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 002020938
007 cr mnu|||uuuuu
008 090723s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002502
Bergman, Elizabeth J.
Service utilization among bereaved spouses and family caregivers
h [electronic resource] /
by Elizabeth J. Bergman.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 99 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: For family caregivers, response to the death of the care recipient is marked by a high degree of variability. In recognition of this variability, a range of services and interventions is available to assist individuals in the adjustment to bereavement. The present dissertation, consisting of three related studies, was conducted to examine the utilization of bereavement services by family caregivers. The first study examined the role of psychological distress in the utilization of bereavement services by spousal caregivers in the Changing Lives of Older Couples (CLOC) study. The second study examined bereavement service utilization among dementia caregiver participants in the Resources for Enhancing Alzheimer's Caregiver Health (REACH) study. Both employed Andersen's Behavioral Model of Health Services Use. The third study examined bereavement service utilization, barriers, and preferences among bereaved spousal caregivers of patients of three hospices in Tampa Bay. Taken together, results of the current studies point to the importance of family physicians and members of the clergy in the provision of services to bereaved family caregivers and to the prominent role of bereavement outcomes (e.g., depressive symptoms, grief) as need factors in the utilization of bereavement services.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: William E. Haley, Ph.D.
t USF Electronic Theses and Dissertations.
Service Utilization among Bereaved Spouses and Family Caregivers by Elizabeth J. Bergman A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy School of Aging Studies College of Arts and Sciences University of South Florida Major Professor: William E. Haley, Ph.D. Tamara A. Baker, Ph.D. Yuri Jang, Ph.D. Victor Molinari, Ph.D. Brent J. Small, Ph.D. Date of Approval: July 11, 2008 Keywords: bereavement, caregiving, depres sive symptoms, anxi ety, grief, social resources Copyright 2008, Elizabeth J. Bergman
Dedication This dissertation is dedicated to the people in my life who encouraged me to follow my passion and who supported and encouraged me along the way: To my parents, Walter and Carol Johnson, remarkable people whose values and example I will always hold up as ideals to be emulated. Thank you for giving me a solid foundation and a love of learni ng and for always believing in me, even when I didnÂ’t believe in myself. To my brother, Pete Johnson, who teaches me through his example that gentle tenacity is the best way to effect change and that I can do whatever I set my mind to. To my friends and family, for your suppor t, mentorship, understanding, and love. To Ed, my husband and biggest champion, for your love and unwav ering support, your consistently keen insight, and advice that is yet to be off the mark. Thank you for listening to me, even when I donÂ’t make se nse, for remaining flexible, and for the countless sacrifices you have made so I could follow my dreams. It will be my privilege to return the favor.
Acknowledgments I would like to express my gr atitude to the people who have helped make this work possible: Dr. William E. Haley: Thank you for your support and mentorship over the past five years and for your consistently high expecta tions. I am grateful to have had such an excellent mentor as a role model and I can only hope to be as influential in the lives and professional development of my students as you have been in mine. Dr. Brent J. Small: Thank you for your support, guidance, friendship, and most of all your patience. Drs. Tamara Baker, Yuri Jang, and Vi ctor Molinari: Thank you for your invaluable feedback and all of your contributions as members of my committee. Dr. Cathy McEvoy: Thank you for lending your ear and for helping me to understand the ways of the academic world. The USF Center for Hospice, Palliativ e Care, and End-of-Lif e Studies: Thank you for providing a setting in which applied end-of -life research is so highly valued and for the financial and research support provided thr ough your graduate assistantship program.
The USF School of Aging Studies faculty a nd staff: Ever since my first days in the MasterÂ’s program in 1996 the faculty a nd staff in the School of Aging Studies have welcomed me and been an amazing source of support Â– personally, professionally, and financially. Thank you to Ross Andel, Bill Haley, Yuri Jang, Mary Kaplan, Cathy McEvoy, Wiley Mangum, Larry Polivka, Sandy Reynolds, Sue Saxon, Hal Sheppard, John Skinner, and Brent Small for being teachers of the highest caliber. Thank you to Ga il Smith, Amy Woodbury, Pam Stobaugh, Rosa Diaz, Lydia Hentschel, Nancy Hallfor d, and Ginny Chaplin for all of your practical support and encouragement. Dr. John Cochran: Thank you for chairing my dissertation defense.
i Table of Contents List of Tables................................................................................................................. ....iii Abstract....................................................................................................................... .......iv Chapter One: Introduction...................................................................................................1 Chapter Two.................................................................................................................... .....8 Abstract....................................................................................................................9 Introduction............................................................................................................10 Methods..................................................................................................................14 Participants.................................................................................................15 Measures....................................................................................................15 Predisposing Characteristics..........................................................15 Enabling Resources........................................................................15 Internal Control..................................................................15 Religious Involvement.......................................................15 Emotional Support.............................................................16 Instrumental Support..........................................................16 Need Factors..................................................................................16 Depressive Symptoms........................................................16 Anxiety...............................................................................17 Grief...................................................................................17 Bereavement Service Utilization...................................................17 Statistical Analysis.....................................................................................17 Results ....................................................................................................................19 Sociodemographics & Descriptive Analyses.............................................19 Univariate Analyses...................................................................................21 Multivariate Analyses................................................................................21 Discussion..............................................................................................................25 Chapter Three ................................................................................................................. ...29 Abstract..................................................................................................................30 Introduction............................................................................................................31 Methods..................................................................................................................34 Participants.................................................................................................35 Measures....................................................................................................36 Predisposing Characteristics..........................................................36 Enabling Resources........................................................................36 Need Factors..................................................................................37
ii Bereavement Service Utilization...................................................37 Statistical Analysis.....................................................................................38 Results....................................................................................................................39 Discussion..............................................................................................................45 Chapter Four................................................................................................................... ...50 Abstract..................................................................................................................51 Introduction............................................................................................................53 Methods..................................................................................................................56 Procedure...................................................................................................56 Participants.................................................................................................57 Measures....................................................................................................57 Demographics & Sample Characteristics......................................57 Depressive Symptoms....................................................................57 Social Network...............................................................................57 Bereavement Service Utilization & Barriers.................................58 Bereavement Service Preferences..................................................58 Statistical Analysis.....................................................................................59 Results....................................................................................................................59 Sample Characteristics...............................................................................59 Bereavement Service Utilization...............................................................60 Barriers to Service Use..............................................................................61 Bereavement Service Preferences..............................................................62 Impact of Depressive Symptoms...............................................................64 Impact of Social Network..........................................................................66 Discussion..............................................................................................................67 Chapter Five................................................................................................................... ....72 Conclusions............................................................................................................72 Limitations.............................................................................................................75 Future Directions...................................................................................................77 References..................................................................................................................... .....79 Appendices..................................................................................................................... ....91 Appendix A: Curriculum Vitae..............................................................................92 About the Author...................................................................................................End Page
iii List of Tables Table 2.1 Sample Background Ch aracteristics and Descriptives....................................20 Table 2.2 Correlation Matr ix for Study Variables...........................................................22 Table 2.3 Cox Regression Model of Bereav ement Service Utilization (Any Service)...23 Table 2.4 Cox Regression Model of Bereavement Service Utilization (Family Doctor)............................................................................................... .24 Table 2.5 Cox Regression Model of Ber eavement Service Utilization (Clergy)............25 Table 3.1 Sample Background Ch aracteristics and Descriptives....................................40 Table 3.2 Cox Proportional Hazard s Model of Support Group Use................................41 Table 3.3 Cox Proportional Haza rds Model of Counseling Use.....................................42 Table 3.4 Cox Proportional Hazards M odel of Psychotropic Medication Use................43 Table 3.5 Cox Proportional Hazards Mode l of Overall Bereavement Service Utilization................................................................................................... ......44 Table 4.1 Sample Characteristics.....................................................................................60 Table 4.2 Bereavement Service Utilization Rates...........................................................61 Table 4.3 Perceived Barriers to Bereavement Service Utilization..................................62 Table 4.4 Mean Preferences Regardi ng Content of Bereavement Services....................63 Table 4.5 Mean Preferences Regardi ng Structure of Bereavement Services..................65 Table 4.6 Differences in Content Prefer ences Based on Level of Social Network.........67
iv Service Utilization among Bereaved Spouses and Family Caregivers Elizabeth J. Bergman ABSTRACT For family caregivers, response to the d eath of the care recipient is marked by a high degree of variability. In recognition of this variability, a ra nge of services and interventions is available to assist individuals in the adjustment to bereavement. The present dissertation, consisting of three re lated studies, was conducted to examine the utilization of bereavement services by family caregivers. The first study examined the role of psychological distress in the utilization of bereavement services by spousal caregivers in the Changing Lives of Older Couples (CLOC) study. The second study examined bereavement service utilization among dementia caregiver participants in the Re sources for Enhancing AlzheimerÂ’s Caregiver Health (REACH) study. Both employed Ande rsenÂ’s Behavioral Model of Health Services Use. The third study examined be reavement service utilization, barriers, and preferences among bereaved spousal caregiver s of patients of three hospices in Tampa Bay. Taken together, results of the current st udies point to the importance of family physicians and members of the clergy in the provision of services to bereaved family caregivers and to the prominent role of bereavement outcomes (e.g., depressive symptoms, grief) as need factors in th e utilization of ber eavement services.
1 Chapter One: Introduction Accompanying growth in the size of the older adult population (U.S. Census Bureau, 2005) is an increase in the number of older adults ex periencing physical limitations, frailty, and chronic and terminal illness. These conditions frequently necessitate reliance upon spouses family members, and other sources of informal care. The long duration and intense level of care fr equently required has led researchers to describe caregiving as a Â“careerÂ” (Pea rlin & Aneshensel, 1994) accompanied by transitions such as institutionalization and bereavement (Aneshensel, Pearlin, Mullan, Zarit, & Whitlatch, 1995). The U.S. Admini stration on Aging estimates that more than 22.4 million Americans act as informal caregivers to an older or disabled person and this number is projected to rise dramatically in the coming decades (U.S. Department of Health and Human Services, 2003). For family caregivers, response to the d eath of the care recipient is marked by a high degree of variability (Aneshensel, Botticello, & Yamamoto-Mitani, 2004; Bonanno et al., 2002; Bonanno, Moskowitz, Papa, & Folkman, 2005; Christakis & Iwashyna, 2003; Goodkin et al., 2001; Sc hulz, Mendelsohn, et al., 2003). Some respond to the transition with resilience, while for others it is a disruptive and painful life experience marked by years of intense physical, psychol ogical, and existential suffering (Bonanno et al.). Most exhibit a bereavement response that falls somewhere between these extremes (Bonanno & Kaltman, 2001). Bereavement outco mes reported in the literature include
2 mental health decrements (e.g., depression, an xiety, complicated grief) (Bonanno et al.; Schulz et al., 2001), physical health decr ements (Kaprio, Koskenvuo, & Rita, 1987), and heightened risk of mortality (Christakis & Iwashyna; Goldman, Korenman, & Weinstein, 1995; Goodkin et al.). Given the stressful nature of family careg iving and bereavement, a wide array of interventions and services has been developed to assist indi viduals with the emotional, practical, and social adjustments that accompa ny the death of the care recipient. These services are provided by a range of professionals and organi zations. Examples include grief therapy, cognitive beha vioral therapy, support groups pastoral counseling, widowto-widow programs, psychoeducatio nal programs, and pharmacotherapy. While a detailed discussion is beyond the scope of this dissertation, recent research has produced some interesting and important results regarding the efficacy of bereavement interventions (e.g., Allumba ugh & Hoyt, 1999; Kato & Mann, 1999; Schut, Stroebe, van den Bout, & Terheggen, 2001; Schut & Stroebe, 2005). The general conclusion that can be drawn from this wo rk is that the efficacy of bereavement interventions increases in direct proportion to the level of complication or distress in bereavement (Schut et al., 2001). Howeve r, we do not know whether the individuals most in need of interventi on are receiving services in bereavement nor is there any widespread, systematic effort in place to target those in the greatest need. Some of the most commonly identified me ntal health outcomes associated with bereavement and experienced at higher ra tes than non-bereaved controls include depression (Bodnar & Kiecolt-Glaser, 1994; McHorney & Morr, 1988; Schulz, Mendelsohn, et al., 2003; Zisook, Paulus, Shuchter, & Judd, 1997), anxiety (Beem,
3 Maes, Cleiren, Schut, & Garssen, 2000; By rne & Raphael, 1997; Zisook & Shuchter, 1985), and complicated grief (B arry, Kasl, & Prigerson, 200 2; Prigerson et al., 1995; Prigerson et al., 1997 ; Zhang, El-Jawahri, & Prigerson, 2006). Some researchers have found an association between elevated distress and service use. For example, Cherlin and colleagues (2007) reported that family caregiv ers who used bereavement services were more likely to have met criteria for Majo r Depressive Disorder at the time of study enrollment than were those who did not. Caserta and Lund (1992) found an association between service utilization and elevated depre ssion in the first three months after spousal bereavement. Similarly, Levy and Der by (1992) found an association between bereavement support group use and higher levels of anxiety and depr essive symptoms at 6 months after the death of a spouse. Ho wever, Prigerson and colleagues (1997) found that heightened distress in be reavement, in the form of se vere depression and complicated grief, was associated with a decreased like lihood of physical and me ntal health service use in bereavement. In light of these contradictory findings, there exists a need to more closely examine whether those with th e greatest opportunity for benefit are indeed aware of and utilizing bereavement interventions available in the community. In learning about factors associated with the utilization of bereavement services, we can better tailor interventions and services to the needs and preferences of those experiencing or belonging to groups at high risk for distress. Further, we can work to achieve more equitable access to services for individuals and groups w ho are currently underserved. One category of factors which may be associated with bereavement service utilization may be viewed collectively as resources, including social resources,
4 religiosity, perceived control, and participation in a caregiv er intervention prior to the death of the care recipient. These resour ces may serve to strengthen ties to the community or formal service delivery sy stem (McIntosh, Silver, & Wortman, 1993; Stroebe & Stroebe, 1987; Coulton & Fros t, 1982; Pearlin, Menaghan, Lieberman, & Mullan, 1981), change self-perceptions of h ealth and well-being and resulting motivation to seek care (Berkman, 1979), or improve acc ess to resources necessary for participation (e.g., transportation) (Toseland, McCallion, Gerber, & Banks, 2002). Sociodemographic characteristics, incl uding age, gender, race/ethnicity, and education, may also be associated with bereav ement service utilization. For example, the subset of older individuals who experience mental health decrements in response to bereavement may be less likely to utilize inte rvention due to an age bias in referral for psychological services (Ginsberg & Goldstein, 1 974) or to stigma associated with mental health service use (G allagher-Thompson et al., 2003). In general, it is believed that women are more likely to recognize a ment al health problem (Kessler, Brown, & Broman, 1981) and utilize mental health se rvices (Coulton & Frost, 1982), suggesting that female gender may predict be reavement service utilization. While the role of education in predicti ng bereavement service utilization is not known, health and human service use research suggests that higher ed ucational level is a predictor of service use among the elderly (Mutran & Ferraro, 1988; Snider, 1980), dementia caregivers (Toseland et al., 2002) and Black caregivers (Williams & DilworthAnderson, 2002). It may be the case that lower educational level diminishes the likelihood of service use through its impact on knowledge of the service delivery system
5 and community resources (Snider; Wallace, Levy-Storms, Kingston, & Andersen, 1998) or on socioeconomic status and income (Wallace et al., 1998). Ethnic and racial minorities are drasti cally underrepresented among those seeking psychological and supportive services (A merican Psychological Association, 2002; Coulton & Frost, 1982; Mui & Burnette, 1994). In response to the historical pattern of denial of access to formal sources of s upport, many racial/ethnic minorities rely on extensive informal support netw orks, including extended family and church, in coping with the demands of providing care for disa bled relatives (Dilworth-Anderson, Williams, & Gibson, 2002), a practice which may extend into bereavement. Research suggests that religious leaders in the Black community play an important role both in delivering and referring to services with the potential to im prove the mental health and quality of life of bereaved individuals (Kart, 1991; Williams & Dilworth-Anderson, 2002). The Andersen model of hea lth services utilization, fi rst proposed by Andersen (1968) and subsequently revised and expa nded (Andersen & Newman, 1973; Aday & Andersen, 1974) to aid in the id entification and description of determinants of health care use, is a major perspective used in health se rvice utilization research. It has been used extensively to examine health care utilization and equity in the health care system (e.g., Andersen, Kravits, & Anderson, 1975), to ma ke cross-national comparisons in health service utilization (Andersen, 1976) and in many studies of the use of nursing homes, inhome nursing, formal and informal helpers, social services, adult day care, and dental services by older adults (e.g., Arling, 1985; Conrad, Hughes, & Wang, 1992; Coulton & Frost, 1982; Counte & Glandon, 1991; Evashw ick, Rowe, Diehr, & Branch, 1984; Eve, 1988; Freedman, 1993; Miller & McFall, 1991; Mutran & Ferraro, 1988; Rabiner, 1992;
6 Rosner, Namazi, & Wykle, 1988, Wan, 1989). T hus, the Andersen model is a useful conceptual tool in the investigation of the determinants of bereavement intervention utilization. The Andersen model encompasses three ca tegories of individual determinants of health service utilization, in cluding predisposing characterist ics, enabling resources, and need factors (Aday & Andersen, 1974; A ndersen, 1968; Andersen & Newman, 1973). Need factors include the level of illness or the need for health care, both as perceived by the individual and as evaluated by health pe rsonnel. Enabling resources are those which allow for the use of services, and include personal resources (e.g., insurance coverage), family resources (e.g., social support), and community resources (e.g., geographic characteristics of the community). Predispos ing characteristics are those which, present prior to the onset of need (A day & Andersen), would deter or predispose the individual from using health care services Examples include demographi c characteristics (e.g., age, gender), aspects of social st ructure (e.g., race/eth nicity, education), and health beliefs (e.g., attitudes, perceptions, and knowledge of disease and medical care). Need factors are believed to be the most immediate predictor of health services use (Aday & Andersen, 1974; Andersen, 1995; Coulton & Frost, 1982; Kart, 1991; Wallace et al., 1998), accounting for the most variance in the use of physician visits, mental health services, personal care, a nd recreational services (Coulton & Frost), informal and formal home care and commun ity services (Kart), and in-home and community-based services (Mui & Burnette, 1994). However, the roles of predisposing characteristics and enabling resources may become more prominent relative to need factors when greater discretion can be exerci sed by the potential consumer in the decision
7 whether or not to utilize health services a nd in the type of serv ice pursued (Aday & Awe, 1997), as may be the case with bereavement services. The present doctoral disserta tion examines bereavement in the context of family caregiving, with specif ic emphasis on service utilization and preferences among bereaved caregivers. A series of thr ee studies was conducted. In the first two, bereavement service utilization rates and the type s of services used were examined using two publicly available datasets. The first is the Cha nging Lives of Older Couples (CLOC) study, a longitudinal, multi-wave, prospective study of spousal bereavement. The second is the Resources for Enhancing AlzheimerÂ’s Caregiver Health (REACH) study, a multi-site study evaluating the longitudinal impact of family caregiver inte rventions. In both studies, guided conceptually a nd analytically by AndersenÂ’s behavioral model of health services use, predictors of bereavement serv ice utilization were examined. In the third study, bereavement service utilization rates, patt erns, preferences, barriers to use, and the roles of depressive symptoms and social netw ork were examined in a sample of bereaved caregivers whose spouses died while under hospice care in the Tampa Bay area.
