Psychosomatics 41:262-268, June 2000
© 2000 The Academy of Psychosomatic Medicine
The Influence of HIV-Related Support Groups on Survival in Women Who Lived With HIV
A Pilot Study
Jacquelyn Summers, Ph.D., M.S.W.,
Renee Robinson, M.A.,
Lisa Capps, Ph.D.,
Sidney Zisook, M.D.,
J. Hampton Atkinson, M.D.,
Emily McCutchan, M.A.,
J. Allen McCutchan, M.D.,
Reena Deutsch, Ph.D.,
Thomas Patterson, Ph.D., and
Igor Grant, M.D.
Received January 28, 1999; revised May 20, 1999; accepted July 29, 1999. From the University of California, Berkeley (UCB) School of Social Welfare, the UCB School of Public Health, and the UCB Department of Psychology; the University of California, San Diego (UCSD) HIV Neurobehavioral Research Center, the UCSD Department of Psychiatry, and the Division of Infectious Diseases, UCSD School of Medicine, and the Veteran Affairs Medical Center, San Diego, California. Address reprint requests to Dr. Summers, Department of Psychology, University of Auckland, Private Bag #92019, Auckland, New Zealand; e-mail: j.summers{at}auckland.ac.nz
|
ABSTRACT
|
To determine the effect of support groups on survival, the authors retrospectively studied 21 HIV-seropositive women who died during the course of participation in a natural history study of HIV. Groups were composed of women who self-selected HIV support groups before death (n=11) and a comparison group (n=10). Survival analysis found group participation to be associated with increased longevity (73 months vs. 45 months; P=0.011). Proportional-hazards regression demonstrated that HIV-related support groups and smaller family size significantly influenced survival (P=0.0002). Factors related to group participation and ways in which support groups might promote longevity are discussed.
Key Words: AIDS/HIV • Support Groups
|
INTRODUCTION
|
Women are among the fastest-growing group at risk for HIV infection.1–3 Importantly, women may have shorter survival times than men. Abbreviated survival in women living with HIV has been attributed to more advanced disease staging at the time of HIV diagnosis and medical misdiagnosis,4 lower socioeconomic status,5 unequal access to HIV treatment,6 and biological sex differences.7 Little attention has been given, however, to the possible influence of social support on survival time among individuals living with HIV. This is surprising in light of research suggesting that social networks prolong survival among individuals with cancer,8–10 cardiovascular disease,11–13 and hemodialysis.14 Whether social ties affect survival outcomes in HIV is an especially relevant question because women have been described as being more isolated from community support than gay men affected by the AIDS epidemic.15,16
To examine this issue, we compared the survival experience in seropositive women who participated in HIV-related support groups before death to seropositive women who did not take part in support groups. We hypothesized that participation in groups would be associated with longer survival. This investigation had the advantage of having access to a well-characterized study population of women who lived with HIV. Although participation in support groups was not randomly assigned in this study, the potential effects of selection bias were likely attenuated by the fact that the comparison group closely matched those in support groups with respect to key demographic and medical characteristics.
|
METHODS
|
This study was part of a larger longitudinal cohort study of the neurobehavioral aspects of HIV at the University of California's HIV Neurobehavioral Research Center (HNRC) in San Diego, California. Of 234 study inquiries, 154 participants were screened for the study. Of these, 129 were enrolled and provided written, informed consent. Between August 1991 and December 1996, 21 of the 129 women participating in the HNRC died (16.3%). This retrospective study consisted of these 21 deceased women. Most deaths were reported by next of kin, friends, postal authorities, or in the cases of women consenting to autopsy, health care providers.
Procedure
Two groups were compared in the current investigation: a group of deceased women who were identified as participating in an HIV-related support group after the time of their HIV diagnosis (n=10) and a comparison group of seropositive women (n=11) who did not participate in support groups between the time of their HIV diagnosis and date of death. In the support group sample, 9 women participated in a 12-week session of HNRC-based supportive group therapy offered as a community service. These groups were cofacilitated by two of three licensed therapists (E.M., R.R., or J.S.). Discussions centered on self-image, family/parenting issues, intimacy, collaborating with healers, disclosure and privacy, death and dying, spirituality, humor, and relaxation. The tenth member of the support group sample was active in an ongoing peer support group offered through a local HIV resource and advocacy center. The HNRC-sponsored supportive group therapy met once weekly for 90 minutes for a minimum of 12 sessions. The community agency support group met weekly for peer-facilitated support in a group mixed with men and women living with HIV. Both groups were offered at no charge. Both groups were open to all seropositive women in the San Diego area. Only women who participated in four or more group meetings within a 12-week period were included in the support group sample. Comparison group members participated in the neurobehavioral study assessments but reported no HIV-related group support between time of HIV diagnosis and death.
