least one of the six pregna ncy complications of interest in the study compared to 7.3% of nonsmokers. Because the rates of pregnancy complications are higher among nonsmokers, caution should be used in the in terpretation of these results. A possible explanation of these findings may be relate d to the fact that 90% of those with a pregnancy complication were classified as non-smokers. The factor of smoking was tested to determine if confounding existed be tween pregnancy complications and SIDS. Because the frequency of each of the six comp lications was so small in the SIDS cases, the diagnoses were grouped together. The results showed that smoking was not a confounder between the two variab les, and that a risk remains between the two variables even after controlling for toba cco use. When taking the in fluence of self report into consideration, it is possible that the women who smoked during pregnancy and experienced a complication were aware of the association between the exposure to smoking and the risks involved, therefore ha d may not reported the tobacco use but

PAGE 42

35 perhaps felt obliged to repor t the pregnancy complications since verification of the medical record was feasible. Hypothesis of Late Initia tion into Prenatal Care Women who initiated pren atal care during the second half of their pregnancy (months 5-9) are 3 times more likely to have an infant die of SIDS when compared to those who receive prenatal care in the first four months of pregnancy. (95% CI= 1.39, 4.66) and the risk increases to 4.37 for those who receive no prenat al care (95% CI= 1.37, 13.89). While it is possible that women do not recall the exact number of prenatal care visits, 40.6% received less than 10 visits when the recommended number with no symptoms of a high-risk pregnancy is 12 visits These results are consistent with other studies that affirm the mothers of SIDS cases are more likely to initiate prenatal care after the first trimester (Pet erson et al., 1979; Buck et al., 1991) In addition, the risk rate of 4.37 in those with no prenatal care is equivale nt to findings based on the same type of analysis reported by Standfast et al. (1980). Limitations of Study The major limitations of this research design are related to the secondary analysis of state managed birth and death certificate data. The quality of the data sets is questionable since it relies heavily on the accu rate self-report of the mother and thorough documentation of the health care provider at time of delivery. A lthough the diagnoses of interest have shown substantial inference to causality in other studies, it is believed that these results were determined because of th e opportunity to verify the data on the birth certificate to the documentation on the medi cal record. The correctness of methods involved in verification of in formation, data cleansing and management were constrained

PAGE 43

36 in this study due to the fact that it was ba sed on a secondary data analysis and the access to medical records was not possible. Furthe rmore, the codes assigned to each of the pregnancy complications were restricted, th e source of data entry is unknown, and the method in which the data was entered was impractical and difficult to analyze. Combining Hillsborough County and Duval County was effective in increasing the sample size, yet this does not change th e fact that the inde pendent and dependent variables of the study are considered to be ra re. The rate of SIDS is approximately 0.8 per 1,000, and a majority of the measured preg nancy complications occur in less than 8 per 1,000 births in these counties. It is necessary to increas e the population further if the intention is to increase the exposures and out comes, yet it is essent ial that the choice of additional populations be comparable to thes e current counties in or der to preserve the generalizability and external validity of the findings. Lastly, while the choice of the univariat e analyses has been beneficial in apprehending a broad understanding of the a ssociations that exist within the populations of Hillsborough County and Duval County, the discovery of the subtle relationships require more complex techniques of analyses that will measure and control for extraneous factors and assist in a more thorough understand ing of the interactions that exist. The improvement in both research de sign and analysis in future studies should address other determinants that may influence these resu lts and facilitate the refinement of the associations that have been presente d, if they do indeed exist. Implications Over thirty years ago, a panel of physicia ns collaborated to review evidence of infants who died suddenly and unexpectedly in the postneonatal period. While the