8 Chapter Two: The Role of Grief and Depressive Symptoms in the Utilization of Bereavement Services
9 Abstract This study examined the role of psychol ogical distress in the utilization of bereavement services by caregivers in th e Changing Lives of Older Couples study. Approximately 57% of caregivers utilized services, most commonly provided by family physicians and clergy members. Elevated grie f was associated with higher likelihood of using any service and of talk ing with a physician. Higher depressive symptoms were associated with decreased likelihood of talk ing with a physician. Findings suggest the need for outreach and education with family physicians and clergy. Individuals experiencing elevated grief may be util izing bereavement services, while those experiencing elevated depressi ve symptoms may be underserved.
10 Introduction Recent demographic trends and projections point to dramatic growth in the size and proportion of the older adult population in the United States (U.S. Census Bureau, 2005). Coupled with the finding that approximat ely 80 percent of care for older adults is provided by family members (National Al liance for Caregiving & AARP, 1997), the critical role of family caregivers is in increasing demand. At the same time, technological and medical advances are dela ying death, prolonging th e end-of-life stage frequently characterized by high levels of im pairment and a need for intense levels of care. Caregiving has been described in the lit erature as a Â“careerÂ” (Aneshensel et al., 1995; Pearlin & Aneshensel, 1994) with the po tential to be a long-term experience (Hurley & Volicer, 2002; Schulz, Mendelsohn, et al., 2003) punctuat ed by transitions, such as institutionalization and bereavement (Burton, Zdaniuk, Schulz, Jackson, & Hirsch, 2003; Gaugler, Anderson, Zarit, & Pear lin, 2004). The transition to bereavement, which occurs when the care recipient dies a nd the caregiving role ceases, is marked by great variability (Bonanno & Kaltman, 1999). Fo r some, bereavement results in years of intense physical, psychological, and existen tial suffering, while ot hers respond with remarkable resilience (Bonanno et al., 2002). Most exhibit a bereavement response that falls somewhere between these extremes. Bereavement outcomes of family caregivers reported in the literature include grief, depr ession, anxiety, diminished quality of life, increased health care utilizat ion, and increased risk of phys ical illness and mortality (Christakis & Iwashyna, 2003; Goodkin et al., 2001; Stroebe, Schut, & Stroebe, 2007).
11 An array of interventions and services has been devel oped to assist individuals with the emotional, practical, and social pr oblems that can accompany the loss of a loved one. These services are provided by a range of professionals a nd organizations and examples include grief therapy, cognitive behavioral therapy, support groups, pastoral counseling, widow-to-widow programs, psychoeducational programs, and pharmacotherapy. Reviews of the efficacy of these interventions have been mixed, with some evidence to suggest that highly-struct ured interventions ta rgeting those who are most vulnerable to or currently experien cing negative bereavement outcomes have the greatest impact (Allumbaugh & Hoyt, 1999; Forte, Hill, Pazder, & Feudtner, 2004; Larson & Hoyt, 2007b; Schut & Stroeb e, 2005; Schut et al., 2001). In previous studies, bereavement servi ce utilization rates have ranged from approximately 10% to 35% (Billings & Ko lton, 1999; Caserta & Lund, 1992; Cherlin et al., 2007; Levy & Derby, 1992; Provini, Everett, & Pfeffer, 2000; Sc hulz, Mendelsohn, et al, 2003). Cherlin and colleagues identified factors associated with service use by bereaved hospice caregivers, including spousal relationship, younger age, more complicated caregiving context, presence of major depressive disorder, and adequate instrumental support. They found, however, th at even among participants identified as experiencing clinically signifi cant levels of depressive symp toms, less than half utilized bereavement services despite the fact that all had received inform ation about available hospice bereavement services. These findings suggest that many spous al caregivers at risk for or experiencing negative bereavement outcomes are unlikely to utilize bereavement services. Additionally, we know of no widespread, systematic effort in place to target those in greatest need. Theref ore, questions remain as to whether those
12 individuals most in need of intervention are seeking out or receiving services in bereavement. The purpose of this study was to examine predictors of bereavement service utilization among bereaved spousal caregivers. We were conceptually and analytically guided by AndersenÂ’s health services u tilization model (Ada y & Andersen, 1974; Andersen, 1968; Andersen & Newman, 1973) to examine the contributions of predisposing characteristics (i.e., sociodemograp hics), enabling factors (i.e., resources), and need factors (i.e., bereavement outcomes) Need variables ar e believed to be the most immediate predictor of health servic es utilization (Aday & Andersen; Coulton & Frost, 1982; Kart, 1991; Wallace et al., 1998). However, predis posing characteristics and enabling resources may take on greater im portance when more discretion may be exercised in the decision to utilize health se rvices and in the type of service pursued (Aday & Awe, 1997), as may be the case with bereavement services. Age is hypothesized to serve as a pred isposing characteristi c, with older age associated with greater utilization of serv ices provided by a family physician and younger age associated with the utiliz ation of other bereavement serv ices, such as counseling and support groups. In a survey of older and younger adults (Robb, Haley, Becker, Polivka, & Chwa, 2003), older adults reported that they were significantly less likely to utilize mental health services for depression, anxiet y, bereavement, or stress. They were also less likely to have had prio r experience with mental hea lth providers and were less confident about their knowledge of mental health care and tr eatment. Older adults were more likely to view their primary physician as the appropriate source of referral to a mental health professional.
13 Female gender is hypothesized to serve as a predisposing characteristic predictive of greater bereavement service utilization. Studies indicate that, in general, women are more likely to recognize mental health probl ems (Kessler et al., 1981) and to utilize mental health services (Coulton & Frost, 1982). Racial or ethnic minority status is hypothesized to serve as a predisposing characteristic in greater u tilization of services provide d by a member of the clergy, whereas non-minority status is expected to predict for greater us e of other types of bereavement services. Ethnic and racial minorities are drastically underrepresented among those seeking psychological and s upportive services (A PA, 2002; Coulton & Frost, 1982; Mui & Burnette, 1994). In additi on, examinations of racial differences in sources of caregiving support and assistance re veal that White individuals demonstrate the highest levels of formal home health care and nursing home use, whereas Black individuals demonstrate the hi ghest levels of reliance on informal helpers (DilworthAnderson et al., 2002; Mui & Burnette). In a study of Black older adults, Kart (1991) found that assistance overwhelmingly came fr om informal, unpaid sources and that participants looked to respect ed leaders to act as brokers between community members and the formal social service system. This suggests that some individuals in the Black community, particularly religious leaders, may play an important role in delivering and referring to services with the potential to im prove the mental health and quality of life of bereaved individuals. Education is expected to serve as a pred isposing characteristic with higher levels of education leading to greater use of servic es. Health and human service use research suggests that educational leve l is a predictor of service use among the elderly (Mutran &
14 Ferraro, 1988; Snider, 1980), dementia care givers (Toseland et al., 2002), and Black caregivers (Williams & Dilworth-Anderson, 2002) It has been hypothesized that lower educational level serves to decrease servic e use through its impact on knowledge of the service delivery system and available commun ity resources (Snider; Wallace et al., 1998) and on socioeconomic status a nd income (Wallace et al.). Internal control, social resources, an d religious involvement are hypothesized to serve as enabling resources, or resources whic h allow for service use, with higher levels of each leading to greater bereavement se rvice utilization. This may occur through strengthening ties to the co mmunity (Coulton & Frost, 1982; Pearlin et al., 1981), changing self-perceptions of health and well-be ing and resulting motivation to seek care (Berkman, 1979), or improving access to re sources necessary for participation (e.g., transportation) (Toseland et al., 2002). Bereavement outcomes including depressi ve symptoms, anxiety, and grief are expected to serve as need factors, with hi gher levels of distress predictive of greater bereavement service utilization. Methods These analyses used data from the Changing Lives of Older Couples (CLOC) study, which was conducted to examine spousal bereavement as a life stressor and to elucidate the mechanisms through which bereav ement affects mental and physical health (Bonanno, Wortman, & Nesse, 2004). The CLOC study is a longitudinal, multi-wave, prospective study of spousal bereavement conducted in the De troit Standardized Metropolitan Statistical Area by the Institute fo r Social Research at the University of Michigan and funded by the National Instit ute on Aging. Baseline interviews were
15 conducted in 1987 and 1988 with both members of each couple. Follow-up interviews were then conducted with survivors in three waves at 6, 18, and 48 m onths after the death of their spouse, concluding in 1994. Participants CLOC participants were community-dwe lling, English-speaki ng, and capable of participating in a 2-hour f ace-to-face interview and they were members of a married couple in which the husband was at least 65 years of age at the time of enrollment. During the study, 319 CLOC participants lo st a spouse and 86% of those (n=276) participated in at least on e follow-up interview. The current analyses include 126 participants who provided care to their spouse in the last six months of life and who completed at least one post-loss wave of follow-up. Measures Predisposing Characteristics The sociodemographic characte ristics of interest include age, gender, race, and education. Due to the small number of minor ities in the sample, race was categorized as White or non-White. Enabling Resources Internal control. The internal control index ( = .71) contained the following 4 items: 1) My life is determined by my own ac tions, 2) When I make plans, I am almost certain to make them work, 3) When I get what I want, itÂ’s usually because I worked hard for it, and 4) I am usually able to protect my own interests. Each item was scored on a 4point Likert scale ranging from Â“not true at allÂ” to Â“very tr ue.Â” Higher scores indicate higher levels of internal control.
16 Religious involvement. The religious involvement index ( = .83) is a composite of 4 items including: 1) In general, how impor tant are religious and spiritual beliefs in your day-to-day life, 2) How often do you us ually attend religious services, 3) When you have problems or difficulties in your fam ily, work, or personal life, how often do you seek spiritual comfort and support, and 4) When you have decisions to make in your everyday life, how often do ask what God woul d want you to do? Hi gher scores indicate higher levels of religious involvement. Emotional support. This variable is a composite index ( = .71) of two items measuring the degree to which friends and re latives made the respondent feel loved and cared for and how willing they were to listen to the respondent talk about their worries or problems. Higher scores indicate highe r levels of positive emotional support. Instrumental support. This variable is a composite index ( = .68) of three items measuring the degree to which the respondent fe lt they could rely on friends or relatives for help if they needed extra help with ge neral housework or home maintenance, needed extra money, or were ill. Higher scores i ndicate higher levels of instrumental support. Need Factors Depressive Symptoms. An 11-item subscale of the CES-D (Radloff, 1977) ( = .84) was used to measure depressive symp toms with responses ranging on a 3-point Likert scale from Â“hardly ever Â” to Â“most of the time.Â” C onverted into z-scores with a mean of 0 and a standard deviation of 1, higher scores indicate higher levels of depressive symptomotology.
17 Anxiety. Anxiety was measured using a 10-item index ( = .87) with responses ranging on a 5-point Likert scale fr om Â“not at allÂ” to Â“extremel y.Â” This variable was also converted to centered z-scores and hi gher scores indicate higher anxiety. Grief. Overall grief was measured using a 19-item scale ( = .88) developed by the CLOC researchers. The s cale utilized a 5-point Likert scale ranging from 0 to 4 with higher scores indicating highe r levels of grief. Bereavement Service Utilization Bereavement service utilization was dete rmined based on participantsÂ’ answers to the following questions: To help handle f eelings of grief, loneliness or missing your husband/wife, have youÂ… a. gone to a self-help group or support gr oup for people who have lost a spouse? b. seen a minister, priest or other clergy for help? c. talked to your family doctor? d. gone to a mental health professional for help Â– someone like a social worker or psychologist? A composite bereavement service utilization variable was also created, representing the use of any one or more of the above bereavement services. Statistical Analysis Correlation analyses were performed in order to examine the bivariate associations among study variable s. Predictors of bereav ement service utilization were examined using Cox proportional hazards models, chosen for their ability to accommodate censored cases, or those for which the event of interest did not occur (i.e., individuals did not utilize be reavement interventions) or for whom occurrence status was
18 not known due to discontinuation of the st udy (Cantor, 2003; Tabachnick & Fidell, 2001). Predictors were categorized as predispo sing characteristics, enabling resources, or need factors. Predisposing characteristics included baseline age, gender, race, and education. Enabling resources incl uded internal control, reli giosity, and instrumental and emotional support as reported at the first post-loss follow-up. Need factors included bereavement outcomes, including depressive sy mptoms, anxiety, and grief as reported at the first post-loss follow-up. For those indivi duals lacking 6 month follow-up data, data from the first available follow-up wave we re used for enabling resources and need factors. Time, measured in days, from the date of the care recipientÂ’s death to the date of the post-loss interview at which service use was reported, was used to account for differences in the interval until first service use. For these analyses, use of bereavement service was coded as Â“1Â” and non-use (i.e., Â“sur vivalÂ”) as Â“0.Â” Separate analyses were conducted for the use of any bereavement service, talking with the family doctor, and talking with a minister, prie st, or other member of the clergy. We were unable to individually examine predicto rs of support group attendance a nd use of a mental health professional due to the small number of pa rticipants who utilized these bereavement services. For those who neve r utilized bereavement servi ces, the date of the 48 month post-loss follow-up interview was used as a right-censoring event. For those who dropped out of the study before 48 month follo w-up, the date of the last available postloss follow-up interview was used as the censori ng event. In the model examining use of any service, event occurrence was coded at the date of the first wave at which use of any one of the four services under investigation was reported.
19 In the Cox proportional hazards models, we entered predisposing characteristics, enabling resources, and need factors in se quential blocks in order to examine the importance of each block to the prediction of time to service us e after statistical adjustment for the effects of the preced ing block or blocks of covariates. Results Sociodemographics and Descriptive Analyses The sample consisted primarily of White (84.1%) females (84.9%) with a median of 12 years of education and an average ag e of 69. Table 2.1 shows sociodemographic and background characteristics of the sample. While the sample c onsisted primarily of females, this is similar to other studies of older bereaved spouses (e.g., Schulz, Mendelsohn, et al., 2003) and gene rally representative of the gender distribution of older bereaved spouses. The sample is not, however representative of the population of older bereaved spouses with regard to race and th e generalizability of th ese findings are thus limited. The mean baseline depressive symptom and anxiety scores of all 1,532 CLOC participants were centered with a mean of zero and SD of 1. Thus, depressive symptom and anxiety scores above zero i ndicate that they were highe r than the baseline mean of the broader sample. As displayed in Tabl e 2.1, caregiver depressive symptoms were approximately one quarter of a standard de viation above the tota l group at baseline, peaked by 6 months post-loss at nearly one half of a standard deviation above the baseline mean, and declined thereafter. Care giver anxiety scores were highest prior to the death of the care recipient and fell at each successive wave, ultimately reaching a
20 level below the baseline mean of the broader sample. Grief scores were highest at 6 months post-loss and declined at each successive wave. Table 2.1. Sample Background Char acteristics and Descriptives % Mean SD Range Predisposing Characteristics Age 68.8 7.0 48 -86 Gender (Female) 84.9 Race (White) 84.1 Education (Median in years) 12 3 17 Enabling Resources Internal Control a b -.10 1.0 -2.8 Â– 1.2 Religious Involvement a c .15 .94 -2.4 Â– 1.3 Instrumental Support a c -.05 1.0 -2.6 Â– 1.2 Emotional Support a c .34 1.1 -2.7 Â– 1.4 Need Factors Depression a Baseline .24 1.1 -1.1 Â– 4.2 6 Months .43 1.3 -1.1 Â– 6.1 18 Months .15 1.2 -1.1 Â– 5.2 48 Months -.28 1.1 -1.1 Â– 3.9 Anxiety a Baseline .22 1.3 -.59 Â– 7.0 6 Months .12 1.0 -.59 Â– 4.5 18 Months -.01 .84 -.59 Â– 3.5 48 Months -.09 .92 -.59 Â– 3.5 Grief 6 Months 2.1 .62 1.1 Â– 3.9 18 Months 1.9 .52 1.0 Â– 3.2 48 Months 1.6 .54 1.0 Â– 3.2 Service Utilization (%) Use of Any Service 57.1 Family Doctor 42.1 Clergy 29.4 Support Group 15.1 Mental Health Professional 5.6 a z-scores b Baseline c Wave 1 (6 months post-loss)
21 Seventy two (57.1%) bereaved spouses utilized at least one of the four bereavement services of interest (i.e., fa mily physician; clergy; bereavement support group; mental health professi onal). Fifty three (42.1%) part icipants talked with their family physician about feelings of grief, l oneliness, or missing their spouse and thirty seven (29.4%) participants met with their mini ster, priest, or cler gy member. Nineteen (15.1%) participants a ttended a bereavement self-help or support group. Only seven (5.6%) saw a mental health professional su ch as a social work er or psychologist. Bivariate Analyses The results of the bivariate correlati on analyses are shown in Table 2.2. The results indicate that female participants we re more likely to be younger (r = -.25, p < .01). Participants with higher levels of education were also more likely to be younger (r = -.25, p < .01) and to report higher levels of em otional support (r = .19, p < .05). Non-Whites reported significantly higher levels of relig ious involvement (r = .20, p < .05), but lower levels of instrumental support (r = -.19, p < .05). Participants who reported higher levels of religious involvement had significantly hi gher levels of emotiona l support (r = .25, p < .01). Social resources were inversely correla ted with bereavement outcome, with higher levels of instrumental and emotional support as sociated with lower le vels of depressive symptoms (r = -.31 and -.28 respectively, p < .01), anxiety (r = -.31 and -.32 respectively, p < .01), and grief (r = -.24, p < .05 and r = -.26, p < .01). Multivariate Analysis Complete data were available for one hundred and thirteen bereaved spouses who participated in at least 1 post-loss interview. In the first proportional hazards model, a composite variable indicating time to use of any one or more of the four bereavement
22 Table 2.2. Correlation Matr ix for Study Variables 1 2 3 4 5 6 7 8 9 10 11 Predisposing Characteristics 1. Age -.25** -.25** -.04 .03 -.05 .09 -.09 .06 -.08 -.10 2. Gender .11 .00 -.14 .15 .11 .11 .02 .12 -.02 3. Education .02 -.11 .05 -.04 .19* -.10 -.09 .11 4. Race .12 .20* -.19* .07 -.16 -.01 -.17 Enabling Resources 5. Internal Control -.02 .18 -.01 -.15 -.16 -.17 6. Religious Involvement .05 .25** -.18 -.17 -.12 7. Instrumental Support .25** -.31** -.31** -.24* 8. Emotional Support -.28** -.32** -.26** Need Factors 9. Depression .70*** .74*** 10. Anxiety .53*** 11. Grief *p<.05, **p<.01, ***p<.001
23 services under examination (family doctor, clergy/pastor, suppor t group, mental health professional) served as the dependent variable. Regarding the use of any bereavement service (see Table 2.3), the propor tional hazards model indicated that grief successfully predicted time to service use after ad justing for the other covariates ( 2 (1) = 8.152, p = .004). As shown in Table 2.3, for each one-point difference in grief at the first post-loss follow-up, the probability of using any bereav ement service increased by approximately 92%. The proportional hazards model for talking with the family doctor about feelings of grief, loneliness, or missing oneÂ’s spous e indicated that the need factor block containing depressive symptoms and grief successfully predicted time to service use after adjusting for predisposing characte ristics and enabling resources ( 2 (2) = 14.250, p = .001). As shown in Table 2.4, for each one-point difference in grief, the probability of Table 2.3. Cox Regression Model of Bereav ement Service Utilization (Any Service) 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age .016 1.02 .98 1.06 Gender -.162 .85 .41 1.75 Race -.424 .65 .36 1.19 Education .039 1.04 .95 1.14 Enabling Resources Internal Control -.114 .89 .67 1.19 Religious Involvement .111 1.12 .82 1.53 Instrumental Support -.054 .95 .72 1.24 Emotional Support -.071 .93 .71 1.22 Need Factors a Grief .650 1.92** 1.23 2.99 **p<.01 a Anxiety and depressive symptoms were excluded from the model.