Measures
Stage of HIV illness was classified using Center for Disease Control 1993 criteria. HIV-related diagnoses were based on medical history, physical examination, and laboratory studies. HIV serologic status was determined by a positive enzyme-linked immunoassay followed by a confirmatory immunofluorescent assay or Western blot. Lymphocyte subsets were obtained by flow cytometry, using immunofluorescent monoclonal antibody staining techniques for CD4+ markers (Ortho Diagnostic Systems; Raritan, NJ). Antiretroviral treatment (e.g., zidovudine, DDI) was dichotomously rated by assigning "1" to individuals reporting any retroviral use for greater than 3 months since the time of diagnosis and "0" for exposure lasting less than 3 months.
Participants were examined for lifetime and current (within 1 month of the final assessment before death) prevalence of psychiatric disorders using the Structured Clinical Interview for DSM-III-R (SCID17). The SCID is a clinician-administered, semistructured assessment that bases the diagnosis of Axis I psychopathology on DSM-III-R criteria. The SCID was used to evaluate for major depression, generalized anxiety disorder, alcohol use disorder, other substance use disorder, panic disorder, and social phobia. Interrater reliability of 0.8 to 0.9 was established for major diagnoses of interest among research clinicians.
Statistical Analysis
Survival was measured from the date of the initial HIV diagnosis to the date of death from any cause. Survival after the diagnosis of HIV infection was calculated with the product-limit survival estimate using months of survival as the time estimate, support group membership as the grouping variable, and ethnicity (white/minority) as the stratum. Survival curves were compared by calculation of the stratified log-rank statistic. Proportional-hazards regression was used to examine the impact of multiple covariates on survival time.
|
Results
|
Demographics
Table 1 details the demographic characteristics of the two groups. Participants were on average high-school educated and in their late 30s. Group-comparison age distributions were similar (27–47 years vs. 23–41 years). Heterosexual women comprised 100% of the support group sample and 82% of the comparison sample. Sixty percent of the support group sample and a comparable 73% of the comparison sample were single. Ninety percent of the support group sample compared to 73% of the comparison group were mothers. At baseline, group-comparison CD4+ counts were comparable (198.9 vs. 121.1). The average duration of months since an HIV diagnosis was 41 months for women involved in support groups and 27 months for the comparison group, a difference that was not statistically significant. Approximately 40% of the women from both groups participated in individual psychotherapy between the time of HIV diagnosis and death, whereas one-third received pharmacotherapy. No statistical difference existed between the groups on either of these indices. Of the women in the support group sample, two-thirds elected to repeat groups more than once before death.
View this table:
[in this window]
[in a new window]
|
TABLE 1. Baseline demographic characteristics (mean±SD) of HIV-seropositive women with and without supportive group therapy before death (N=21)
|
Ethnicity was the sole group-comparison difference to emerge. Ten percent of the support group sample self-identified as an ethnic minority compared to 64% percent of the comparison sample (Fisher exact, P<0.03). This ethnic difference enters a potential bias into subsequent analysis. For this reason, we examined the survival experience of white and minority members in the comparison group. Because the median survival of minority comparison subjects was actually longer than that of whites (51 months vs. 36 months), we believe that this mismatch on ethnicity did not bias the results; if anything, the effect was to diminish the likelihood of finding differences in survival between the supportive therapy and comparison groups.
Medical Prognostic Indicators
Medically, the two groups were similar. Both groups were diagnosed with HIV in their 30s. Upon entry, over 70% of the women in both groups met CDC criteria for an AIDS diagnosis; at the final assessment, all women except one member of the comparison sample had an AIDS diagnosis. All the women except one member of the comparison group participated in a retroviral treatment lasting longer than 3 months. For both groups, final research assessments preceded death by an average of 12 months.