PAGE 44

37 identification of the term Sudden Infant D eath Syndrome or SIDS was simple, the issue involved in the classification of infants ha s not been an easy task. SIDS remains a ‘diagnosis of exclusion’, and de spite the vast amount of research from multiple health disciplines, the definition continues to be imprecise and adaptable to each pathology department across the na tion (Beckwith, 2003). While the problem with the definition of SIDS continues, immense progress has been made in recent years to reduce the number of infants who die of SIDS. Research has proven that SIDS is undeniably mulitfactor ial and that it is influenced by various potential factors as evidenced by the triple ri sk model. Because the model begins with the vulnerable infant, a preference was made to focus on events that occur during pregnancy that may lead to the outcome of the vulnerable infant. Since the medical recommendations, public awareness campaigns and education on prevention have not been effective in the complete eradica tion of SIDS, the events surrounding fetal development and birth call for additional res earch regarding their role in SIDS. The outcome of SIDS continues to devastate approximately 9 families a day when a baby is found dead of “unexplained” reasons. The fi ndings of this study have contributed to the current evidence relating prenatal events to SIDS. The public health implications can guide the necessary intervention towards a greater awareness and more effective preventable measures in hopes to eliminate SIDS overall. Recommendations for Future Study The obvious goal of SIDS research is to determine the etiology of the event, establish preventative measures and to improve maternal and child health so that SIDS will no longer exist. For these objectives to be successful, research such as this must

PAGE 45

38 continue so that the framework of knowle dge and resources can expand beyond what is currently known. The findings of this current study can direct in future proceedings specific to the populations of Hillsborough C ounty and Duval County, Florida. While the birth and death certific ate data has proven beneficial in developing a broad depiction of the prenatal events that occur, it is reco mmended that a more detailed investigation of other types of records be explored to veri fy the findings of this study. Although lengthy and expensive, the most accurate method to obtain vital information on the events that take place during pregnancy as well as identif y potential factors that categorize a woman as high risk is to follow them prospectiv ely from initiation into prenatal care. Furthermore, the benefits of the periodic a ssessment of women in this population will not only assist in the development of a high risk profile, but will provide an optimal opportunity and environment to educate th e woman on the importance of maintaining good health during pregnancy. An alternativ e method, though not as accurate, would be to examine the medical records and interview the mothers at time of delivery, when the events of the pregnancy are easily remembered and attainable. It is recommended that funding increase at the state level for community-based organizations such as Healthy Start to devel op social marketing sche mes that will target the women who are smoking during their pregnanc y and initiating prenat al care later than the recommended time. The evaluation of Healthy Start sponsored programs have proven to be quite beneficial in the reduction of infant mort ality since home visits to new mothers were implemented (Florida Depart ment of Health, 2000-2003), yet flexibility and resources are necessary to ensure cove rage of the population during the prenatal period as well as after the infant is born. In conclusion, the community needs to take

PAGE 46

39 aggressive measures to reach these wome n and guarantee that they are receiving the necessary healthcare if work constraints and time limitations are preventing the early initiation into prenatal care

PAGE 47

40 References Ananth, C. V., Smulian, J. C., & Vintzile os, A. M. (1999). Incidence of Placental Abruption in Relation to Cigarette Sm oking and Hypertensive Disorders During Pregnancy: A Meta-Analysis of Observational Studies. Obstetrics and Gynecology, pp. 622-628. Andres, R. L., & Day, M. (2000). Peri natal Complications Associated with Maternal Tobacco Use. Seminars in Neonatalogy, 5, pp. 231-241. American SIDS Institute (2002). Retrieved October 29, 2002, www.sids.org Beckwith, J. B.(March 2003). Defining the Sudden Infant Death Syndrome. Archives of Pediatric an d Adolescent Medicine, 157, pp.286-290. Beers, MH. & Berkow, R (Eds.). (1999). The Merck Manual of Diagnosis and Therapy (17th ed.) Chapter 260. Disturban ces in Newborns and Infants Retrieved October 24, 2002, http://www.merck.com/pubs/ mmanual/section18/chapter250.htm Beggs, A. H., Bucks, G. Corwin, M. Dailey, D. Eaglestaff, M. L., Fifer, W. P., & Jacobson, J. (2001). From Cells to Selves. Targeting Sudden Infant Death Syndrome (SIDS): A Strategic Plan. Retrieved September 7, 2002, http://www.nichd.nih.gov Buck, G. M., Cookfair, D. L., Michalek, A. D., Nasca, P. C., Standfast, S. J., & Severs, L.E. (1989). Assessment of in utero Hypoxia and Risk of Sudden Infant Death Syndrome. Paediatric and Perinatal Epidemiology 3, pp. 157-173. Bulterys, M. G., Greenland, S. & Kraus, J. K. (1990). Chronic Fetal Hypoxia and Sudden Infant Death Syndrome: Interac tion between Maternal Smoking and Low Hematocrit During Pregnancy. Pediatrics, 86, pp. 535-540. Centers for Disease Contro l and Prevention (2001). Women and Smoking: A Report of the Surgeon General. Retrieved March 15, 2003, www.cdc.gov Filiano, J. J., & Kinney, H. C. (1994). A Perspective on Neuropathologic Findings in Victims of the Sudden Infant Deat h Syndrome: The Triple-Risk Model. Biology of the Neonate, 65, pp. 194-197. Filiano, J. J., & Kinney, H. C. (1995) Sudden Infant Death Syndrome and Brainstem Research. Pediatric Annals, 24, pp. 379-83.