24 talking with the family doctor increased near ly four and a half times. The effect of depressive symptoms was in the opposite direction, however, with each standard deviation difference resulting in a nearly 33% decrease in the probability of talking with the family doctor. Table 2.4. Cox Regression Model of Bereavem ent Service Utilization (Family Doctor) 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age .043 1.04 .99 1.09 Gender -.084 .92 .40 2.10 Race -.543 .58 .29 1.15 Education .025 1.03 .92 1.14 Enabling Resources Internal Control .188 1.21 .86 1.69 Religious Involvement .176 1.19 .83 1.72 Instrumental Support -.185 .83 .61 1.14 Emotional Support -.221 .80 .59 1.08 Need Factors a Depressive Symptoms -.397 .67* .46 .98 Grief 1.522 4.58** 1.93 10.91 *p<.05, **p<.01 a Anxiety was excluded from the model. The proportional hazards model for talki ng with a minister, priest, or other member of the clergy about feelings of grief, loneliness, or missing oneÂ’s spouse indicated that the predisposing characteristics block successfully predicted time to service use ( 2 (4) = 11.451, p = .022) and the probability of talking with a clergy member was 67% less for Whites than for non-Whites (see Table 2.5). The enab ling resources block also predicted time to service use ( 2 (4) = 12.765, p = .012) after adjusting for
25 predisposing characteristics, and with each unit difference in religious involvement the probability of talking with a clergy member increased by more than two times. Table 2.5. Cox Regression Model of Ber eavement Service Utilization (Clergy) 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age -.023 .98 .93 1.03 Gender -.573 .56 .17 1.90 Race -1.116 .33** .16 .67 Education -.024 .98 .84 1.13 Enabling Resources Internal Control -.267 .77 .50 1.16 Religious Involvement .751 2.12** 1.21 3.72 Instrumental Support .092 1.10 .76 1.59 Emotional Support .132 1.14 .75 1.74 Need Factors Depressive Symptoms .328 1.39 .83 2.33 Anxiety -.281 .76 .41 1.38 Grief -.068 .94 .33 2.63 *p<.05, **p<.01 Discussion In the current study, approximately 57% of caregivers utilized bereavement services after the death of th eir spouse. This fairly high proportion may indicate that bereaved spousal caregivers who are experi encing distress recognize their need for and seek out bereavement services consistent with their preferences. However, nearly half of bereaved spousal caregivers did not utilize bereavement se rvices from which they may have benefited. While we found a bereavem ent service utilization rate higher than previous studies (Billings & Kolton, 1999; Caserta & Lund, 1992; Cherlin et al., 2007; Levy & Derby, 1992; Provini et al., 2000; Schulz, Mendelsohn, et al., 2003), this may be
26 accounted for by the fact that the current study focused on spousal caregivers, a group who may have been more closely linked to or accustomed to the use of formal supportive services. It could also be attributed to the liberal definition of bereavement services employed in the current study, including the characterization of talking with a clergy member and the family doctor as use of a bereavement service. The most commonly used services incl uded talking with the family physician (42.1%) and seeing a minister, priest, or othe r clergy for help (29.4%). These findings indicate the need for outreach and educati on with family doctors and members of the clergy, as they are likely to ha ve regular contact with older adults and to be viewed by older adults as important gatek eepers to other mental health services (Robb et al., 2003). In many cases, they may be the only professiona ls with the opportunity to identify an individual who is suffering in bereavement. Ideally, family physicians and members of the clergy should receive information and trai ning regarding risk factors for negative bereavement outcomes, the identification of negative outcomes and distress, and resources available in the community to provi de specialized bereav ement services which are beyond their scope of resources or training. A comprehensive review of the efficacy of bereavement interventions (Schut et al., 2001) and a more recent update to this revi ew (Schut & Stroebe, 2005) concluded that interventions were more effective when targ eted to individuals in high-risk groups or individuals already suffering fr om complicated grief, as opp osed to primary preventive interventions offered routinely to bereaved individuals. The authors advocated for inreaching interventions, or those provided to bereaved individuals who seek out and ask for help, as they are likely to be in the gr eatest need. The curr ent study suggests that
27 individuals experiencing a high level of grie f may indeed perceive a need for and seek out bereavement intervention. However, indi viduals suffering from higher levels of anxiety were not more likely to utilize be reavement services than those exhibiting low anxiety. Furthermore, indivi duals experiencing higher levels of depressive symptoms were actually less likely to talk with thei r family doctor about feelings of grief, loneliness, or missing their spouse. Those e xperiencing elevated leve ls of depression or anxiety in bereavement may well be in a high risk group for complicated bereavement and yet they may not seek out or receive intervention. This is an important consideration in the planning of screening and outreach e fforts with bereaved spousal caregivers. This research has some limitations which should be noted. First, the dataset does not contain information about whether or not the spouse died while under hospice care. The provision of hospice care is likely to impact knowledge of available bereavement services, the lack of which may be a significan t barrier to bereavemen t service utilization as it was in the use of long term care (Bra dley et al., 2002). Fu ture research should address this in an effort to understand the ro les played by knowledge of available services and by hospice use in general. Second, inform ation about the utiliz ation of bereavement services was collected retrospe ctively and was self-reported by participants. This is a common approach to the study of health beha vior and the only option for the collection of data pertaining to utilization of bereavem ent services which are not reimbursable under Medicare, Medicaid, or ot her insurance plans (e.g., bereavement support group, clergy/pastoral counse ling). However, studies indicat e that when compared with information about actual utilization as obtaine d from administrative records, self-reported health services utilizatio n is frequently underestimated (Dunlop, Manheim, Song, &
28 Chang, 2002). Finally, the available data offers little insight into the specifics of the bereavement services of interest. For exam ple, information was not available about the credentials of service providers, the dura tion of the services, or the theoretical foundations of the services under examination. Attention to these and other details in future studies will help us to better understand preferences, motivations, and choices regarding bereavement service utilization as we ll as the specific features of services which lead to positive outcomes for diverse groups of bereaved individuals.
29 Chapter Three: Who Uses Bereavement Services? An Examination of Service Use by Bereaved Dementia Caregivers
30 Abstract Despite the availability of a wi de range of services and inte rventions for bereaved family caregivers, little is known about the characteristics of thos e who use them. The current study employed Andersen's Behavioral Model of Health Services Use in an examination of bereavement service util ization among dementia caregiv er participants in the Resources for Enhancing AlzheimerÂ’s Caregi ver Health (REACH) study. Approximately one in three participants utilized bereav ement services, primarily during the first 9 months after the death of the care r ecipient. Support groups, counseling, and psychotropic medications were used in sim ilar proportions. Depre ssive symptomatology and complicated grief served as need factor s, with higher levels resulting in higher probability of service utilizati on. Race/ethnicity served as a predisposing characteristic, with Black caregivers less likely and Hispanic caregivers more likely than White caregivers to utilize suppor t groups. Future research should focus on interventions designed to assure timely access to those bere aved family caregivers who are most in need.
31 Introduction It has been well establishe d that caring for a loved one with AlzheimerÂ’s disease or dementia can be difficult and stressful, potentially leading to a range of negative outcomes for family caregivers (Pinquart & S rensen, 2003). With bereavement (i.e., the death of the care recipient) increasingly rec ognized as an experience not separate from the caregiving Â“careerÂ” (Pearlin & Aneshensel 1994), many studies of the bereavement transition among family caregivers (see Schulz, Newsom, Fleissner, Decamp, & Nieboer, 1997 for a review) have focused on mental heal th outcomes. Recent studies suggest wide individual variation in both the initial response a nd the long-term trajectory of adjustment to bereavement (Aneshensel et al., 2004; Bonanno et al., 2002; Zhang, Mitchell, Bambauer, Jones, & Prigerson, 2008). While some family caregivers respond to bereavement with remarkable resilience, othe rs experience varying degrees of depressive symptomatology, anxiety, complicated grief, and other mental health problems. While the range of available services varies by community, bereaved family caregivers who are experiencing negative ment al health outcomes have a variety of choices regarding services de signed to facilitate their ad justment. Some consult a psychologist, psychiatrist, or mental health prof essional; others may pr efer the services of their family physician; while still others look to a trusted spiritual or religious advisor. Hospice, hospitals, and other community organizations offer self-help and support groups, bereavement counseling, and widow-to-wi dow programs. Despite the availability of these and other services for bereaved family caregivers, we know little about the characteristics of those who do, and those w ho do not, utilize bereavement services.
32 A review of the literature reveals mixed findings regarding variables associated with formal bereavement serv ice utilization. In a study of be reaved hospice caregivers, Cherlin and colleagues (2007) found a ber eavement service util ization rate of approximately 30%. Spousal relationship, younge r age, presence of Major Depressive Disorder, having discussed the care recipien tÂ’s prognosis with a doctor, having assisted the care recipient with Instrumental Activitie s of Daily Living (IADLs), and the presence of adequate instrumental support were associ ated with bereavement service utilization. Levy and Derby (1992) found that social support was not asso ciated with service use, while the experience of stressful events after the death and higher levels of depressive symtpomatology and anxiety were associat ed with bereavemen t support group use among bereaved spouses, of whom approximately 30% utilized the servic e. Caserta and Lund (1992) found a formal service utilization rate of nearly 12% in th e first two to three months after the death of a spouse. Demogr aphic characteristics, social network, and social support were not associated with th e use of formal services. Those who utilized formal services reported higher depression sc ores, lower coping abil ity, poorer perceived health, and lower levels of instrumental a nd resource-identification personal competency skills than non-users. The small number of studies of bereavem ent service utilizatio n conducted to date have not used a theoretical framework to guide the selection of measures and have methodological limitations which highlight the need for more research in this area. Andersen's Behavioral Model of Health Services Use (Aday & Andersen, 1974; Andersen, 1968; Andersen, 1995; Andersen & Newman, 1973), developed to aid in the identification and description of determinants of health care use, is a widely used
33 perspective in health service utilization resear ch and is therefore a useful theoretical and conceptual tool in the investigation of the determinants of bereavement services utilization. Within AndersenÂ’s model, predisposing characteristics are those which, present prior to the onset of need, dete r or predispose individuals from using health care services. Examples include demographic characteristics (e.g., age), aspects of social structure (e.g., race/ethnicity), and health beliefs (e. g., perceptions and knowledge of disease and medical care). Enabling resources are those which allow for the use of services, which include personal resources (e.g., income), family resources (e.g., social network characteristics), and community resources (e.g., geographic characteristics of the community). Need variables include the level of illness (e.g., depression) or the need for health care, both as perceive d by the individual and as evaluated by health personnel (Aday & Andersen, 1974; Andersen, 1968; Andersen & Newman, 1973). Some researchers have argued that need variables are the most immediate predictor of health services utilization (Aday & Andersen, 1974; Coulton & Frost, 1982; Kart, 1991; Wallace et al., 1998). However, the roles of predisposing characteristics and enabling resources may take on greater importa nce relative to need factors when greater discretion can be exercised by the potential consumer in the decision to utilize health services and in the type of service purs ued (Aday and Awe, 1997), as may be the case with bereavement services. The Resources for Enhancing AlzheimerÂ’s Caregiver Health (REACH) study provides a valuable means to examine determinants of bereavement service utilization for a number of reasons. First, the study include s a comprehensive range of caregiver and
34 care recipient characteristics, resources, a nd bereavement outcomes. The multi-site sample represents a broad geographic sample of diverse family caregivers who were followed over a period of more than 18 months. Finally, the sample is of sufficient size to employ a range of multivariate statistical techniques. The current study was conducted to examine the determinants of bereavement service utilization in a diverse sample of bereaved dementia family caregivers. Guided by the literature reviewed above and the lite rature on general ment al health services utilization, we hypothesized th at caregivers who were younge r (Cherlin et al., 2007), female (Rhodes, Goering, To, & Williams, 2002), non-minority (Harris, Edlund, & Larson, 2005), and more highly educated (C rabb & Hunsley, 2006) would be more highly predisposed to the util ization of bereavement services. We also hypothesized that prior participation in a caregiver intervention, higher re ligiosity, and higher levels of social resources (Cherlin et al., 2007) would serve to strengthen an indi vidualÂ’s community and formal service network ties and enable the ut ilization of bereavement services. Finally, we expected that greater distress in ber eavement, as indicated by higher levels of depressive symptoms (Caserta & Lund, 1992; Cherlin et al.; Levy & Derby, 1992), anxiety (Levy & Derby), and complicated grie f, would influence need and result in a higher probability of be reavement service use. Methods The Resources for Enhancing AlzheimerÂ’s Caregiver Health (REACH) Study is a longitudinal, multisite, prospective study f unded by the National Institute for Nursing Research and the National Institute on Ag ing and conducted with AlzheimerÂ’s disease caregivers and patients begi nning in 1995. A primary goal was to evaluate the
35 effectiveness of interventions with family caregivers of individuals with AlzheimerÂ’s disease and related disorders (Schulz, Burg io et al., 2003). Baseline and 6, 12, and 18month post-baseline follow-up data used in this study, as well as in troductory information and codebook, were downloaded from the Inte r-university Consortium for Political and Social Research (ICPSR), part of the Institute for Social Re search of the University of Michigan. Participants In order to be eligible for REACH, partic ipants had to be a family caregiver living with a person with dementia for at least 6 months and providing care for a minimum of four hours each day. Participants were random ly assigned to the intervention or control conditions. Interventions va ried by study site, including Bi rmingham, Memphis, Boston, Philadelphia, Miami, and Palo Alto. A tota l of 1,222 caregiv er-care recipient dyads were recruited for participation (W isniewski et al., 2003). During the course of the study, 265 participants experienced the death of the care recipient. Bereavement outcome and service utilization data were obtained fr om 224 of these bereaved caregivers. For the purposes of our study, waves of data were reconceptualiz ed as either preor post-loss according to the timing of the d eath of the care recipient. Therefore, participants varied in the number of post-loss waves of data available. Specifically, some participants contributed three waves of post-loss data (N=60), as the death of their care recipient occurred between the baseline and 6-month follow-up interviews. For others, we had 2 or 1 waves of post-loss data (N=64 and N=100 respectively), depending on whether the care recipient died between 6 and 12-months follow-up or between 12 and 18-months follow-up. Twelve bereaved caregiver s did not participate in the first wave of
36 bereavement follow-up, but were available for subsequent waves of post-loss data collection. Measures Predisposing Characteristics The predisposing characteristics on which we focused were baseline sociodemographic characteristics including care giver age, gender, education (measured in years), and race/ethnicity (White, Black, Hispanic). Enabling Resources Enabling resources included caregiver inte rvention status (treatment vs. control) and religiosity. Religiosity was measured at baseline with a single item in which caregivers reported the importance of religion in their lives using a 4-point Likert scale ranging from Â“not importantÂ” to Â“very importa nt.Â” Also included were two measures of social resources social network and satis faction with support Â– obtained at the first available post-loss interview. Social networ k was measured with 6 items from LubbenÂ’s Social Network Scale (LSNS; Lubben, 1988). M easures included the number of relatives or friends seen at least once a month (0 to 9 or more), frequency of contact (less than monthly to daily), and the number of relatives or friends the subj ect felt close to (0 to 9 or more). Satisfaction with support was measured with 3 items measuring satisfaction with tangible, emotional, and informational support received (Krause, 1995; Krause & Markides, 1990), with modified response options ranging from Â“ not at allÂ” to Â“veryÂ” on a 4-point Likert scale. Mean imputation wa s used to avoid losing observations due to missing data. Social network (LSNS) had 9 missing values (4%) and satisfaction with support had 10 missing values (4.5%).