Mode of HIV exposure was similar for the two groups, being heterosexual intercourse for a majority of women from both groups. Almost 20% of the women from both groups attributed intravenous drug use to their risk of exposure, although both intravenous drug users from the support group sample described a concurrent risk of heterosexual sex. One woman was exposed to HIV through a blood transfusion. Of the remaining two women, one attributed exposure to a blood spill and the other was uncertain about her exposure.
No relationship existed between group membership and cause of death. A known cause of death was available for 16 of the women and all of these were attributed to HIV-related complications. Pneumocystis carinii pneumonia, the most common disease attributed to death, occurred in 44% (7/16) of the women with known deaths, followed by HIV encephalopathy in 25% (4/16) and cytomegalovirus retinitis in 19% (3/16). Other diagnoses were singular cases of non-Hodgkins lymphoma, pulmonary complications, and seizures. More than one cause of death was possible for each woman. Cause of death was unknown for 5 women, including 1 participant who was incarcerated at the time of death.
Psychiatric Diagnoses
Psychodiagnostic information was available for 71% (15/21) of the sample. Group-comparison lifetime and current (1 month) psychiatric morbidity at the time of the final longitudinal assessment were strikingly similar. A lifetime history of at least one major Axis I psychiatric diagnosis was evident among all of the women in both groups (15/15). No differences in the proportions of major lifetime diagnoses were evident between support group participants and the comparison group. The combined lifetime prevalence of the psychiatric diagnoses were major depression (93%, 14/15), alcohol use disorder (27%, 4/15), other substance use disorder (33%, 5/15), generalized anxiety disorder (27%, 4/15), panic disorder (27%, 4/15), and social phobia (7%, 1/15).
A current Axis I psychiatric disorder at the final assessment before death was no more likely in women who participated in support groups (33%, 3/9) than in women in the comparison group (50%, 3/6). The most common psychiatric diagnosis for both groups was major depression. The joint prevalence of current psychiatric diagnoses for the two groups were major depression (33%, 5/15), generalized anxiety disorder (13%, 2/15), panic disorder (7%, 1/15), and social phobia (7%, 1/15). None of the women in either group met the diagnostic criteria for current alcohol use disorder or other substance use disorder at the time of the final assessment.
Survival
The Kaplan-Meier plot in Figure 1 shows that women from support groups had a longer survival time compared to women without group support (P=0.011). The median survival time from the time since the diagnosis of HIV to date of death for women in the support group sample was 120 months for minorities and 73 months for whites. Women not participating in support groups had a median survival time of 51 months for minorities and 36 months for whites. The overall median survival for support group participants was 73 months compared to 45 months in the comparison group. Survival distributions between the support group and comparison sample adjusted for ethnicity were significantly different (Log Rank Test; P=0.0107).
View larger version (22K):
[in this window]
[in a new window]
|
FIGURE 1. Duration of survival as measured by number of months after diagnosis of HIV to date of death in women with (n=10) and without (n=11) support group therapy
|
Cox regression analysis was performed to determine which, if any, of the following factors accounted for the women's survival experiences: ethnicity, age, education, marital status, CDC staging at baseline, use of retrovirals, family size, individual psychotherapy, and HIV-related support group participation. Demographic indicators were considered in the first set of analyses. Race/ethnicity was not related to survival in Cox regression analysis. CDC staging, age, education, use of retrovirals, and marital status also failed to emerge as explicative covariates in Cox proportional-hazards regression. Second, the social support variables of support group status, family size as measured by number of children, and individual therapy were analyzed to evaluate their association with survival time. Participation in support groups (P=0.0039) and smaller family size (P=0.0134) both significantly increased the odds of longer survival. Individual therapy did not significantly account for survival.
Because the groups were relatively homogeneous in CDC staging at baseline and at their final assessment, months from the time of HIV diagnosis to baseline was added as a covariate in the Cox regression. Time since HIV diagnosis was our closest substitute marker for months of known infection, given that the time of incubation and actual date of seropositivity was unknown for most participants (except in cases of blood transfusions and accidental needle sticks). Ethnicity was also retained in the final analysis as a covariate due to the group-comparison difference that was detected in group membership distributions. For this final model, ethnicity continued to materialize as a nonsignificant factor (P=0.99). Despite the significant influence of months HIV+ prior to baseline on survival (P=0.0013), support group participation (P<0.02) and smaller family size (P<0.03) continued to emerge as explicative covariates of survival (Table 2).