PAGE 48

41 Florida Department of Health (1996-2001). Florida Vital Statistics Annual Report: Fetal and Infant Deaths. Retrieved February 2, 2003, www.myfla.gov Florida Department of Health (2000). Florida Vital Statistics Annual Report: Live Births Retrieved February 2, 2003, www.my.fla.gov Florida Department of Health (2000-2003). Healthy Start Fact Sheet. Retrieved April 30, 2003, www.my.fla.gov Haglund, B., Cnattingius, S. (1990). Ciga rette Smoking as a Risk Factor for Sudden Infant Death Syndrome: A Population-Based Study. American Journal of Public Health 80 pp.29-34. Hillsborough County Government Online. Frequently Asked Questions about your Medical Examiner. Retrieved October 31, 2002, http://www.htv22.org/medexam.home.html Hoffman, H. J., & Hillman, L. S. (1992, December). Epidemiology of the Sudden Infant Death Syndrome: Maternal, Neona tal, and Postneonatal Risk Factors. Clinics in Perinatology, 19 (4), pp. 717-736. Horne, R. S., Ferens, D. Walts, A. Vitkovic, J. Lacey, B. Andrews, S. Cranage, S. M.,Chau, B. & Greaves, R. (2002, September). Effects of Maternal Tobacco Smoking, Sleeping Positions and Sleep State on Arousal in Healthy Term Infants. Archives of Disease. Child, Fetal and Neonatal Edition, 87 (2), pp. F100-F105. Kinney, H. C., Filiano, J. J., & Harper, R. M. (1992, March). The Neuropathology of the Sudden Infant Death Syndrome. A Review. Journal of Neuropathology and Experimental Neurology, 51 (2), pp. 115-126. Li, D. & Wi, S. (1999, April 1). Matern al Placental Abnormality and the Risk of Infant Deaths. Pediatrics, 149 (7), pp. 608-611. Li, D. & Wi, S. (2000). Maternal Preeclampsia/Eclampsia and the Risk of Sudden Infant Death Syndrome in Offspring. Paediatric and Perinatal Epidemiology, 14, pp. 141-144. National Center for Health Statistics (2001, October 12). Nati onal Vital Statistics Report Retrieved September 13, 2002, www.cdc.gov/nchs/data National Institute of Child Health an d Human Development (2000, November). Babies Sleep Safest on Their Backs: A Resource Kit for Reducing the Risks of SIDS in African American Communities. Retrieved September 13, 2002, www.nichd.nih.gov/sids/ Naeye, R.L. (1980). Sudden Infant Death. Scientific American, 242, pp. 56-62.