37 Need Factors Need factors included three outcomes frequently examined in studies of bereavementÂ– depressive symptoms, anxiety, an d complicated grief, as measured at the first available post-loss follow-up interview. Depressive symptoms were measured using the Center for Epidemiological Studies-Dep ression (CES-D) scal e (Radloff, 1977), a 20item measure of the frequency with which re spondents experienced depressive symptoms within the past week. Anxiety was measured w ith the 10-item Â“stateÂ” anxiety subscale of the State-Trait Personality Inve ntory (Spielberger, 1979; Spie lberger, Gorsuch, Lushene, Vagg, & Jacobs, 1983). Participants were asked to report their experience of symptoms of anxiety during the past week us ing a 4-point scale of intensit y ranging from Â“not at allÂ” to Â“very much.Â” Complicated grief was measur ed using Prigerson and colleaguesÂ’ (1995) well-validated 19-item Inventory of Comp licated Grief designed to measure the frequency of grieving thoughts and behaviors considered to be maladaptive (e.g., anger, disbelief, hallucinations). Responses ranged on a 5-point scale from Â“neverÂ” to Â“always.Â” Bereavement Service Utilization At each post-loss follow-up, REACH participants were asked whether they were currently receiving any type of counseling or other support services to help in coping with the death of the care recipi ent. If so, they were then prompted with further questions to describe the nature of the services being utilized. For thes e analyses, participants were coded as using a support group if they confir med the use of a support group for family members of bereaved dementia caregivers or other support group. They were coded as using counseling if they confirmed the use of individual counseling, family counseling, or clergy/pastoral counseling.
38 At each wave, participants were asked to report all prescription and nonprescription medications they were currently taking. Participants were coded as using psychotropic medication if they confirmed the use of any medication identified as a psychiatric medication using the Systems Key of the Instant Drug I ndex (Aloisi, 2001). Participants were further code d as using any bereavement se rvice if they confirmed the use of any one or more of the above bereavement services. Statistical Analysis Descriptive statistics were calculated to examine the sampleÂ’s demographic and background characteristics and correlational an alyses were used to examine bivariate associations among study variables. Multiv ariate analyses were conducted using Cox proportional hazards modeling (Allison, 1995; Ta bachnick & Fidel1, 2001). This method was chosen for its utility in examining the li kelihood of occurrence of a given outcome as well as covariates associated with risk of event occurrence. In addition, Cox proportional hazards models can accommodate censored cases or those in which the event is not experienced during the observational period. The dependent variable for the present study was defined as time to bereavement serv ice utilization following the death of the care recipient. For those cases in which be reavement services were utilized, time was calculated as the number of days between the date of the care recipientÂ’s death and the date of the interview at which use was fi rst reported. For censored cases in which bereavement service utilization did not occur, time was calculated as the number of days between the date of the care recipientÂ’s deat h and the date of part icipation in the final follow-up interview.
39 Four Cox proportional hazard models were constructed to examine the relationship of hypothesized pr edisposing character istics, enabling resources, and need factors with the likelihood of (a) any bereavement servic e utilization, (b ) support group use, (c) counseling use, and (d) psychotropi c medication use. The hierarchical method was chosen to control the order of entrance of blocks of covariates in accordance with the Andersen model of health services use. Cox proportional hazards models provide estimates of the effect of a covariate on th e likelihood of the occurrence of the outcome. Results Table 3.1 displays the desc riptive and backgr ound characteristic s of the 224 study participants. The sample consisted primarily of White females with a mean of 12 years of education and an average age of 63 years. On average, the amount of time between the care recipientÂ’s death and the first bere avement follow-up interview was 99 days (N = 212; SD = 52.8), or 3.3 months. Average time until the second bereavement follow-up interview was 261 days (N = 124; SD = 63.7), or 8.5 months, and until the third bereavement follow-up was 445 days (N = 66; SD = 65.0), or 14.6 months. The results of the univariate correlation analyses revealed that the three need factors depressive symptoms, anxiety, and complicated grief Â– were significantly and positively correlated. Those with higher depressive symptom scores were also more likely to have higher anxiety scores (r=.80, p<.000) and highe r complicated grief scores (r=.59, p<.000). Similarly, those with higher an xiety scores were al so more likely to have higher complicated gr ief scores (r=.48, p<.000).
40 Table 3.1. Sample Background Char acteristics and Descriptives N (%) Mean SD Range Predisposing Characteristics Age (years) a 63.2 13.3 28 89 Gender Female Male 189 (84.4) 35 (15.6) Race/Ethnicity White/Caucasian Black/African American Hispanic/Latino 148 (66.1) 45 (20.1) 31 (13.8) Education (years) 12.3 3.0 0 17 Enabling Resources Religiosity a 3.5 .85 14 Caregiver Intervention Status Intervention Group Control Group 143 (63.8) 81 (36.2) Social Network (LSNS) b c 18.8 5.0 5 30 Satisfaction with Support b 6.4 2.3 0 9 Need Factors Depressive Symptoms (CES-D) b 16.2 12.3 0 Â– 60 Anxiety (STPI anxiety subscale) b 20.0 6.7 10 40 Grief (ICG) b 37.3 12.8 19 86 SD = Standard Deviation LSNS = LubbenÂ’s Social Network Scale CES-D = Center for Epidemiol ogical Studies-Depression scale STPI = State-Trait Personality Inventory ICG = Inventory of Complicated Grief a baseline b first bereavement follow-up The support group utilization rate reported in the curre nt study was 17% (N = 38), reflecting participation in a support group fo r family members of bereaved dementia caregivers or other support gr oup at any bereavement follow-up interview. First use of a support group was reported by participants who used them an average of 154 days (SD = 92.1), or 5.1 months, after the death of the care r ecipient. Ninety percent (N = 34) utilized a support group during the first 9 months after the care recipientÂ’s death.
41 The Cox proportional hazards model for support group use is displayed in Table 3.2. The model indicated that the predispos ing characteristics block successfully predicted time to support group use ( 2 (5) = 12.75, p = .026). Bl ack caregivers were 77% less likely to participate in a support group than White caregivers, while Hispanic caregivers were more than two times more likely than White caregivers to participate. In addition, the needs factor block containing co mplicated grief successfully predicted time to support group use after adjusting for pr edisposing characteristics and enabling resources ( 2 (1) = 6.72, p = .010), with each one-un it difference in complicated grief score resulting in a 4% increase in the probability of suppor t group participation. Table 3.2. Cox Proportional Hazards Model of Support Group Use 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age .001 1.001 .975 1.028 Gender a -.232 .793 .268 2.345 Race/ethnicity White/Caucasian b Black/African American -1.468 .230* .054 .982 Hispanic/Latino .837 2.310* 1.028 5.193 Education .027 1.028 .927 1.139 Enabling Resources Caregiver Intervention c .136 1.145 .571 2.296 Religiosity .351 1.421 .894 2.260 Social Network -.028 .972 .908 1.042 Satisfaction with Support .005 1.005 .865 1.169 Need Factors d Complicated Grief .036 1.037** 1.009 1.065 *p<.05, **p<.01 a Reference category: females b Reference category c Reference category: control group d Depressive symptoms and anxiet y were excluded from the model
42 The counseling utilization ra te, or use of individual, family, or clergy/pastoral counseling at any wave of bereavement fo llow-up, was 16% (N = 36). On average, participants who used the service reported first use of counseling 126 days (SD = 69.6), or 4.1 months, after the death of the care recipient. All but one (i.e., N = 35; 97%) used counseling services within the first 9 months of the death. As shown in Table 3.3, the Cox proportional hazards model indicated that age served as a significant predisposing characte ristic, with increasing age resulting in a lower probability of counseling use. Fu rther, the needs factor block containing depressive symptoms successfully predicted time to counseling use after adjusting for Table 3.3. Cox Proportional Hazards Model of Counseling Use 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age -.026 .974* .949 .999 Gender a .357 1.429 .577 3.538 Race/ethnicity White/Caucasian b Black/African American -.482 .618 .248 1.536 Hispanic/Latino -.110 .895 .291 2.754 Education .066 1.068 .938 1.216 Enabling Resources Caregiver Intervention c .153 1.165 .588 2.310 Religiosity .007 1.007 .691 1.468 Social Network -.022 .978 .911 1.051 Satisfaction with Support .012 1.012 .868 1.179 Need Factors d Depressive Symptoms .050 1.051** 1.021 1.082 *p<.05, **p<.01 a Reference category: females b Reference category c Reference category: control group d Anxiety and complicated grief were excluded from the model
43 other covariates ( 2 (1) = 11.55, p = .001), with each one-unit difference in depressive symptoms resulting in a 5% increase in the probability of counseling use. During bereavement follow-up, 19% of study participants (N = 42) reported the utilization of psychotropic medi cation, all within the first 9 m onths after the death of the care recipient. First medicat ion use was reported an average of 122 days (SD = 69), or 4 months after the death. The Cox proportional hazards model for psychotropric medication use (displayed in Table 3.4) indicated that the needs factor block containing de pressive symptoms Table 3.4. Cox Proportional Hazards Mode l of Psychotropic Medication Use 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age -.003 .997 .972 1.022 Gender a -.225 .798 .303 2.104 Race/ethnicity White/Caucasian b Black/African American -.724 .485 .185 1.271 Hispanic/Latino .012 1.012 .386 2.652 Education .000 1.000 .902 1.109 Enabling Resources Caregiver Intervention c .457 1.579 .839 2.971 Religiosity -.292 .747 .532 1.048 Social Network .027 1.027 .961 1.097 Satisfaction with Support -.016 .984 .851 1.137 Need Factors d Depressive Symptoms .054 1.055*** 1.027 1.085 *p<.05, **p<.01 a Reference category: females b Reference category c Reference category: control group d Anxiety and complicated grief were excluded from the model
44 successfully predicted time to psychotropic medication use after adjusting for other covariates ( 2 (1) = 14.77, p = .000), with each one-point difference in depressive symptoms resulting in a 6% increase in the probability of psychotropic medication use. Table 3.5. Cox Proportional Hazards Model of Overall Bereavement Service Utilization 95% Confidence Interval B Odds Ratio Lower Upper Predisposing Factors Age -.007 .993 .975 1.012 Gender a .034 1.035 .516 2.074 Race/ethnicity White/Caucasian b Black/African American -.497 .608 .306 1.210 Hispanic/Latino .531 1.700 .889 3.254 Education .018 1.018 .945 1.098 Enabling Resources Caregiver Intervention c .070 1.073 .660 1.745 Religiosity -.044 .957 .727 1.261 Social Network -.006 .994 .946 1.045 Satisfaction with Support .033 1.033 .926 1.153 Need Factors d Depressive Symptoms .035 1.035** 1.015 1.056 **p<.01 a Reference category: females b Reference category c Reference category: control group d Anxiety and complicated grief were excluded from the model The overall bereavement service utilization rate, reflecting use of any one or more of the above bereavement serv ices during the course of th e study, was 33.9% (N = 76). The first use of any bereavement service was reported an average of 137 days (SD = 83.8), or 4.5 months, after th e care recipientÂ’s death and 93% (N = 70) reported use within the first 9 months.
45 In the overall Cox proportional hazard s model of bereavement service use (displayed in Table 3.5), only the needs fact or block containing de pressive symptoms successfully predicted time to service use ( 2 (1) = 11.36, p = .001). After adjusting for other covariates, each one-point difference in depressive symptoms resulted in a 4% increase in the probability of bereavement service utilization. Discussion Overall, approximately one-third of bereav ed family caregivers utilized services, almost entirely during the first 9 months after the death of the care recipient. Participants utilized support groups, counseling, and psychot ropic medications in similar proportions. Slightly less than one fifth of participants ut ilized each of these bereavement services to facilitate their adjustment to bereavemen t after providing care to a loved one with AlzheimerÂ’s disease or dementia. These ra tes of bereavement se rvice utilization are consistent with rates found in other studi es (Billings & Kolton, 1999; Caserta & Lund, 1992; Cherlin et al., 2007; Levy & Der by, 1992; Provini et al., 2000; Schulz, Mendelsohn, et al, 2003). The most striking finding in the present analyses is the prominent role of need factors in the utiliza tion of bereavement services by fam ily caregivers. Consistent with studies of other types of hea lth services utilization empl oying the Andersen model (Aday & Andersen, 1974; Coulton & Frost, 1982; Kart 1991; Wallace et al., 1998) need factors, including depressive symptoms and complicated grief, appear to be important in driving service use. In another recent study of bereavement se rvice utilization grounded in AndersenÂ’s model (Bergman, Haley, & Small, 2008) grief served as a need factor in the use of any bereavement service and in talking with a physician for help in coping with the
46 death of the care recipient. Bergman and Haley (2008) also found no relationship between depressive symptoms and bereavemen t service utilization in a study of bereaved hospice caregivers. Taken together, these findings suggest that while bereavement outcomes are an important consideration in the planning and design of bereavement services, more research is needed to unders tand the contributions of a broad range of potential need factors. We are encouraged by our findings in th e current study which indicate that those who are most distressed in bereavement are seeking out and utilizing bereavement services. This suggests the need for continue d and increased outreach and other efforts to assure timely access to those bereaved family caregivers who are most in need. This could be achieved through the standardization of regular sc reenings by family physicians of bereaved family caregivers and referral to bereavement services for those identified as experiencing mental health decrements. Also of potential benefit are educational efforts designed to inform family caregivers and th e general public that, while some emotional distress is to be expected, excessive psyc hological distress is not a normal part of bereavement and services are available to ai d in the adjustment to this transition. Furthermore, efforts to assure the availabil ity of a wide range of carefully-evaluated services which appeal to the needs and pr eferences of diverse family caregivers are important. Also of interest in the cu rrent study is the finding that race/ethnicity served as a predisposing characteristic for support group ut ilization, with Black caregivers being the least likely and Hispanic careg ivers the most likely to util ize a support group. Research suggests that Black individuals possess unique strengths in coping with family caregiving
47 (for a review see Dilworth-Anderson et al., 2002), strengths which may be carried into bereavement (Schulz et al., 2001) resulting in decreased need for bereavement services. In a study using REACH data of racial di fferences between Caucasian and AfricanAmerican caregivers (Haley et al., 2004), Caucasian car egivers reported significantly higher levels of anxiety but no difference emerged in the experience of depressive symptoms. No ethnic difference on either of these measures was found when comparing the experiences of Latina a nd White caregivers in the REACH study (Coon et al., 2004). Similarly, post hoc analyses in the current study revealed no differe nces in depressive symptoms, anxiety, or complicated grief be tween White, Black, and Hispanic bereaved caregivers. Given that Black caregivers in the present study were not suffering less in bereavement than their White or Hispanic peers, Black caregivers may have lacked knowledge of or access to support group servic es (Bradley et al., 2002), held stronger preferences against s upport group use, or relied more h eavily instead on their informal support network as has been found in studies comparing the formal service utilization rates of different racial and ethnic groups (Dilworth-Anderson et al., 2002; Mui & Burnette, 1994; Williams & Dilworth-Anderson, 2002). In general, Hispanic adults in the U. S. report fewer mental health problems and less mental health service use than White adults (Harris et al., 2005). A recent metaanalysis (Pinquart & Srens en, 2005) reveals, however, that Hispanic caregivers experience worse physical and mental health than White caregivers. Thus, for many Hispanic caregivers, bereavement may represen t an end to the role captivity experienced during caregiving and the opportuni ty to address long-standing mental health issues and grief. An alternative potential explanation fo r our finding that bereaved Hispanic family
48 caregivers utilized supp ort groups at a rate more than tw ice that of their White peers may be that Spanish-speaking support groups we re more accessible or highly preferred by Hispanic caregivers than other types of bereavement services. To date, the REACH dataset represents the largest and most diverse sample of family caregivers with which to examine bereavement service utilization. However, further research with larger sample sizes and more specific measures is necessary in order to better understand the reasons behind a nd implications of our findings regarding race/ethnicity and bereavement service utilization. Contrary to our expectat ions, few other variables in the current study served as predisposing characteristics or enabling res ources in the utilization of bereavement services in the current study. More research is needed in this area to examine the roles of demographic characteristics, re ligiosity, social resources, an d other potential covariates affecting bereavement service utilization among diverse samples of bereaved family caregivers. One such potential addition to the model is health beliefs, described by Andersen (1995) as a predis posing characteristic includ ing Â“attitudes, values, and knowledge that people have about health and h ealth services that might influence their subsequent perceptions of need and use of health servicesÂ” (p 2). Related to the current study, health beliefs may include individual at titudes about Â“normalÂ” and Â“pathologicalÂ” grief. For example, an individual who be lieves that a certain degree of psychological suffering is a normal part of bereavement may be less inclined to ut ilize services than someone who does not hold this health belief. The Andersen model has been used exte nsively to examine health care utilization (e.g., Andersen, 1968) and equity in the health care system (e.g., Andersen et al., 1975),
49 to make cross-national comparisons in health service utilization (Andersen, 1976), and in many studies of the use of nursing home s (e.g., Freedman, 1993), in-home nursing (e.g., Rabiner, 1992), adult day care (e.g., Conrad et al., 1992), and other formal and informal services by older adults. The primary strength of the curren t study is the extension of the application of the Andersen model to the exam ination of bereavement service utilization. As with any research conducted using secondary data, the current research has some limitations which should be noted. Specifically, in the current study need was conceptualized as the self-report of sympto ms of depression, anxiety, and complicated grief at the fist interview after the death of the care recipient. We were not able to examine the role, if any, of need as evaluated by a healthcare professional. Communication with a physicia n or other healthcare provide r in the early stages of bereavement may result in the identification of the need for intervention, referral to bereavement services, or increased knowledge of available services and these factors may in turn increase the probability of bereavemen t service utilization (C herlin et al., 2007). Hospice enrollment of the care recipient may al so lead to increased bereavement service utilization through increased acce ss to health personnel trained in end-of-life issues and the provision of bereavement follow-up mandate d by Medicare regulations. Future study is needed regarding the influence of communi cation with a physicia n or other healthcare provider, such as hospice bereavement pers onnel, on perceived and evaluated need for bereavement services.