View this table:
[in this window]
[in a new window]
|
TABLE 2. Predicting duration of survival by support group participation, family size, and ethnicity in HIV-seropositive women with and without supportive group therapy before death (n=21)
|
|
CONCLUSION
|
This observational pilot study examined the influence of HIV-related support groups on survival time in women who lived with HIV. These preliminary findings indicate that participation in HIV-related support groups for women living with HIV may be associated with longer survival.
Because this was a pilot study, a number of significant limitations may have influenced our results. First, because group membership was based on self-selection, caution must be exercised in interpreting the results. As noted in other studies of HIV longevity,2,18,19 tertiary factors other than the status of support group participation may have driven the differential survival patterns between groups. For example, factors not assessed in this study, such as a "fighting spirit" or other social support resources (e.g., churches, pets, extended kinship networks), may have influenced longer survival rather than support group participation. Additionally, group participants may have enjoyed better diets or more stable sleep patterns than comparison group members, which may have bolstered survival differentially. Lacking random assignment, the study outcomes must be interpreted conservatively.
Second, data collection was incomplete and the sample size was extremely small. A larger sample size would have provided greater statistical power to help determine group-comparison differences. The relatively small number of participants in this study and missing data may have led to Type II errors in some comparisons (e.g., weak or absent relationships between psychiatric sequelae and support group status). Assessing a larger, more diverse sample of women would provide a more useful and generalizable characterization of women living with HIV.
Two findings in particular warrant further comment. The first involves the benefit of HIV-related supportive group therapy among women living with HIV. The second concerns the identification of social and demographic factors that may account for differential survival patterns. First, with respect to survival, women in HIV-related support groups exhibited a significant survival advantage over the comparison group. These findings are consistent with studies of support groups for individuals living with cancer, in which support group members survive almost twice as long as their counterparts who are not in support groups.10,18,20 Supportive group therapy may promote health and provide a sense of belonging for women who may otherwise be socially and geographically isolated during the HIV disease process. A necessary next step in establishing the benefits of HIV-related social support will be randomized, controlled trials. Nevertheless, since the observational design of this study focuses on community support available to women living with HIV and the "real world" use of such resources, it provides a realistic assessment of the true impact of HIV-related support groups on the natural history of HIV in women.
The role of family size also bears mention. In multivariate modeling, the fewer children in a family, the longer a woman tended to survive. Although we had not predicted that family size would be a significant factor in survival, these findings may be interpreted in light of the literature on social support. Studies of social support in seropositive gay men have suggested that social contact may serve as a buffer against stress.21,22 However, for women, other investigators purport that "social ties may also bring social strains."23 Social strain refers to the additional burdens that can be laden upon seropositive women who frequently serve multiple social roles as mothers, spouses, employers, and careproviders. For the women in this study, the demands of these roles, particularly motherhood, was evident. Despite the fact that many women verbalized a maternal drive to survive for the sake of their children (e.g., "I've got to be here to help Reggie start walking"; "I will survive to see a cure for my baby Anne"; "I plan to live for Maria's graduation from college"), an increased number of children in the family did not favorably influence life expectancy. Future research investigating the quality of family relationships may be an important next step in understanding this outcome.
This sample represents some of the first women in the United States to participate in a neurobehavioral study of HIV. Other HIV-seropositive women not involved in the rigors of a longitudinal cohort study may be experiencing less support and more social isolation than the women described here. Replicating this study with more stringent experimental methodology on a larger and more diverse sample of women is a necessary next step.
In summary, this observational pilot study of HIV-seropositive women revealed an extended duration of survival after the diagnosis of HIV infection for self-selected participants of HIV support groups. The evidence from this study suggests that a seropositive woman's connectedness to her peers and a smaller family size may have salutary effects on survival. For women living with HIV, ties to other individuals living with HIV may be among the factors that extend life.
|
ACKNOWLEDGMENTS
|
The authors gratefully acknowledge the biostatistical contributions of Drs. Mark Van der Laan and Geoffrey Keppel of UC Berkeley. The authors also extend gratitude to Ms. Nancy Newhouse of UCSD for her clinical supervision. And most importantly, the authors acknowledge the generosity of these 21 female pioneers with HIV whose memories are cherished and whose presence is missed.