PAGE 49

42 Naeye, R.L., Landis, B. & Drage, J.S. (1976, November). Sudden Infant Death Syndrome. American Journal of Disease in Childhood, 130 pp. 1207-1210. National SIDS Resource Center Retrieved September 13, 2002, www.sidscenter.org Olds, S.B., London, M. L., Ladewig, P.W. (1996). Maternal-Newborn Nursing: A Family Centered Approach (5th ed) Menlo Park: Addison-Wesl ey Nursing, A Division of The Benjamin/Cummings Publishing Company, Inc. Perrin, K.M., Stanley, L., Myers, P., Bern ecki-Dejoy, S., Harris, J., Perrin, S. (2002, December 11). Florida SIDS death certificate data: A profile of risk. Presented at the Eighth Annual Maternal and Child Health Epidemiological Conference, Clearwater, FL. Peterson, D. R., Van Belle, G. & Chinn, N. M. (1979, April 9). Epidemiologic Comparisons of the Sudden Infant Death Syndr ome with other Major Causes of Infant Mortality. American Journal of Epidemiology, 110 (6), pp. 699707. Schuler-Maloney, D. & Lee, S. (1998). The Placenta: To Know Me Is To Love Me. A Reference guide for gross placental examination. Retrieved October 20, 2002, http//showcase.netins.net/web/placenta SIDS Network (1995, July). SIDS Research: Current and Future Directions. Retrieved October 24, 2002, http://sids-network.org Standfast, S. J., Jereb, S. & Janerich, D. T. (1980). The Epidemiology of Sudden Infant Death Syndrome in Upstate New York. American Journal of Public Health, 70, pp. 1061-1067. Sudden Infant Death Syndrome Alliance (1998). Reducing the Risk of SIDS. Retrieved September 13, 2002, www.sidsalliance.org Sudden Infant Death Syndrome Autopsy Protocol, 4 Florida Stat. Ann. § 11G2.0031-2.004 (October 14, 1996). Thomas, C. L. (Ed.), (1993). Taber's Cyclopedia Medical Dictionary (17th ed., p. 1580). Philadelphia: FA Davis Company. Thompson, D., Simmons M., Gr aham, C. (August 2002). Florida Infant Death and Low Birth Weight Attribut able to Absence of Prenat al Care and Tobacco Use of Mother. Retrieved April 25, 2003, www.doh.state.fl.us/family/mch/docs/documents.html United States Department of Health a nd Human Services (1999, October 26,). Back to Sleep Campaign Seeks to Reduce Incidence in African American Populations. Retrieved January 14, 2003, www.nichd.nih.gov

PAGE 50

43 United States Census Bureau (2000). Profile of General Demographic Characteristics. Geographical Area: Duval County, Florida. Retrieved April 2, 2003. www.census.gov United States Census Bureau (2000). Profile of General Demographic Characteristics. Geographical Area: Hillsborough County, Florida. Retrieved April 2, 2003. www.census.gov Voto, L. S., Lapidus, A. M., & Marquiles, M. (1999). Effects of Preeclampsia on the Mother, Fetus, and Child. Retrieved October 23. 2002, http://www.obgyn.net Washington, J. (January 2003). Fetal and Neonatal Hypoxia. Retrieved February 21, 2003, www.continuingeducation.com Wisborg, K.W., Kesmodel, U., Henrikson, T. B., Olsen, S.F., Secher, N.J. (2000). A Prospective Study of Smoking during Pregnancy and SIDS. Archives of Disease in Childhood 83, pp. 203-206. Willinger, M. James, L. S., & Catz, C. (1991). Defining the Sudden Infant Death Syndrome (SIDS): Deliberations of an Expe rt Panel Convened by the National Institute of Child Health and Human Development. Pediatric Pathology, 11, pp. 677-684. Retrieved September 7, 2002 from OVID database.