50 Chapter Four: Depressive symptoms, social network, and bereavemen t service utilization and preferences among spouses of former hospice patients
51 Abstract Background : Bereavement services are an importa nt part of comprehensive end-of-life care with potential to ameliorate physical, psychological, and spiritual distress. We studied bereaved spouses of hospice patients to examine bereavement service utilization, barriers, and preferences regarding conten t, structure, and delivery of potential bereavement services. We also examined the impact of depressive symptoms and social network. Methods: Retrospective cohort study of be reaved spousal caregivers of patients of three hospices in Tampa Bay. De scriptive and univaria te analyses assessed demographics, depressive symptoms, social network, service utiliz ation, barriers, and preferences. Results: Nearly half utilized at least one type of specialized professional bereavement intervention to ai d in coping with th eir loss. The most frequently used services were provided by clergy members and physicians. Primarily attitudinal in nature, barriers included the finding that more than one third felt available services did not fit their needs or interest s. Individual and sp iritually-based services were highly endorsed, as were services de signed to provide tools to re frame the loss and cope with accompanying changes and emotions. Lower social network was associated with higher content preferences for services consistent pr imarily with restorati on-oriented coping. Conclusion: Clinicians and service providers may facilitate coping by routinely screening for depressive symptoms and soci al network and tailor ing interventions to those identified as experiencing elevated distre ss or lacking social re sources. Attitudinal barriers and preferences suggest that even in the servicerich environment of hospice some modification of bereavement services mi ght reach more bereaved spouses. Future
52 studies might address whether preferences lead individuals to servic es of the greatest benefit.
53 Introduction Nearly one million individuals in the United States each year experience the death of their spouse (Kung, Hoyert, Xu, & Murphy, 2008). This can be a difficult and stressful transition, leading to considerable phys ical and psychological distress for some individuals (Bonanno et al., 2002; Genevro, 2004). Bereavement outcomes which have been reported in the literatur e include grief, depression, an xiety, diminished quality of life, increased health care utilization, and incr eased risk of physical illness and mortality (Christakis & Iwashyna, 2003; Goodkin et al., 2001; Stroeb e et al., 2007). An array of services and interventions ha s been developed in an effort to assist widows and widowers in their adjustment to be reavement (Forte et al., 2004; Schut et al., 2001). Although most studies have focused on a pa rticular service or se rvices of interest rather than examining use of a comprehensiv e range of services, previous estimates of bereavement service utilization rates range from 10% to 35% (Billings & Kolton, 1999; Caserta & Lund, 1992; Cherlin et al., 2007; Le vy & Derby, 1992; Provi ni et al., 2000; Schulz, Mendelsohn, et al., 2003). However, litt le is known about the types of services individuals choose or about whether they face ch allenges or barriers to the utilization of these services. Further, it is important to c onsider the preferences of bereaved individuals regarding the content and struct ure of bereavement services, as they are likely to impact whether or not an individual initiates use, follows through upon referral, or maintains their participation in a service from which they could potentially benefit. In a study of attitudes towards mental h ealth care (Robb et al., 2003), older adults reported favorable attitudes, but lacked experience and knowle dge of mental health care as compared to younger adults. Many view ed their family doctors and clergy as
54 important sources for help with mental hea lth issues and for referral to other mental health services. Because most individuals experience widowhood in later life, these findings are likely to have bearing in the ut ilization of services for bereaved spouses. Other services frequently available to surv ivors of hospice patient s include bereavement support groups and bereavement counseling. Research indicates that bereavement inte rventions can be of benefit in mitigating the potential negative outcomes associated with spousal loss (Larson & Hoyt, 2007a; Zisook & Shuchter, 2001). However, while cl aims of treatment induced deterioration effects of bereavement counseling (Fortner 1999; Neimeyer, 2000) have largely been dismissed (Larson & Hoyt, 2007b), reviewers of the efficacy of bereavement services (Raphael, Middelton, Martinek, & Misso, 1993; Schut & Stroebe, 2005) argue against routine intervention for bereaved individuals, citing a lack of justification for their need by those who respond to bereavement with resilience (Chentsova Dutton & Zisook, 2005). Instead, experts advocate for the targeti ng of resources and in terventions to those suffering in bereavement, those in a highrisk group, and to those lacking access to supportive persons and resources (Parkes, 1998). Thus, interv entions and services which are not tailored to fit the n eeds of these individuals and wh ich they do not find appealing are not likely to be utilized (Mittelman, 2008; Robb et al., 2 003; Zhang et al., 2006), and represent an inefficient use of resources. It is important to study the preferences and motivations of bereaved indivi duals and to incorporate these into the design of services which will appeal to and facilitate ad justment among those most in need. Two risk factors which may simultaneously serve as a reason bereaved spouses may benefit from services and why they do not s eek services are high le vels of depressive
55 symptoms (Jordan & Neimeyer, 2003) and low leve ls of social resour ces. Depression is a frequently studied bereavement outcome (Stroe be et al., 2007), and it is estimated that between 25% and 45% of bereav ed individuals experience m ild levels and between 10% and 20% experience clinically significant le vels of depressive symptoms (Hansson & Stroebe, 2003). In the general mental health services literature, elevated distress is associated with increased likelihood of utiliz ing mental health services (Mojtabai & Olfson, 2006; Rhodes, Jaakkimainen, Bondy, & Fung, 2006). However, Prigerson and colleagues (2001) found that those suffering th e most in bereavement tended to be the least likely to seek out formal services. Older bereaved spouses experiencing elevated levels of depressive symptoms may favor services viewed as psychoeducational or supportive (e.g., support groups) rather than serv ices viewed as treatment for a mental health diagnosis (e.g., psychologist/psychia trist), due to conc erns about stigma (Gallagher-Thompson et al., 2003 ) and perceptions of their n eed for professional help as indication of their inability to live in dependently (Bambauer & Prigerson, 2006). Research indicates that social network members are an important factor in the decision to utilize mental h ealth services in general (V ogel, Wade, Wester, Larson, & Hackler, 2007) and that those with larger social networks are less likely to use formal, outpatient mental health serv ices (Kang et al., 2007; Pesc osolido, Wright, Alegra, & Vera, 1998), whereas those with greater per ceived deficits in the amount of support available from social network members are more likely to use formal mental health services (Phillips & Mu rrell, 1994). In bereavement, thos e with larger so cial networks characterized by more frequent contact and greater perceived clos eness with network members may be less likely to need services as they r eceive support informally from
56 network members. Alternatively, larger so cial networks characterized by closer relationships and more frequent contact may facilitate service use as network members observe suffering and encourage the use of fo rmal resources or make their use possible through the provision of instrumental support, such as transportation. The present study was conducted in an effo rt to examine the bereavement service utilization rates and perceive d barriers to servi ce use reported by former spousal hospice caregivers and to examine their preferences regarding the conten t, structure, and delivery of potential bereavement services. An addi tional goal was to examine group differences in bereavement service utilization rates, ba rriers, and preferences based on levels of depressive symptoms and social network. We hypothesized th at those with higher levels of depressive symptoms and smaller social neworks would report higher bereavement service utilization rates and ratings reflec ting stronger preferences regarding potential bereavement services. Methods Procedure This study was conducted in collaborat ion with three hospice organizations serving the Tampa Bay community, includi ng LifePath Hospice and Palliative Care, Hernando Pasco Hospice, and Hospice of S outhwest Florida (now TideWell Hospice and Palliative Care). The spouses of patients who died unde r the care of these hospices between 6 and 20 months prior received a letter from the hospice bereavement department inviting them to pa rticipate in the study. Data were collected during in-home interviews conducted by the lead author.
57 Participants The final sample included 61 bereaved spousal caregivers of former hospice patients. The mean age of par ticipants was 72 years (SD=10). Measures Demographics & Sample Characteristics Basic demographic and sample ch aracteristics included age, gender, race/ethnicity, education, religious affilia tion, the number of years married, and the amount of time elapsed since the death. Depressive Symptoms We utilized the 20-item version of the Center for Epidemiologic StudiesDepression Scale (CES-D), a se lf-rated measure of depressive symptoms (Radloff, 1977). Participants were asked to provide information as to th e frequency of behaviors and feelings during the past week. Responses ra nged from Â“rarely or none of the timeÂ” to Â“most of the timeÂ” on a 4-point Likert scale. The CES-D has been used in a broad range of studies, including those of caregiving and bereavement, and validated cutpoints for clinical significance have been established. Specifically, a score of 16 or higher on the 20-item version of the CES-D is indicativ e of clinically relevant depressive symptomatology (Lewinsohn, Seeley Roberts, & Allen, 1997). Social Network LubbenÂ’s Social Network Scale (LSNS) is a composite scale fo r measuring social network in adult populations (Lubben, 1988). The first six items of the LSNS, which assess social network size, frequency of cont act, and closeness with family and friends, were selected for use in this study. Possibl e scores range from 0-30, with higher scores
58 indicating greater size, more fr equent contact, and more cl oseness with members of the social network. Bereavement Service Ut ilization and Barriers Participants were asked a series of ques tions designed to lear n about their use of bereavement services. Specifically, they were asked whether or not they received help from a list of professionals (see Table 4.2) to aid in dealing with their loss and about support group attendance and use of bereavem ent counseling. They were also presented with a range of potential barri ers (see Table 4.3) to bereav ement service utilization and asked to retrospectively report whether or not they experienced any of these barriers. Bereavement Service Preferences In the first part of this measure, usin g a 7-point Likert s cale ranging from Â“least appealingÂ” to Â“most appealing,Â” participants were asked to rate 19 items (see Table 4.4) pertaining to the content of potential bereavement services. Items included in this measure were intended to reflect a variety of perspectives on be reavement intervention, including grief work (Bonanno, 2001), cognitive-behavioral therapy and cognitive restructuring (Boelen, de Keijser, va n den Hout, & van den Bout, 2007), and psychoeducational, interpers onal resource and practical sk ill building perspectives (Caserta & Lund, 1993). The measure was also designed to reflect the dual process model of grief (Stroebe & Schut, 1999), in cluding features of interventions which emphasized loss-oriented and restorationoriented coping among the bereaved. The second part of this measure asked participan ts to rate 11 items (see Table 4.5) pertaining to the structure and delivery of bereavement services.
59 Statistical Analysis Basic descriptive analyses were used to assess participantsÂ’ demographic characteristics, depressive symptoms, and soci al network as well as bereavement services utilization, barriers, and prefer ences. Independent samples ttests and chi-square analyses were used to examine group differences in se rvice utilization, barri ers, and preferences based on level of depressive symptoms and so cial network. The sample was divided into low and high social network groups using a me dian split. Those with scores of 21 or higher on the LSNS were categorized as high so cial network and those with scores below 21 were categorized as low so cial network. The sample wa s divided into low and high depressive symptoms groups us ing a cutoff score of 16 or higher on the CES-D, a score suggestive of clinically relevant depressive symptomatology. Since bereavement service utilization and perceived barriers were meas ured dichotomously, chi-square tests were used to examine associations. Group differe nces in preferences were examined using independent samples t-tests. Preference scor es for each of the 30 items were centered using z-transformation in order to standardi ze the interpretation of mean comparisons, but for simplicity of presentation raw means are shown in the tables. Results Sample Characteristics Table 4.1 displays the charac teristics of the sample, which consisted primarily of Caucasian females. Participants ranged in age from 46 to 87 years and most reported at least some education beyond high school. Th e length of time part icipants had been married to their spouse ranged from 2 to 65 years, with a median of 40 years. Bereavement occurred between 7 and 18 mont hs prior to taking part in the study. The
60 median length of time since the death was 11 months. The mean CES-D depressive symptoms score for participants was 15.7 (SD = 10.2; range 0-39). The mean LSNS score was 20.2 (SD = 5.6; range 2-29). Table 4.1. Sample Characteristics (N = 61) Variable N (%) Gender Female 45(73.8%) Male 16(26.2%) Race/Ethnicity Caucasian 60(98.4%) Hispanic 1(1.6%) Education 8th Grade or Less 6(9.8%) High School/GED 14(23%) Some College 17(27.9%) College Graduate 16(26.2%) Post-Graduate Study 8(13.1%) Religious Affiliation Protestant 34(55.7%) Roman Catholic 17(27.9%) Jewish 2 (3.3%) Other 8(13.1%) Bereavement Service Utilization We found that nearly half (45.9%) of the bereaved caregivers w ho participated in this study utilized at least one type of speci alized professional be reavement intervention, including either a support group, bereavem ent counseling, or a psychologist or psychiatrist to aid in coping with the death of their spouse. Nearly all participants
61 (88.5%) reported some bereavement service ut ilization when we employed a more liberal definition of some kind of formal professi onal service utilization, which also included receiving help from a physician, member of the clergy, social worker, or nurse. The most frequently used bereavement se rvices were provided by members of the clergy and by physicians, followed by be reavement support groups and individual bereavement counseling. Smaller proportions of participants repor ted talking with a social worker, volunteer, or a psychologist or psychiatrist a bout their loss. Table 4.2. Bereavement Service Utilization Rates Service N (%) Reporting Use (N = 61) Talked with Clergy 34(55.7%) Talked with Physician 30(49.2%) Support Group 24(39.3%) Bereavement Counseling 19(31.1%) Talked with Nurse 18(29.5%) Talked with Social Worker 13(21.3%) Talked with Volunteer 11(18%) Talked with Psychologist or Psychiatrist 6(9.8%) Barriers to Service Use The most commonly reported barriers to se rvice utilization were attitudinal, in that participants did not rec ognize the utility of bereavement services, they did not feel they needed assistance in coping with bereavement beyond that provided by their informal social network, or th ey perceived that the availabl e services did not match their
62 needs or interests. It is of particular note that 41% of bereaved spouses reported that available services did not fit their needs or inte rests, suggesting that even in the relatively service-rich environment of hospice some m odification of bereavement services might reach more bereaved spouses. A small minor ity of participants reported experiencing practical barriers to bereavement service utiliz ation, such as a lack of transportation or the inability to pay for services. Table 4.3. Perceived Barriers to Bereavement Service Utilization Perceived Barrier Yes N (%) DidnÂ’t see the use in participating 32 (54.2%) DidnÂ’t think I needed any help beyond family and friends 27 (44.3%) Services available didnÂ’t fit my needs or interests 25 (41%) Too depressed 4 (6.8%) Lack of transportation/Require d traveling too far 2 (3.3%) CouldnÂ’t afford a service I was interested in 1 (1.6%) My family did not want me to participate 0 (0%) Bereavement Service Preferences The mean preference ratings of all 19 items regarding the content of potential bereavement services are shown in rank orde r in Table 4.4. Results reflect preferences for services designed to give pa rticipants tools to reframe th eir loss and to cope with the changes and feelings that accompanied the loss. They also favored services focused on engagement with community and a broade ned social network and on distraction, relaxation, and stress management.
63 Table 4.4. Mean Preferences Regardi ng Content of Bereavement Services Preference M SD Find things to laugh about 4.22 2.23 Starts before your spouse dies and help you know better what to expect 4.18 2.30 Talk about your feelings related to the loss 4.13 1.95 Help you make new friends 4.08 2.07 Provide information about community resources 4.00 2.15 Teach you how to relax 3.95 2.28 Rethink or work through challenging situations 3.95 2.27 Focus on growth and things youÂ’ve learned 3.83 1.99 Give you tools to help with loneliness or depression 3.79 2.09 Stress management 3.75 2.22 Give you confidence to know you can make it through the loss 3.64 2.23 Express your grief over the loss 3.59 2.20 Make meaning of your loss 3.33 2.38 Teach you how to do things your spouse used to do 3.33 2.26 Explore your continued relati onship with your spouse 3.22 2.22 Get involved in your community and with your friends again 3.17 2.01 Help you stop thinking about your loss all the time 2.87 2.18 Work through Â“unfinished businessÂ” with your spouse 2.62 2.29 Relate better or participate more in lives of family members 2.49 1.86
64 Regarding structure and delivery, participan ts favored individual, spiritually based interventions delivered in th eir homes. Their preferences also reflected a desire for diversity with regard to gender and age am ong participants, and very low ratings for online and instructional formats. The mean preference ratings for all 11 items regarding the structure of potential bereavement serv ices are shown in rank order in Table 4.5. Impact of Depressive Symptoms Using a cutoff score of 16 on the CESD, 29 (47.5%) participants were included in the high depressive symptoms category and 32 (52.5%) were included in the low depressive symptoms category. FisherÂ’s exact and PearsonÂ’s chi-square tests were then used to compare the bereavement service ut ilization rates and re ports of barriers to service use among high and low depressive symptoms groups. Contrary to our expectations, depressive symptoms had littl e relationship with se rvice utilization or perceived barriers. No significant diffe rences between low and high depressive symptoms groups were found in utiliza tion rates of support groups, bereavement counseling, or in talking with a physicia n, clergy member, nurse, social worker, volunteer, psychologist or psyc hiatrist. The only significan t difference in barriers to service use was in the proportion of participan ts in the low and high depressive symptoms groups who felt too depressed to participate (0% and 13.8% respectivel y; FisherÂ’s exact test p = .045)
65 Table 4.5. Mean Preferences Regarding Structure of Bereavement Services Preference M SD Individual 5.19 1.91 Spiritually-based 4.42 2.31 Provided in your home 4.27 2.02 Group 3.73 2.27 Widow-to-widow 3.70 2.32 Limited by age group 3.37 2.15 Therapy 3.30 2.38 Self-administered 3.03 2.05 Instructional or lecture format 3.02 2.18 Limited by gender 3.02 2.09 Computer-based or online 2.24 1.93 The mean preference score for services desi gned to help participants find things to laugh about for the low depressive sympto ms group was 3.61 (N=31; SD=2.33) and for the high depressive symptoms group it wa s 4.86 (N=29; SD=1.94). A statistically significant difference was found between th e two scores, t(58) = 2.25, p=.029. There were no significant differences in preferen ces regarding the other 18 items related to content (see Table 4.4) or the 11 items relate d to the structure and delivery (see Table 4.5) of potential services based on level of depressive symptoms.