PAGE 51

44 Bibliography Breslow, N. & Day, N. E. (1980). Statis tical methods in cancer research, Vol I: The analysis of case-control studies. Lyon: IARC. Cody, R. P. & Smith, J. K. (1997). App lied Statistics and the SAS Programming Language. (4th ed.). New Jersey: Prentice Hall. Holm, S. (1979). Simple Sequentially Rejective Multiple Test Procedure. Scandinavian Journal of Statistics 6 : pp. 65-70. Mantel, N. & Haenszel, W. (1959). Sta tistical aspects of the analysis of data from retrospective studies of di sease. Journal of the National Cancer Institute, 22, 719 748. SAS Institute Inc. (1999-2000) (Version 8.1) [SAS System for Windows]. Cary, NC SPSS Inc. (2002) (Version 11.5) [SPSS for Windows]. Chicago, IL: LEAD Technologies Inc. (Contract No. GS-35F-5889H) University of Florida, Perinatal Data Re search Center and Florida Department of Health, Office of Planning, Evaluation, and Data Analysis (October 2000). 1998 Birth and Fetal Death Records Linked to Healthy Start Prenatal Screen s and Infant Deaths. Tallahassee, FL. University of Florida, Perinatal Data Re search Center and Florida Department of Health, Office of Planning, Evaluati on, and Data Analysis (June 2001). 1999 Birth and Fetal Death Records Linked to Healthy St art Prenatal Screen s and Infant Deaths. Tallahassee, FL. University of Florida, Perinatal Data Re search Center and Florida Department of Health, Office of Planning, Evaluati on, and Data Analysis (May 2002). 2000 Birth and Fetal Death Records Linked to Healthy St art Prenatal Screen s and Infant Deaths. Tallahassee, FL.

PAGE 52

45 Table 1 2000 Demographic Profile by County Characteristic Hillsborough Duval Total Population 998,948 778,879 Male 488,772 (48.9%) 377,781 (48.5%) Female 510,176 (51.07%) 401,098 (51.49%) White 750,903 (75.2%) 512,469 (65.8%) Black 149,923 (15.0%) 216,780 (27.8%) Asian 21,947 (2.2%) 21,137 (2.7%) Hispanic/Latino 179,692 (18.0%) 31,946 (4.1%) Age under 5 68,444(6.85%) 56,247 (7.22%) Table 2 2000 Vital Statistics of Resident Live Births and Deaths by County Characteristic Hillsborough (%of FL) Duval (% of FL) Total Live Births 14,662(14.9%) 12,169(15.7) White 11,146(13.4%) 74289(13.2%) Black 2982(22.6% 4223(22.4%) Other 494 (24.8%) 511(23.1%) Total Neonatal Deaths 65 (5.3%) 73(5.0%) White 22(3.0%) 40(3.6%) Black 42(9.9%) 31 (10.4%) Other 1(2.0%) 2(4.0%) Total Infant Deaths 118(9.7%) 116(7.9%) White 46(6.2%) 62(5.6%) Black 71(16.8% 51(17.1%) Other 1(2.0%) 3 (6.1%)

PAGE 53

46 Table 3 Frequency of Pregnancy Co mplication by County and Year Hillsborough Duval _________________ ________________ Pregnancy Complication 1998 1999 2000 Total (%) 1998 1999 2000 Total (%) Anemia 156 143 140 439 (7.03%) 259 347 463 1069 (17.11%) PIH 620 806 754 2180 (34.89%) 318 298 305 921 (14.74%) Eclampsia 22 26 30 78 (1.25%) 222 211 318 751 (12.02%) Uterine Bleeding 39 25 35 99 (1.58%) 17 26 27 70 (1.12%) Placenta Previa 67 77 50 194 (3.10%) 72 71 55 198 (3.17%) Abruptio Placentae 48 49 61 158 (2.53%) 34 30 28 92 (1.47%) Total (n=6249) 952 1126 1070 3148 (50.38%) 922 983 1196 3101 (49.62%) PIH=Pregnancy Induced Hypertension

PAGE 54

47 Table 4 Distribution of Infants who died of SIDS by County between 1998-2000. Variable Hillsborough Duval Total 34 35 Male 20 25 Female 14 10 White 26 16 Black 8 18 Indian 0 1 Age at Death (in days) 93.5+/-68 66.6+/-107 Maternal Smoking Yes 10 8 No 23 27 Number of Cigarettes Smoked 10.5/day 9.25/day Month PNC began 2.78 2.2 Number of PNC visits 12 9.37