66 Impact of Social Network Group differences based on social networ k were then examined for each item on the bereavement service utilization measure and the measure of perceived barriers to service use. Of those in the low social ne twork group, 44.8% (n=13) utilized bereavement counseling, compared to 16.1% (n=5) of those in the high social network group, 2 (1, N=60) = 5.88, p=.015. No significant diffe rences emerged in the proportion of participants in the low and high social network groups who utilized the other seven bereavement services (see Table 4.2). For those in the high social network gr oup, 71.0% (n=22) reported that they did not see the use in participating in bereavemen t services, compared w ith 35.7% (n=10) of those in the low social network group, 2 (1, N=59) = 7.37, p=.007. There were no significant differences based on social netw ork in the proportion of participants who reported experiencing the other six perceived ba rriers to bereavement service utilization (see Table 4.3). As shown in Table 4.6, the mean preferen ce ratings for ten of the items pertaining to the content of potential services were rated significantly higher for the low social network group when compared with the high social network group. Services that were more highly favored by those in the low soci al network group incl uded those intended to help participants reframe or distract from the loss, to help them express or cope with their grief, or to promote self-efficacy, skill build ing, or social engagement. There were no significant differences in preference ratings fo r the 11 items related to the structure and delivery of potential services based on social network group.
67 Table 4.6. Differences in Content Prefer ences Based on Level of Social Network Low Social Network High Social Network Preference M SD M SD df t Find things to laugh about 5.10 1.80 3.39 2.29 57 -3.55** Help you stop thinking about your loss all the time 3.59 2.32 2.13 1.83 56 -2.83** Express your grief over the loss 4.34 2.11 2.87 2.11 57 -2.68* Give you tools to help with loneliness or depression 4.48 1.82 3.23 2.14 57 -2.60* Give you confidence to know you can make it through the loss 4.31 2.27 3.00 2.07 57 -2.56* Teach you how to do things your spouse used to do 4.00 2.39 2.71 1.97 57 -2.48* Relate better or participate more in lives of family members 3.11 2.10 1.97 1.45 55 -2.39* Teach you how to relax 4.66 1.97 3.29 2.42 57 -2.38* Rethink or work through challenging situations 4.62 2.11 3.33 2.31 56 -2.17* Focus on growth and things youÂ’ve learned 4.34 1.78 3.35 2.09 57 -2.04* p< .05, ** p < .01 Discussion In the current study, 45.9% of bereaved spouses used specialized professional bereavement services (i.e., support group, bereavement counseling, or
68 psychologist/psychiatrist) and 88.5% reported any formal service use (also including physician, clergy, social worker, nurse) to aid in the adjustment to bereavement. This is a higher rate of bereavement service utilizati on than has been found in previous studies (Billings & Kolton, 1999; Caserta & Lund, 1992; Cherlin et al., 2007; Levy & Derby, 1992; Provini et al., 2000; Schul z, Mendelsohn, et al., 2003). This is likely due to the fact that all spouses died wh ile enrolled in hospice, a syst em of care which includes the availability of bereavement follow-up and services for surviving family members and other loved ones. These careg ivers likely had access to in formation and services which are not routinely available to non-hospice caregivers. The most frequently used services we re provided by members of the clergy (55.7%) and by physicians (49.2%). These results are consistent with previous studies of mental health service utilization by older ad ults in highlighting th e importance of these professionals as sources of support (Cas erta & Lund, 1992; Phillips & Murrell, 1994; Robb et al., 2003). Despite thei r importance as sources of s upport to the bereaved, fewer than 25% of physicians take a course in d eath in dying while in medical school (Wass, 2004) and many express lack of confidence a nd training in the provision of bereavement care (Genevro, 2004; Joanna Briggs Collabor ating Center for Ev idence-based Multiprofessional Practice, 2006). Wide variation al so exists in the level of bereavement care training of clergy and other religious serv ice providers (Joanna Briggs Collaborating Center for Evidence-based Multi-professiona l Practice, 2006). Our findings suggest the merit of the inclusion of bereavement tr aining in the professi onal and continuing education curricula of these pr ofessionals. At a minimum, su ch training might consist of effective methods of communication with ber eaved individuals, risk factors for and
69 identification of distress both before and after the loss, the assessment of bereavement service needs, and professional collaborations and referral sources available in the effort to ensure that patients and parishioners receive comprehens ive end-of-life care, including bereavement services for survivors (Genevro, 2004). We are encouraged by our finding that few participants experienced practical barriers to bereavement service utilization. The higher rate of reported attitudinal barriers to service use, however, sugge sts the need for education a nd outreach with bereaved individuals in an effort to dispel these attitudes. The finding that the majority of participants did not see the us e in participating in bereavem ent services may reflect many participantsÂ’ resilience and ability to effectiv ely utilize support provided by the informal network in coping with and adjusting to the loss of a spouse. However, similar proportions of individuals e xperiencing low and high depressive symptomatology failed to see the utility in service use. Thus, sc reening for depressive symptoms as well as outreach and education with be reaved individuals who are experiencing high levels of depressive symptoms are important consider ations for bereavement service providers. Nearly half of participants reported that the available services did not match their needs or interests. This attitudinal barrier, reported in similar proportions by low and high depressive symptoms and social ne twork groups, suggests a need for careful attention to the preferences of potential participants in planning and delivering bereavement services, as they are likely to impact initiation an d ongoing participation. In the current study, preferen ces regarding the content, structure and delivery of bereavement services reflected a wide variety of perspectives and pa rticipantsÂ’ desire to be actively engaged in their adjustment to wi dowhood as well as in other aspects of their
70 lives. Preferences also reflected a desire fo r services designed to enhance both types of coping, loss-oriented and restor ation-oriented, included in th e dual process model of grief (Stroebe & Schut, 1999). Of note, the low so cial network group e xpressed significantly higher preferences than the high social netw ork group for services consistent primarily with restoration-oriented coping. Previous re search has suggested th at those lacking in social resources may be especially vulnerab le in bereavement given their diminished opportunities for the expression of grief, an act ivity consistent with loss-oriented coping (Raphael, Minkov, & Dobson, 2001). Our findings s uggest that social resources are also vital in providing an environment which allows for activities consistent with restorationoriented coping, including attending to life changes, establishing new roles and relationships, and distractions from grief. Future studies should examine whether th e provision of services viewed as more appealing and applicable to th e needs and preferences of be reaved individuals lead to more favorable adjustment in bereavement. For example, participants highly favored services designed to give them the opportunity to talk about their fe elings related to the loss. However, research (Stroebe, Schut, & Stroebe, 2005) suggests th at talking about the loss or oneÂ’s feelings about the loss may not facilitate adjustment to bereavement. On the other hand, the highest preference rating in th is study was given to services designed to help participants find things to laugh about. Genuine la ughter by bereaved individuals has been associated with an increase in posit ive emotions, as well as a reduction in the experience of negative emotions (Keltner & Bonanno, 1997). Therefore, it will be important to understand whether or not prefer ences lead individuals to services which will be of greatest benefit.
71 We found little relationship between social network and depressive symptoms and bereavement service utilization rates. For example, only 41.4 % of bereaved spouses with high depressive symptoms used support groups, and only 37.9% used bereavement counseling, despite these servi ces being readily available in the hospices. This suggests that many with the potential to benefit may not be accessing bereavement services. Bereaved older adults may view negative expe riences in bereavement as a normative part of growing older and may benefit from expa nded outreach and educational efforts, as well as improved access to bereavement services. While limited by sample size and crosssectional design, this study was intended to set the stage for more in-depth examina tions of the needs a nd characteristics of distressed bereaved individuals as well as individual motivations and preferences for bereavement service utilization. We view th is study as an important step in the development and evaluation of appealing bereav ement interventions that are accessible to and widely utilized by those most in need.
72 Chapter Five Conclusions In recognition of the wide variability in the experiences of bereaved family caregivers and the potential for distress, the present series of three studies examined bereavement services utilization among ber eaved family caregivers. Current service utilization as well as preferences regarding the content, structure, and delivery of potential bereavement services were exam ined. Collectively, these studies were undertaken in an effort to better understand curre nt predictors of serv ice utilization and to determine whether those experiencing the highest levels of distress were accessing bereavement services designed to help ameliorate this distress. It is also hoped that the current studies will contribute to the effort to develop and deliver bereavement services which appeal to the preferences and are utilized by those most in need. Guided conceptually and analytically by AndersenÂ’s behavioral model of health services use (Aday & Ande rsen, 1974; Andersen, 1968; Andersen & Newman, 1973), the first study examined bereavement service utilizat ion rates, the types of services used, and predictors of service utilizati on among bereaved spousal care givers using data from the Changing Lives of Older Couples (CLOC) study. Results indicated that a large proportion utilized some type of bereavement service, including talki ng with their family doctor or a member of the clergy, attending a support group, or seei ng a mental health professional. The most commonly utilized services were provided by members of the
73 clergy and by family physicians. Grief was foun d to be a need factor, with higher levels of grief playing a prominent role in talki ng with the family docto r and in the use of services in general. Higher levels of depressive symptoms re sulted in a decreased likelihood of talking with the family doctor. While need factors were not associated with use of clergy for bereavement care, racial minority status served as a predisposing characteristic and higher reli gious involvement served as an enabling resource in the utilization of services provide d by members of the clergy. Similar to the first, the second study exam ined service utilization rates, the types of services used, and predictors of servi ce utilization. This study included bereaved family caregivers, regardless of relationship, who had provi ded care to a loved one who suffered from AlzheimerÂ’s disease or dementia Data from the Resources for Enhancing AlzheimerÂ’s Caregiver Health (REACH) were utilized in th e second study. One-third of study participants utilized ber eavement services after the de ath of the care recipient. Similar rates Â– nearly one in five Â– were found for the use of support groups, bereavement counseling, and psychotropic medication. Depressive symptoms, a need factor, played a prominent role in the use of counseling (individual, family, and pastoral) and psychiatric medication, and in service use in general. Complicated grief served as a need factor in the use of a support group, with higher complicated grief associated with greater likelihood of using this service. Black caregivers were less likely and Hispanic caregivers were more likely than White caregivers to use a support group. The third study examined service utiliza tion rates, the types of services used, barriers to service use, and pr eferences regarding the content, structure, and delivery of bereavement services in a sample of bereav ed caregivers whose spouses died while under
74 the care of one of three local hospice organi zations. Hospice is a relatively service rich environment in which programs are mandated to offer bereavement services for family members for one year after the death of the hospice patient. Thus in hospice services lack of service use is not due to lack of availability. The majo rity of participants utilized bereavement services, including support groups and counseling, as well as talking with a psychologist or psychiatrist, a physician, member of the cler gy, social worker, or nurse for help in coping with the death of their s pouse. The most commonly used services were provided by members of the clergy and physicians Barriers were primarily attitudinal, rather than practical, in nature. Participan ts preferred services designed to give them tools to reframe their loss and adjust to acco mpanying changes, to help them engage with the community, and to manage stress. They favored interactive services over those offering little opportunity to engage with ot her bereaved individua ls or with service providers. A significantly high er proportion of participants in the low social network group utilized bereavement counseling than di d those in the high social network group, a group more likely to report that they di d not recognize the utility of bereavement services. Services that were more highly favored by those in the low social network group and by those with higher levels of depr essive symtpomatology were those intended to help participants reframe or distract from the loss. Those in the low social network group also favored services design ed to help them express or cope with their grief or to promote self-efficacy, skill-buildi ng, or social engagement. Collectively, the studies in this disser tation suggest that need factors play a prominent role in driving the utilization of be reavement services and that the experience of grief and depressive symptoms is associ ated with seeking out and accessing supportive
75 services in bereavement. However, ther e is still a significant gap, with many highly depressed bereaved caregivers not using services More research is necessary in order to more clearly describe the full range of n eed factors and to understand how need is determined and ultimately results in servic e utilization. While social resources did not play an enabling role in the use of servi ces, they were associated with preferences regarding the content of be reavement services. These findings, combined with recommendations that bereavement services target those at highest risk for and experiencing distress in bereavement (Genev ro, 2004; Schut & Stro ebe, 2005; Schut et al., 2001), have some important implications. First, outreach with and routine screening to identify individuals with el evated depressive symptoms, complicated grief, and other need factors and for decrements in social resources are needed. Findings regarding the importance of physicians and members of the clergy in the lives of bereaved family caregivers suggest that these professionals are critical part ners in these outreach and screening efforts. In addition, services should be designed and delivered in a manner consistent with preferences and should take leve l of social resources into consideration. Tailoring interventions to the needs and preferences of those in greatest need will maximize the likelihood of their utilization in facilitating ad justment to the bereavement transition and will contribute to ensuring the highest possible quality of life for bereaved family caregivers. Limitations The research described in this dissert ation suffers from some limitations which should be noted. While the datasets used in the secondary data anal ysis portions of the current project were chosen specifically because they best allowed for examination of the
76 research questions, they nonetheless omitted so me data which would likely strengthen the explanatory power of the model. First, th e CLOC and REACH datasets did not contain information about whether or not the care r ecipient died while under hospice care. The provision of hospice care is likely to impact knowledge of available bereavement services, the lack of which may be a significan t barrier to bereavemen t service utilization as it was in the use of long term care (Bra dley et al., 2002). Fu ture research should address this in an effort to understand the ro les played by knowledge of available services and by hospice use in general. In both the CLOC and REACH studies, information about th e utilization of bereavement services was collect ed retrospectively and was se lf-reported by participants. This is a common approach to the study of health behavior and the only option for the collection of data pertaining to utilizati on of bereavement services which are not reimbursable under Medicare, Medicaid, or other insurance plans (e.g., bereavement support group, clergy/pastoral counseling). However, st udies indicate that when compared with information about actual ut ilization as obtained from administrative records, self-reported health se rvices utilization is freque ntly underestimated (Dunlop et al., 2002). The available data offered little insigh t into the specifics of the bereavement services of interest. For example, informa tion was not available a bout the credentials of service providers, the durati on of the services, or the th eoretical foundations of the services under examination. In addition, limitations in the av ailable data meant that we could only focus on the first reported use of be reavement services, rather than on patterns of use over time. Attention to these and othe r details in future st udies will help us to
77 better understand preferences, motivations, a nd choices regarding bereavement service utilization as well as the specific features of services which lead to positive outcomes for diverse groups of bereaved individuals. Finally, the small sample size and st udy design in study three of the present dissertation restricted the t ypes of statistical analyses we could employ to examine service use, preferences, and ba rriers. While practical consid erations drove the choices in study design, future research should include larger, more diverse samples as well as multiple waves of data collection in an effort to understand patterns of use and changes in preferences over time. Future Directions This dissertation hi ghlights some important next st eps in the study of bereavement interventions and services. Firs t, more research is needed in an effort to more fully identify and describe factors associated w ith bereavement service utilization. While current study findings pointed primarily to need factors, future efforts should aim to determine the roles played by factors which may predispose individuals to or enable bereavement service utilization. In studying f actors associated with bereavement service use, careful consideration should be given to their mutability, or the degree to which these factors are subject to change or alteration through polic y changes or other intervention (Andersen, 1995). This may enable better targeting of se rvices to those in greatest need or increase access or the likelihood of service use among bereaved individuals who were pr eviously underserved. Researchers face a remarkable challenge in identifying and developing culturallyrelevant bereavement interventi ons. Perhaps the most promising approach is to develop
78 highly targeted bereavement services available both prior to and afte r the death (Schulz, Burgio, et al., 2003). It is also important no t to use race or ethnicity as proxies for cultural values (Dilwor th-Anderson et al., 2002), but rath er to carefully consider the many cultural and individual characteristic s and circumstances which impact the bereavement experience and to make modificatio ns to interventions in order to maximize their relevance to the particular individual or group targeted (Burgio, Stevens, Guy, Roth, & Haley, 2003). More research with diverse samples of be reaved individuals is thus called for. This research was conducted in an effort to learn about the char acteristics of those who utilize bereavement services and to aid in refining policy and interventions with the goal of helping those most in need. Future work is needed to understand the mechanisms behind predictors of bereavemen t service utilization and to co ntribute to the development of bereavement interventions which are a ppropriately targeted, lend themselves to methodologically sound evaluation of their efficacy, and are a ppealing to individuals who are distressed in bereavement. Changes to policy and practice should be informed by preferences regarding the conten t, structure, and delivery of bereavement services as well as by their demonstrated efficacy. We must first determine whether or not preferences lead bereaved individuals to select the most efficacious services.
79 References Aday, L. A., & Andersen, R. (1974). A framework for the study of access to medical care. Health Services Research, 9 208-220. Aday, L. A., & Awe, W. C. (1997). Health serv ice utilization models. In D. S. Gochman (Ed.), Handbook of health behavior research I: Personal and social determinants (pp. 153-172). New York: Plenum Press. Allison, P. D. (1995). Survival analysis using the SA S system: A practical guide Cary, NC: SAS Institute. Allumbaugh, D. L., & Hoyt, W. T. (1999). E ffectiveness of grief counseling: A metaanalysis. Journal of Counseling Psychology, 46 370-380. Aloisi, P. (2001). Instant Drug Index. Williston, VT: Blackwell Science, Inc. American Psychological Association. (2002). Guidelines on multi-cultural education, training, research, practice, and or ganizational change for psychologists Washington, DC: Author. Andersen, R. (1968). A behavioral model of familiesÂ’ us e of health services (Research Series 25) Chicago: The University of Chica go Center for Health Administration Studies. Andersen, R. (1976). A framework for crossnational comparisons of health services systems. In M. Pflanz & E. Schach (Eds.), Cross-national sociomedical research: Concepts, methods, practice (pp. 25-35), Stuttgart, Germany: Georg Thieme. Andersen, R. M. (1995). Revisiting the beha vioral model and access to medical care: Does it matter? Journal of Health and Social Behavior, 36 1-10. Andersen, R., Kravits, J., & Anderson, O. (1975). Equity in health services: Empirical analyses in social policy. Boston: Ballinger. Andersen, R., & Newman, J. F. (1973). Societal and individual determinants of medical care utilization in the United States. Milbank Memorial Fund Quarterly, 51 95124.