PAGE 55

48 Table 5 Maternal anemia during pregnancy and the risk of SIDS Anemia SIDS Non-SIDS Crude OR (95% CI) Yes 3 1508 OR= 2.55 (0.81, 11.18) No 66 84765 Totala 69 86273 a Subjects with missing data were excluded Table 6 Maternal hypertension associated wi th pregnancy and the risk of SIDS PIH SIDS Non-SIDS Crude OR (95% CI) Yes 3 3101 OR= 1.22 (0.39, 3.72) No 66 83172 Totala 69 86273 PIH= Pregnancy Induced Hypertension a Subjects with missing data were excluded Table 7 Maternal eclampsia during pre gnancy and the risk of SIDS Eclampsia SIDS Non-SIDS Crude OR (95% CI) Yes 3 755 OR= 4.67 (1.67, 16.29) No 66 85518 Total 69 86273 a Subjects with missing data were excluded

PAGE 56

49 Table 8 Maternal uterine bleeding during pregnancy and the risk of SIDS Uterine Bleeding SIDS Non-SIDS Crude OR (95% CI) Yes 0 144 OR= 0.00 No 69 84129 Totala 69 86273 a Subjects with missing data were excluded Table 9 Maternal placenta previa during pregnancy and the risk of SIDS Placenta Previa SIDS N on-SIDS Crude OR (95% CI) Yes 0 168 OR= 0.00 No 69 86105 Totala 69 86273 a Subjects with missing data were excluded Table 10 Maternal abruptio placentae during pregnancy and the risk of SIDS Abruptio Placentae SIDS N on-SIDS Crude OR (95% CI) Yes 1 254 OR= 5.04 (0.69, 35.62) No 68 86019 Totala 69 86273 a Subjects with missing data were excluded

PAGE 57

50 Table 11 Maternal tobacco use during pregnancy and the risk of SIDS Tobacco Use SIDS Non-SIDS Crude OR (95% CI) Yes 18 8705 OR= 3.13 (1.83, 5.36) No 51 77318 Totala 69 86273 a Subjects with missing data were excluded Table 12 Late Initiation into prenatal care and the risk of SIDS Month of Initiation SIDS Non-SIDS Crude OR (95% CI) 5-9 or none 13 6879 OR= 2.70 (1.39, 4.66) 1-4 54 77235 Totala 69 86273 a Subjects with missing data were excluded Table 13 Lack of prenatal care during pr egnancy and the risk of SIDS Month of Initiation SIDS Non-SIDS Crude OR (95% CI) None 3 890 OR= 4.37 (1.38, 13.89) 1-9 64 83224 Totala 67 86273 a Subjects with missing data were excluded

PAGE 58

51 Table 14 Pregnancy Complication and the risk of SIDS while controlling for smoking Smoking Non-Smoking PG COMP SIDS Non-SIDS PG COMP SIDS Non-SIDS Yes 2 577 Yes 8 5662 No 16 8128 No 43 71656 Totala 18 8723 Total 51 77318 CRUDE OR= 1.76 CRUDE OR=2.35 95% CI (.41, 7.62) 95% CI (1.11, 5.00) ADJUSTED OR=2.2 95% CI (1.13, 4.31) a Subjects with missing data were excluded

PAGE 59

52 Appendices

PAGE 60

53 Appendix A Modified Description of Data Fields The following fields are from the BIRTH CERTIFICATE BLINK De-identified Number for Births A random number between 1000000 and 1999999 If the same number appears more than once, this indicates that the same birth was matched to duplicate prenatal screens. BDOB Date of Birth Month: 01-12 or 99 for Not classifiable Day: 01-31 or 99 for Not classifiable Year: 1998, 1999, or 2000 BCNTYOB County of Birth 26……Duval 39……Hillsborough BSEX Child’s Sex 1.......Male 2.......Female 9.......Not classifiable BCHRACE Child’s Race 1........White 2........Black 3........Indian (See Appendix H) 4........Chinese 5........Japanese 6........Hawaiian 7........Other entries 8........Filipino 9........Not reported 0........Other Asian or Pacific Islander The child’s race is determined by using the mother’s race and father’s race. See Appendix E for details. BTOBUSE Tobacco Use 1.......Yes 2.......No 9.......Not classifiable BTOBNUM Number of Cigarettes Smoked 00......None 01-97......As shown 98......98 or more 99......Not classifiable

PAGE 61

54 Appendix A (Continued) BMEDFAC Medical History Factors for Pregnancy Up to seventeen 2-digit codes are recorded in this field. 00....None 10....Eclampsia 01....Anemia (Hct. < 30/Hgb < 10) 11....Incompetent cervix 02....Cardiac disease 12....Previous infant 4000+ grams 03....Acute or chronic lung disease 13....Previous preterm or smallfor04....Diabetes gestational-age infant 05....Genital herpes 14....Renal disease 06....Hydramnios/Oligohydramnios 15....Rh sensitization 07....Hemoglobinopathy 16....Uterine bleeding 08....Hypertension, chronic 17....Other specified 09....Hypertension, pregnancy-assoc 99....Not classifiable BCOMPFAC Complications of Labor & Delivery Up to sixteen 2-digit codes are recorded in this field. 00....None 09....Prolonged labor (> 20 hrs) 01....Febrile (> 100 F or 38 C) 10....Dysfunctional labor 02....Meconium, moderate/heavy 11....Breech/Malpresentation 03....Premature rupture of membranes 12....Cephalopelvic disproportion (> 12 hrs) 13....Cord prolapse 04....Abruptio placenta 14....Anesthetic complications 05....Placenta previa 15....Fetal distress 06....Other excessive bleeding 16....Other specified 07....Seizures during labor 99....Not classifiable 08....Precipitous labor (< 3 hrs) BMOPNBGN Month of Pregnancy Prenatal Care Began 0........None 1-8........First through eighth month 9........Ninth month or later U........Not classifiable BNOPNVIS Number of Prenatal Visits 00........None 01-48........As shown 49........49 or more 99........Not classifiable The following fields are from the DEATH CERTIFICATE DDOD Decedent’s Date of Death Month: 01-12........Jan-Dec 99........Not classifiable Day: 01-31........Day of month 99........Not classifiable Year: 1998 or 1999 DCAUSDTH Underlying Cause of Death Four-digit cause of death code assigned in conjunction with the International Classification of Diseases, Ninth Revision (ICD-9), World Health Organization.

PAGE 62

55 Appendix B University of South Florida IRB Exemption Certification MEMO: Patricia Myers, MSPH 700 Dockview Way, #1126 Tampa, FL 33602 FROM: Institutional Review Board / KK:cas SUBJECT: Exemption Certification for Protocol No. 101139 DATE: May 22, 2003 On February 4, 2003, it was determined that your project entitled, "The Impact of Adverse Prenatal Events on Sudden Infant Death" met federal crite ria which exempts it from the regulations specified in the Common Rule. On May 14, 2003 you requested the following change(s): A change in study title from "The Impact of Adverse Prenatal Events on Sudden Infant Death" to "The Association of Maternal Pregnancy Complications and Sudden Infant Death Syndrome" These changes have been noted in the file and do not impact the eligibility for exemption. The study continues to have Exempt Certification. Please re member that any grants co nnected to this project must be submitted to the Institutional Review Board for review. Because the study has been certifie d as exempt, you will not be requir ed to complete continuation or final review reports. However, it is your responsibility to notify the IRB prior to making any changes to the study. Please note that changes made to an exempt protocol may disqualify it from exempt status and may require an expedited or full review. If you have any questions, please contact the Division of Research Compliance at (813) 974-5638 Office of Research, Division of Research Compliance University of South Florida • 12901 Bruce B. Downs BlvdMDC035 • Tampa, Florida 33612-4799 (813) 974-5638• FA X (813) 974-5618 The University of South Flo rida is an Affirmative Action/Equal Access/Equal Opportunity Institution