80 Aneshensel, C. S., Botticello, A. L., & Ya mamoto-Mitani, N. (2004). When caregiving ends: The course of depressive symptoms after bereavement. Journal of Health and Social Behavior, 45 422-440. Aneshensel, C. S., Pearlin, L. I., Mullan, J. T., Zarit, S. H., & Whitlatch, C. J. (1995). Profiles in caregiving: The unexpected career San Diego, CA: Academic Press. Arling, G. (1985). Interaction effects in a mu ltivariate model of physician visits by older people. Medical Care, 23 361-371. Barry, L. C., Kasl, S. V., & Prigerson, H. G. (2002). Psychiatric disorders among bereaved persons. The role of perceive d circumstances of death and preparedness for death. American Journal of Geriatric Psychiatry 10 447Â–457. Bambauer, K. Z., & Prigerson, H.G. (2006). The Stigma Receptivity Scale and its association with mental health serv ice use among bereaved older adults. Journal of Nervous and Mental Disease, 194 139-141. Beem, E. E., Maes, S., Cleiren, M., Schut H., & Garssen, B. (2000). Psychological functioning of recently bereaved, middl e-aged women: the first 13 months. PsychologicalReports, 87, 243Â–254. Bergman, E. J., & Haley, W. E. (2008). Depressive symptoms, social network, and bereavement service utilization and prefer ences among spouses of former hospice patients Manuscript submitted for publication. Bergman, E. J., Haley, W. E., & Small, B. J. (2008). The role of grief and depressive symptoms in the utilization of bereavement services. Manuscript submitted for publication. Berkman, L. (1979). Social networ ks, host resistance and mortality. American Journal of Epidemiology, 109 186-203. Billings, J. A., & Kolton, E. (1999). Family satisfaction and bereavement care following death in the hospital. Journal of Palliative Medicine, 2 33-49. Bodnar, J. C., & Kiecolt-Glaser, J. K. ( 1994). Caregiver depression after bereavement: Chronic stress isnÂ’t over when itÂ’s over. Psychology and Aging, 9, 372-380. Boelen, P. A., de Keijser, J., van den Hout, M. A., & van den Bout, J. (2007). Treatment of complicated grief: A comparison between cognitive behavioral therapy and supportive counseling. Journal of Consulting and Clinical Psychology, 75, 277-284.
81 Bonanno, G. A. (2001). Grief a nd emotion: A social-functional perspective. In M. S. Stroebe & R. O. Hansson (Eds.), Handbook of bereavement research: Consequences, coping, and care (pp. 493-515). Washington, DC: American Psychological Association. Bonanno, G. A., & Kaltman, S. (1999). To ward an integrative perspective on bereavement. Psychological Bulletin, 125 760-776. Bonanno, G. A., & Kaltman, S. (2001). The varieties of grief experience. Clinical Psychology Review, 21, 705-734. Bonanno, G. A., Moskowitz, J. T., Papa, A., & Folkman, S. (2005). Resilience to loss in bereaved spouses, bereaved pa rents, and bereaved gay men. Journal of Personality and Social Psychology, 88, 827-843. Bonanno, G. A., Wortman, C. A., Lehman, D. R., Tweed, R. G., Haring, M., Sonnega, J., et al. (2002). Resilience to loss and ch ronic grief: A prospective study from preloss to 18-months postloss. Journal of Personality and Social Psychology 83 1150-1164. Bonanno, G. A., Wortman, C. B., & Nesse, R. M. (2004). Prospective patterns of resilience and maladjustment during widowhood. Psychology and Aging, 19 260-271. Bradley, E. H., McGraw, S. A., Curry, L., Buck ser, A., King, K. L., Kasl, S. V., et al. (2002). Expanding the Andersen model: the role of psychosocial factors in longterm care use. Health Services Research, 37, 1221-1242. Burgio, L., Stevens, A. B., Guy, D., Roth, D. L., & Haley, W. E. (2003). Impact of two interventions on White a nd African-American family caregivers of individuals with dementia. The Gerontologist, 43 568-579. Burton, L. C., Zdaniuk, B., Schulz, R., Jacks on, S., & Hirsch, C. (2003). Transitions in spousal caregiving. The Gerontologist, 43 230-241. Byrne, G. J. A., & Raphael, B. (1997). The psychological symptoms of conjugal bereavement in elderly men over the first 13 months. International Journal of Geriatric Psychiatry 12 241Â–251. Cantor, A. B. (2003). Survivor analysis techniques for medical research (2nd ed.). Cary, NC: SAS Institute, Inc. Caserta, M. S., & Lund, D. A. (1992). Bereaved older adults who seek early professional help. Death Studies, 16 17-30.
82 Caserta, M. S., & Lund, D. A. (1993). Intrap ersonal resources and the effectiveness of self-help groups for bereaved older adults. The Gerontologist, 33 619-629. Chentsova Dutton, Y., & Zisook, S. (2005). Adaptation to bereavement. Death Studies, 29, 877-903. Cherlin, E. J., Barry, C. L., Prigerson, H. G., Schulman-Green, D., Johnson-Hurzeler, R., Kasl, S. V., et al. (2007). Bereavement se rvices for family ca regivers: How often used, why, and why not. Journal of Palliative Medicine, 10 148-158. Christakis, N., & Iwashyna, T. (2003). The hea lth impact of health care on families: A matched cohort study of hospice use by decedents and mortality outcomes in surviving, widowed spouses. Social Science & Medicine, 57, 465-475. Conrad, K., Hughes, S., & Wang, S. (1992). Prog ram factors that infl uence utilization of adult day care. Health Services Research, 27 481-502. Coon, D. W., Rubert, M., Solano, N., Mausbac h, B., Kraemer, H., Argulles, T., et al. (2004). Well-being, appraisal, and coping in Latina and Caucasian female dementia caregivers: Findings from the REACH study. Aging and Mental Health, 8, 330-345. Coulton, C., & Frost, A. (1982). Use of so cial and health services by the elderly. Journal of Health and Social Behavior 23, 330-339. Counte, M., & Glandon, G. (1991). A panel study of life stress, social support, and the health services utili zation of older persons. Medical Care, 29, 348-361. Crabb, R., & Hunsley, J. (2006). Utilization of mental health care services among older adults with depression. Journal of Clinical Psychology, 62, 299-312. Dilworth-Anderson, P., Williams, I. C., & Gibson, B. E. (2002). Issues of race, ethnicity, and culture in caregiving research : A twenty year review (1980-2000). The Gerontologist, 42 237-272. Dunlop, D. D., Manheim, L. M., Song, J ., & Chang, R. W. (2002). Gender and ethnic/racial disparities in health care utilization among older adults. Journal of Gerontology: Social Sciences, 57B S221-S233. Evashwick, C., Rowe, G., Diehr, P., & Bran ch, L. (1984). Factors explaining the use of health care services by the elderly. Health Services Research, 19 357-382. Eve, S. (1988). A longitudinal study of use of health care services among older women. Journal of Gerontology, 43 M31-M39.
83 Forte, A. L., Hill, M., Pazder, R., & Feudtne r, C. (2004). Bereavement care interventions: A systematic review. BMC Palliative Care, 3. Fortner, B.V. (1999). The effectiveness of grief counseling and therapy: A quantitative review. Dissertation Abstracts International, 60 4221B. Freedman, V. (1993). Kin and nur sing home lengths of stay: A backward recurrence time approach. Journal of Health and Social Behavior, 34 138-152. Gallagher-Thompson, D., Haley, W. E., Guy, D ., Rubert, M., Arguellas, T., Tennstedt, S., et al. (2003). Tailori ng psychological interventi ons for ethnically diverse dementia caregivers. Clinical Psychology: Science and Practice, 10, 423-438. Gaugler, J. E., Anderson, K. A., Zarit, S. H ., & Pearlin, L. I. (2004). Family involvement in the nursing home: Effects on stress and well-being. Aging & Mental Health, 8 65-75. Genevro J. L. (2004). Report on be reavement and grief research. Death Studies, 28, 491575. Ginsberg, A., & Goldstein, S. (1974). Age bias in referral for psyc hological consultation. Journal of Gerontology, 29, 410-415. Goldman, N., Korenman, S., & Weinstein, R. (1995). Marital status and health among the elderly. Social Science & Medicine, 40, 1717-1730. Goodkin, K., Baldewicz, T., Blaney, N., Asthan a, D., Kumar, M., Shapshak, P., et al. (2001). Physiological effects of ber eavement and bereavement support group interventions. In Stroebe, M. et al. (Eds.), Handbook of bereavement: Consequences, coping, and care (pp. 671-703). Washington, DC: American Psychological Association. Haley, W. E., Gitlin, L. N., Wisniewski, S. R., Feeney Mahoney, D., Coon, D. W., Winter, L., et al. (2004). Well-being, appraisal, and coping in African-American and Caucasian dementia caregiver s: Findings from the REACH study. Aging and Mental Health, 8 316-329. Hansson, R., O. & Stroebe M. S. (2003). Grie f, older adulthood. In T. P. Gullotta, & M. Bloom (Eds.) The encyclopedia of primary prevention and health promotion (pp. 515-521). Boston: Kluwer. Harris, K. M., Edlund, M. J., & Larson, S. (2005). Racial and ethni c differences in the mental health problems and use of mental health care. Medical Care, 43, 775-784.
84 Hurley, A. C., & Volicer, L. (2002). Alzheime r Disease: "It's okay, Mama, if you want to go, it's okay". Journal of the American Medical Association, 288 2324-2331. Joanna Briggs Collaborating Center for Ev idence-based Multi-professional Practice. (2006, January). Literature review on berea vement and bereavement care. Aberdeen: The Joanna Briggs Institute. Jordan, J. R., & Neimeyer, R. A. (2003). Does grief counseling work? Death Studies, 27, 765-786 Kang, S. H., Wallace, N. T., Hyun, J. K., Morri s, A., Coffman, J., & Bloom, J. R. (2007). Social networks and their relationshi p to mental health service use and expenditures among Medicaid beneficiaries. Psychiatric Services, 58, 689-695. Kaprio, J., Koskenvuo, M., & Rita, H. ( 1987). Mortality after bereavement: A prospective study of 95,647 widowed persons. American Journal of Public Health, 77 283-287. Kart, C. S. (1991). Variation in long-te rm care services use by aged blacks. Journal of Aging and Health, 3, 511-526. Kato, P. M., & Mann, T. (1999). A synthesi s of psychological interventions for the bereaved. Clinical Psychology Review, 19, 275-296. Keltner, D., & Bonanno, G. A. (1997). A study of laughter and disso ciation: Distinct correlates of laughter and smiling during bereavement. Journal of Personality and Social Psychology, 73, 687-702. Kessler, R., Brown, R., & Brom an, C. (1981). Sex differenc es in psychiatric help seeking: Evidence from four large scale surveys. Journal of Health and Social Behavior, 22 49-64. Krause, N. (1995). Negative interaction and satisfaction with social support among older adults. Journal of Gerontology: Psychological Sciences, 50B 59-73. Krause, N., & Markides, K. (1990). Meas uring social support among older adults. International Journal of Aging and Human Development, 30, 37-53. Kung, H. C., Hoyert, D. L., Xu, J. Q., & Mu rphy, S. L. (2008). Deaths: Final data for 2005. National vital statistics reports, 56 Retrieved April 20, 2008, from National Center for Health Statistics Web site: http://www.cdc.gov/nchs/data/nvsr/nvsr56/nvsr56_10.pdf
85 Larson, D. G., & Hoyt, W. T. (2007a). Th e bright side of grief counseling: Deconstructing the new pessimism. In K. J. Doka (Ed.), Living with grief: Before and after the death (pp. 157-174). Washington DC: Hospice Foundation of America. Larson, D. G., & Hoyt, W. T. (2007b). What has become of grief counseling? An evaluation of the empirical f oundations of the new pessimism. Professional Psychology: Research and Practice, 38 347-355. Levy, L. H., & Derby, J. F. (1992). Bereav ement support groups: Who joins; who does not; and why. American Journal of Community Psychology, 20, 649-662. Lewinsohn, P. M., Seeley, J. R., Roberts, R. E., & Allen, N. B. (1997). Center for Epidemiologic Studies-Depression Scale (C ES-D) as a screening instrument for depression among community-residing older adults. Psychology and Aging, 12 277-287. Lubben, J. (1988). Assessing social networks among elderly populations. Family and Community Health, 11 42-52. McHorney, C. A., & Mor, V. (1988). Predictors of bereavement depression and its health services consequences, Medical Care, 26, 882-893. McIntosh, D. N., Silver, R., & Wortman, C. B. (1993). ReligionÂ’s role in adjustment to a negative life event: Coping with the loss of a child. Journal of Personality and Social Psychology, 65 812-821. Miller, B., & McFall, S. (1991). The effect of caregiverÂ’s burd en on change in frail older personÂ’s use of formal helpers. Journal of Health and Social Behavior, 32 165179. Mittelman, M. S. (2008). Psychosocial interven tion research: Challenge s, strategies and measurement issues. Aging & Mental Health, 12 1-4. Mojtabai, R., & Olfson, M. (2006). Treatment seeking for depression in Canada and the United States. Psychiatric Services, 57 ,631Â–639. Mui, A. C., & Burnette, D. (1994). Long-te rm care service use by frail elders : Is ethnicity a factor? The Gerontologist, 34, 190-198. Mutran, E., & Ferraro, K. (1988). Medical need and use of services among older men and women. Journal of Gerontology, 43 S162-S171.
86 National Alliance for Caregi ving and AARP (June, 1997). Family Caregiving in the U.S.: Findings from a National Survey National Alliance for Caregiving, Bethesda, MD and AARP, Washington, DC. Neimeyer, R. (2000). Searching for the mean ing of meaning: Grief therapy and the process of reconstruction. Death Studies, 24, 531-558. Parkes, C. M. (1998). Editorial. Bereavement Care, 17 18. Pearlin, L. I. & Aneshensel, C. S. (1 994). Caregiving: The unexpected career. Social Justice Research, 7, 373-390. Pearlin, L. I., Menaghan, E. G., Lieberman, M. A., & Mullan, J. T. (1981). The stress process. Journal of Health & Social Behavior, 22, 337-365. Pescosolido, B. A., Wright, E. R., Alegra, M., & Vera, M. (1998). Social networks and patterns of use among the poor with ment al health problems in Puerto Rico. Medical Care, 36, 1057-1072. Phillips, M. A., & Murrell, S. A. (1994). Im pact of psychological and physical health, stressful events, and social support on subsequent mental health help seeking among older adults. Journal of Consulting and Clinical Psychology, 62 270-275. Pinquart, M., & Srensen, S. (2003). Differen ces between caregivers and noncaregivers in psychological health and phys ical health: A meta-analysis. Psychology and Aging, 18, 250-267. Pinquart, M., & Srensen, S. (2005). Ethnic differences in stressors, resources, and psychological outcomes of family caregiving: A meta-analysis. The Gerontologist, 45 90-106. Prigerson, H. G., Bierhals, A. J., Kasl, S. V ., Reynolds, C. F. III, Shear, M. K., Day, N., et al. (1997). Traumatic grie f as a risk factor for me ntal and physical morbidity. American Journal of Psychiatry, 54, 617-623. Prigerson, H. G., Frank, E., Kasl, S. V., Re ynolds, C. F. III, Anderson, B., Zubenko, G. S. et al. (1995). Complicated grief and ber eavement-related depression as distinct disorders: Preliminary empirica l validation in elderly bereaved spouses. American Journal of Psychiatry 152, 22Â–30. Prigerson, H. G., Maciejewski, P. K., Reynolds, C. F. III, Bierhals, A. J. Newsom, J. T., Faciszka, A., et al. (1995). Inventory of Complicated Grief: A scale to measure maladaptive symptoms of loss. Psychiatry Research, 59 65-79.
87 Prigerson, H. G., Silverman, G. K., Jacobs, S. C., Maciejewski, P. K., Kasl, S. V., & Rosenbeck, R. A. (2001). Disability, trau matic grief, and th e underutilization of health services. Primary Psychiatry, 8, 61-69. Provini, C., Everett, J. R., & Pfeffer, C. R. (2000). Adults mourning suicide: Selfreported concerns about bereavement, needs for assistance, and help-seeking behavior. Death Studies, 24 1-19. Rabiner, D. (1992). The relationship between program participation, use of formal inhome care, and satisfaction with care in an elderly population. The Gerontologist, 32 805-812. Radloff, L. S. (1977). The CESD scale: A self-report depressi on scale for research in the general population. Applied Psychological Measures, 1, 385-401. Raphael, B., Middelton, W., Martinek, N., & Misso, V. (1993). Counseling and therapy of the bereaved. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of Bereavement Research : Consequences, Coping and Care (pp. 427453). New York: Cambridge. Raphael, B., Minkov, C., & Dobson, M. (2001). Psychotherapeutic and pharmacological intervention for bereaved persons. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement: Co nsequences, coping, and care (pp. 587-612). Washington, DC: American Psychological Association. Rhodes, A. E., Goering, P. N., To, T., & Williams, J. I. (2002). Gender and outpatient mental health service use. Social Science & Medicine, 54 1-10. Rhodes, A., Jaakkimainen, R. L., Bondy, S ., & Fung, K. (2006). Depression and mental health visits to physiciansÂ—a prospective reco rds-based study. Social Science & Medicine, 62 828Â–834. Robb, C., Haley, W. E., Becker, M. A., Poli vka, L. A., & Chwa, H.-J. (2003). Attitudes towards mental health care in young er and older adults: Similarities and differences. Aging & Mental Health, 7, 142-152. Rosner, T., Namazi, K., & Wykle, M. (1988) Physician use among the old-old: Factors affecting variability. Medical Care, 26 982-991. Schulz, R., & Beach, S. R. (1999). Caregivi ng as a risk factor for mortality: The Caregiver Health Effects Study. JAMA, 282 2215-2219. Schulz, R., Beach. S. R., Lind, B., Martire, L. M., Zdaniuk, B., Hirsch, C., et al. (2001). Involvement in caregiving and adjustment to death of a spouse: Findings from the Caregiver Health Effects Study. JAMA, 285 3123-3129.
88 Schulz, R., Burgio, L., Burns, R., Eisdorfer, C., Gallagher-Thompson, D., Gitlin, L. N., et al. (2003). Resources for Enhancing Al zheimerÂ’s Caregiver Health (REACH): Overview, site-specific outcomes and future directions The Gerontologist, 43 514-520. Schulz, R., Mendelsohn, A. B., Haley, W. E., Mahoney, D., Allen, R. S., Zhang, S., et al. (2003). End of life care and the effects of bereavement among family caregivers of persons with dementia. New England Journal of Medicine 349, 1936-1942. Schulz, R., Newsom, J. T., Fleissner, K., D ecamp, A. R., & Nieboer, A. P. (1997). The effects of bereavement after family caregiving. Aging & Mental Health, 1, 269282. Schut, H., & Stroebe, M. S. (2005). Interventi ons to enhance adaptation to bereavement. Journal of Palliative Medicine, 8 S140-147. Schut, H., Stroebe, M., van den Bout, J., & Terheggen, M. (2001). The efficacy of bereavement interventions: Determining who benefits. In Stroebe, M. et al. (Eds.), Handbook of bereavement: Conse quences, coping, and care (pp. 705-737). Washington, DC: American Psychological Association. Snider, E. (1980). Factors influencing h ealth service knowledge among the elderly. Journal of Health and Social Behavior, 21 371-377. Spielberger, C. D. (1979). Preliminary manual for the Stat e-Trait Persona lity Inventory Unpublished manual, University of South Florida, Tampa. Spielberger, C. D., Gorsuch, R. L., Lushene, P. R., Vagg, P. R., & Jacobs, G. A. (1983). State-Trait Anxiety Inventory for Adults (STAIS-AD) Manual. Palo Alto, CA: Consulting Psychologists Press, Inc. Stroebe, M. S., & Schut, H. (1999).The dual process model of coping with bereavement: Rationale and description. Death Studies, 23, 197-224. Stroebe, W., Schut, H., & Stroebe, M. S. ( 2005). Grief work, disclo sure and counseling: Do they help the bereaved? Clinical Psychology Review, 25 395-414. Stroebe, M., Schut, H., & Stroebe, W. (2007). Heath outcomes of bereavement. Lancet, 370 1960-1973. Stroebe, W., & Stroebe M., (1987). Bereavement and health New York: Cambridge University Press.
89 Tabachnick, B. G., & Fidell, L. S. (2001). Using Multivariate Statistics (4th ed.). Boston: Allyn & Bacon. Toseland, R. W., McCallion, P., Gerber, T., & Ba nks, S. (2002). Predictors of health and human services use by persons with dementia and their family caregivers Social Science and Medicine, 55 1255-1266. U.S. Census Bureau. (2005). 65+ in the United States: 2005. Washington, DC: U.S. Government Printing Office. U. S. Department of Health and Human Services, Administration on Aging (May 2003). Family Caregiving Fact Sheet. Retrieved November 1, 2004. Website: www.aoa.gov/press/aom/May_2003/me dia/fact_sheets/National%20 Family%20Caregiver%20Support%20Program.pdf. Vogel, D. L., Wade, N. G., Wester, S. R., Larson, L., & Hackler, A. H. (2007). Seeking help from a mental health professiona l: The influence of oneÂ’s social network. Journal of Clinical Psychology, 63, 233-245. Wallace, S. P., Levy-Storms, L., Kingston, R. A., & Andersen, R. M. (1998). The persistence of race and ethnicit y in the use of long-term care. Journal of Gerontology: Social Sciences, 53B S104-S112. Wan, T. (1989). The effect of managed care on health services use by dually eligible elders. Medical Care, 27 983-1001. Wass, H. (2004). A perspective on the current state of death education. Death Studies, 28 289-308. Williams, S. W., & Dilworth-Anderson, P. (2002). Systems of social support in families who care for dependent Af rican American elders. The Gerontologist, 42 224-236. Wisniewski, S. R., Belle, S. H., Coon, D. W ., Marcus, S. M., Ory, M. G., Burgio, L. D., et al. (2003). The Resources for Enha ncing AlzheimerÂ’s Caregiver Health (REACH): Project design a nd baseline characteristics. Psychology and Aging, 18, 375-384. Zhang, B., El-Jawahri, A., & Prigerson, H. (2006). Update on bereavement research: Evidence-based guidelines for the dia gnosis and treatmen t of complicated bereavement Journal of Palliative Medicine, 9 1188-1203. Zhang, B., Mitchell, S. L., Bambauer, K. Z., Jones, R., & Prigerson, H. G. (2008). Depressive symptom traject ories and associated risks among bereaved Alzheimer disease caregivers. American Journal of Ge riatric Psychiatry, 16, 145-155.
90 Zisook, S., Paulus, M., Shuchter, S. R., & Judd, L. L. (1997). The many faces of depression following spousal bereavement. Journal of Affective Disorders, 45, 8595. Zisook, S., & Shuchter, S. R. (1985). Ti me course of spousal bereavement. General Hospital Psychiatry 7, 95Â–100. Zisook, S., & Shuchter, S. (2001). Treatment of the depressions of bereavement. American Behavioral Scientist, 44 ,782-792.
92 Appendix A: Curriculum Vitae CURRICULUM VITAE Elizabeth J. Bergman, M.A., Ph.D. Candidate School of Aging Studies University of South Florida 4202 E. Fowler Avenue, MHC 1308 Tampa, FL 33620 Tel: (813) 974-1337 Fax: (813) 974-9754 firstname.lastname@example.org Education 2003-present Doctoral Candida te, Ph.D. in Aging Studies University of South Florida Advisor: William E. Haley, Ph.D. Dissertation: Service Utilization amo ng Bereaved Spouses and Family Caregivers 1996-1998 M.A., Gerontology, University of South Florida, Tampa, FL 1991-1996 B.A., Psychology, Auburn University, Auburn, AL Research Interests Family caregiver and bereavement interventions and service utilization; end-of-life issues; hospice and palliative ca re; stress and coping in the co ntext of the aging family; gender and multicultural aspects of aging Research Experience 2007-present Project Manager Understanding Disparities in Service Use among White, African American, and Hispanic Caregivers PI: William E. Haley, Ph.D. Florida AlzheimerÂ’s Disease Research Center 2005-2006 Graduate Research Assistant Brent Small, Ph.D. School of Aging Studies, University of South Florida 2004-2005 Principal Investigator Hospice Bereavement Service Utilization and Preferences among Spouses in Three Hospices
93 Appendix A (continued) Center for Hospice, Palliativ e Care, and End-of-Life Studies, University of South Florida 2004-2005 Graduate Research Assistant Center for Hospice, Palliativ e Care, and End-of-Life Studies, University of South Florida 2004 Project Manager Physical and Mental Well-Bei ng of Older Adults in Assisted Living Facilities PI: Yuri Jang, Ph.D. Department of Aging and Mental Health, Florida Mental Health Institute, University of South Florida 2003-2004 Graduate Research Assistant Quality of Life in Older Breast Cancer Survivors (NIA #1R03AG21210-01) PI: William E. Haley, Ph.D. School of Aging Studies, University of South Florida Teaching Experience (Univ ersity of South Florida) Spring 2008 Instructor, Directed Readings in Family Caregiving (GEY 4900) Fall 2007 Instructor, Introduc tion to Gerontology (GEY 2000) Summer 2007; Summer 2006; Instruct or, Death and Dying (GEY 4641) Spring 2006; Fall 2004 Spring 2007; Fall 2005; Graduate Teachi ng Assistant, Sociocultural Aspects of Aging Spring 2005 (GEY 3625) Fall 2006 Instructor, Soci ocultural Aspects of Aging (GEY 3625) Instructor, Directed Readings in Family Caregiving (GEY 4900) 2003-2007 Guest Lecturer, The Life Cycle (DEP 2004), Psychology of Aging (GEY 4612), Women and Aging (GEY 4935)
94 Appendix A (continued) Professional Experience 2002-2003 Foster Grandparent Program Director Seniors in Service of Tampa Bay, Inc., Tampa, FL Supervised employees and stipended volunteers; wrote grants (resulting in $300,000 in grant awards); managed contracts; conducted ongoing program evaluation; reported pr ogram outcomes; conducted trainings and orientation sessions; oversaw monthly payroll of $45,000. 2000-2002 Program Coordinator Seniors in Service of Tampa Bay, Inc., Tampa, FL Supervised, trained, and placed Bill Payer Program and Senior Companion Program volunteers; c onducted in-home client assessments; reported program outcomes. 1998-2000 Information and Referral Specialist West Central Florida Area Agency on Aging, Inc., Tampa, FL Provided information and referral service to seniors and their caregivers; maintained I&R database; managed Gatekeeper Program; coordinated activities in celebration of Older Americans Month; coordinated Â“My Favorite Older PersonÂ” essay contest in Polk and Hillsborough County public school s; Ceridian Employee Assistance Program counselor. 1997-1998 Director of Activities Palm Terrace Resident Care, Inc. (assisted living facility) Trained and supervised activities department staff and volunteers; implemented and maintained activiti es programs in assisted living and specialized memory support units. Grants and Fellowships Principal Investigator, Â“Hos pice Bereavement Service Util ization and Preferences among Spouses in Three Hospices,Â” Center for Hospice, Palliative Ca re, and End-of-Life Studies, University of S outh Florida, 2004-2005 ($12,000). Graduate Fellowship, Institute on Aging, University of South Florida, 2003-2004 ($15,000 & Graduate Tuition Waiver). Peer Reviewed Journal Publications Haley, W. E., Bergman, E. J., Roth, D. L., McVie, T., Gaugler, J. E., & Mittelman, M. S. (in press). Long-term effects of ber eavement and caregiver intervention on caregiver depressive symptoms. The Gerontologist.
95 Appendix A (continued) Jang, Y., Bergman, E., Schonfeld, L., & Molinar i, V. (2007). The mediating role of health perceptions in the relation betw een physical and mental health: A study of older residents in assisted living facilities. Journal of Aging and Health, 19, 439452. Jang, Y., Bergman, E., Schonfeld, L., & Molin ari, V. (2006). Depr essive symptoms among older residents in a ssisted living facilities. International Journal of Aging & Human Development, 63 299-315. Book Chapters Allen, R. S., Haley, W. E., Roff, L. L., Sc hmid, B., & Bergman, E. J. (2006). Responding to the needs of caregivers near the end of life: Enhancing benefits and minimizing burdens. In J. L. Werth & D. Blevins (Eds.), Psychosocial issues near the end of life: A resource for professional care providers. Washington, DC: American Psychological Association. Papers Under Review or In Preparation Bergman, E. J., Haley, W. E., & Small, B. J. (under review). The role of grief and depressive symptoms in the ut ilization of bereavement services. Bergman, E. J., & Haley, W. E. (under revi ew). Depressive symptoms, social network, and bereavement service utilization a nd preferences among spouses of former hospice patients. Bergman, E. J., & Perkins, E. A. (in prepar ation). Older womenÂ’s experience of surviving breast cancer. Bergman, E. J., Haley, W. E., & Small, B. J. (in preparation). Outcomes and Service Utilization among Bereaved Dementia Caregivers: Findings from the REACH Study. Bergman, E. J., & Polubinski, J. (in prepara tion). Access to the hospi ce care option at the end of life: A review of reported barriers. Bergman, E. J. (in preparation). Grief and spousal loss: A qualitative study of former hospice caregivers. Invited Presentations Bergman, E. J. (2007, July). Coping. Invited workshop for the AlzheimerÂ’s Association Florida Gulf Coast ChapterÂ’s Early Al zheimerÂ’s Support and Education Program.
96 Appendix A (continued) Bergman, E. J., & Haley, W. E. (2005, November). Hospice bereavement service utilization and preference s among spouses in hospice. Invited presentation at the Center for Hospice, Palliative Care, and End-of-Life Studies, University of South Florida, Tampa, FL. Conference Presentations Bergman, E. J., Perkins, E. A., & Haley, W. E. (2007, November). Patient perspectives on breast cancer: The qualitative experi ence of older breast cancer survivors. Poster presentation at the 60th Annual Scientific Meeti ng of the Gerontological Society of America, San Francisco, CA. Bergman, E. J., & Haley, W. E. (2006, Nove mber). Bereavement service utilization and preferences of bereaved spousal care givers. Poster presentation at the 59th Annual Scientific Meeting of the Gerontolog ical Society of America, Dallas, TX. Kuchta, L., Bergman, E. J., & Davis, B. ( 2006, April) W.O.M.E.N. working for women. Panel presentation at the Many Fl oridas: Women Envisioning Change Conference, Tampa, FL. Bergman, E. J. (2006, April). Breaking down barriers to the hospice care option: Opportunities for education & advoc acy. Poster presentation at the 28th Annual Scientific Meeting of the Associat ion for Death Education and Counseling, Tampa, FL. Bergman, E. J. (2006, February). The journe y through grief following the loss of a spouse and the qualitative experiences of hospi ce caregivers. Paper presentation at the 32nd Annual Scientific Meeting of the A ssociation for Gerontology in Higher Education, Indianapolis, IN. Bergman, E. J., & Mills, W. L. (2006, Februa ry). Student perspec tives of teaching and professional development best practices wi thin an interdisciplinary aging studies program. Resource exchange at the 32nd Annual Scientific Meeting of the Association for Gerontology in Highe r Education, Indianapolis, IN. Bergman, E. J., Small, B. J., & Haley, W. E. (2005, November). Depression, social support, and utilization of formal bereav ement services. Poster presentation at the 58th Annual Scientific Meeting of the Ge rontological Society of America, Orlando, FL.
97 Appendix A (continued) Bergman, E. J., Jang, Y., Schonfeld, L., & Mo linari, V. (2005, May). The role of psychosocial resources in physical and me ntal health among older residents in assisted living facilities. Paper presentation at the 16th Annual Scientific Meeting of the Southeastern Regional Student Mentoring Convention in Gerontology and Geriatrics, Savannah, GA. Jang, Y., Bergman, E. J., Schonfeld, L., & Mo linari, V. (2005, March). Depressive symptoms among older residents in assisted living facilities. Paper presentation at the 26th Annual Scientific Meeting of th e Southern Gerontological Society, Orlando, FL. Bergman, E. J., Jang, Y., Schonfeld, L., & Molinari, V. (2005, March). Health perceptions and depressive symptoms among older residents in assisted living facilities. Poster presentation at the 26th Annual Scientific Meeting of the Southern Gerontologic al Society, Orlando, FL. Bergman, E. J., & Polubinski, J. (2005, March) Both sides of the coin: Factors in the decision to choose or not choose hospice care. Poster presenta tion at the Center for Hospice, Palliative Ca re, & End-of-Life Studies A nnual Scientif ic Meeting, Tampa, FL. Robb, C., Haley, W. E., Balducci, L., Exterm ann, M., Perkins, E. A., Small, B. J., Bergman, E. J., Jang, Y., & Mortimer, J. (2004, November). Examining quality of life in older breast cancer surviv ors. Paper presentation at the 57th Annual Scientific Meeting of the Gerontologi cal Society of America, Washington, DC. Haley, W. E., Robb, C., Balducci, L., Exterm ann, M., Perkins, E. A., Bergman, E. J., Small, B. J., Mortimer, J., & Jang, Y. (2004, July). Psychological, social, and health impact of breast cancer survivorsh ip in older women. Poster presentation at the National Cancer Institute and the American Cancer Society Annual Cosponsored Scientific Meeting, Washington, DC. Awards and Honors 2007 ProvostÂ’s Commendation for Outstanding Teaching by a Graduate Teaching Assistant University of South Florida 2006, 2005 Travel Grant Award Graduate and Professional Stude nt Council, University of South Florida 2006 Graduate Student Paper Award Association for Death Education & Counseling
98 Appendix A (continued) 1998 Dr. Tom Rich Scholarship University of South Florida and West Central Florida Area Agency on Aging 1997 Â– Present Sigma Phi Omega National Gerontology Honor and Professional Society Committees and Service Ad hoc abstract reviewer for the Thirty-f ourth Annual Meeting of the Association for Gerontology in Higher Education (2007) Research Committee Member, Pilot Research Gran t Program reviewer, Center for Hospice, Palliative Care, & End-of-Life Studies (2007) Conference Planning Committee Member, Program Committee Member, & Film Committee Member, Florida without Borders: Women at the Intersections of the Local and Global, Conference sponsored by the Florida Consortium for Women's Studies, February 2007. Ad hoc abstract reviewer for the Annual Meeting of the Geront ological Society of America Conference (2005-2008) Ad hoc abstract reviewer for the Gerontologi cal Health Section, American Public Health Association Annual Meeting (2005-2008) Student Association for Aging Studi es, University of South Florida President, 2004-2005 Member, 2003-present School of Aging Studies Ph.D. Governance Committee, University of South Florida Student Association for Aging Studies representative, 2004-2005 Professional Affiliations Gerontological Society of America, 1996-1998, 2003-present Center for Hospice, Palliative Care, a nd End-of-Life Studies, USF, 2003-present Southern Gerontological Society, 1998-present Florida Council on Aging, 1998-present Volunteer Experience Memory Walk Team Captain, The AlzheimerÂ’s Association Florida Gulf Coast Chapter, 2006, 2007
99 Appendix A (continued) Volunteer Driver, Meals on Wh eels of Tampa, 2006-present WomenÂ’s Organization for Medica l Emergency Needs, Inc. Board Chair, 2004-2007 Board Secretary, 2000-2004 Board Member, 1999-present Patient Care/Respite Volunteer, LifePath Hospice & Palliative Care, Inc. 1996-1999
About the Author Elizabeth J. Bergman received a Bachel orÂ’s Degree in Psychology from Auburn University in 1996 and a MasterÂ’s Degree in Gerontology from the University of South Florida in 1998. She worked as a gerontologi st for five years in service delivery and program administrative roles before entering the Ph.D. in Aging Studies program at the University of South Florida in 2003. While in the Ph.D. program at the Univer sity of South Florid a, Ms. Bergman was employed as a Graduate Teaching Associat e, teaching undergraduate Introduction to Gerontology, Sociocultural Aspects of Aging, and Death and Dying courses. She also served as Project Manager for two grant-funde d research projects, co-authored three peerreviewed journal articles and one book chap ter, and presented her work at several national and regional conferences including the Gerontological Society of America, the Southern Gerontological Society, the Associ ation for Gerontology in Higher Education, and the Association for Deat h Education and Counseling.