This project was supported by NIMH Grant 5-P50-MH45294.
The San Diego HIV Neurobehavioral Research Center (HNRC) Group is affiliated with the University of California, San Diego, the Naval Hospital, San Diego, and the San Diego Veterans Affairs Healthcare System, and includes: Igor Grant, M.D., Director; J. Hampton Atkinson, M.D., Co-Director; Thomas Marcotte, Ph.D., Center Manager; Mark R. Wallace, M.D. and James L. Chandler, M.D., Co-Investigators Naval Medical Center; J. Allen McCutchan, M.D., P.I. Neuromedical Core; Robert K. Heaton, Ph.D., P.I. Neurobehavioral Core; John Hesselink, M.D. and Terry Jernigan, Ph.D., Co-P.I.s Imaging Core; J. Allen McCutchan, M.D., J. Hampton Atkinson, M.D., and Ronald J. Ellis, M.D., Ph.D., Clinical Trials Core; Daniel R. Masys, M.D., P.I. Data Management Unit; Ian Abramson, Ph.D., P.I. Biostatistics Core; Michele Frybarger, B.A., Data Manager. The views expressed in this article are those of the authors and do not reflect the official policy or position of the Department of the Navy, Department of Defense, nor the United States Government.
|
REFERENCES
|
-
Centers for Disease Control and Prevention: Update: AIDS among women—United States, 1994. MMWR 1995; 44:5
-
Lemp GF, Hirozawa AM, Cohen JB, et al: Survival for women and men with AIDS. J Infect Dis 1992; 166:74–79[Medline]
-
Smith DK, Moore JS: Epidemiology, manifestations, and treatment of HIV infection in women, in Women and AIDS, Coping and Care. Edited by O'Leary A, Jemmott LS. New York, Penguin, 1996, pp 1–24
-
Chu SY, Buehler JW, Berkelman RL: Impact of human immunodeficiency virus epidemic on mortality in women of reproductive age in the United States. JAMA 1990; 264:225–229[Abstract]
-
Melnick SL, Sherer R, Louis TA, et al: Survival and disease progression according to gender of patients with HIV infection: The Terry Beirn Community Programs for Clinical Research on AIDS. JAMA 1994; 272:1915–1921
-
Ickovics JR, Rodin J: Women and AIDS in the United States: epidemiology, natural history, and mediating mechanisms. Health Psychol 1992; 11:1–16
-
Richardson JL, Shelton DR, Krailo M, et al: The effect of compliance with treatment on survival among patients with hematologic malignancies. J Clin Oncol 1991; 8:356–364[Abstract]
-
Farzadegan H, Hoover DR, Astemborski J, et al: Sex differences in HIV-1 viral load and progression to AIDS. Lancet 1998; 352:1510–1514
-
Reynolds P, Boyd PT, Blacklow RS, et al: The relationship between social ties and survival among black and white breast cancer patients: NCI Black/White Survival Study. Cancer Epidemiol Biomarkers of Prevention 1994; 3:253–259
-
Spiegel D: Can psychotherapy prolong cancer survival? Psychosomatics 1990; 31:361–366
-
Anderson D, Deshaies G, Jobin J: Social support, social networks, and coronary artery disease rehabilitation: a review. Can J Cardiol 1996; 12:739–744[Medline]
-
Kawachi I, Colditz GA, Ascherio A, et al: A prospective study of social networks in relation to total mortality and cardiovascular disease in men in the USA. J Epidemiol Comm Health 1996; 50:245–251[Abstract]
-
Thomas SA, Friedmann E, Wimbush F, et al: Psychological factors and survival in the cardiac arrythmia suppression trial: a reexamination. Am J Crit Care 1997; 6:116–126
-
Christensen AJ, Wiebe JS, Smith TW, et al: Predictors of survival among hemodialysis patients: effect of perceived family support. Health Psychol 1994; 13:521–525[CrossRef][Medline]
-
Semple SJ, Patterson TL, Temoshok LR, et al: Identification of psychobiological stressors among HIV-positive women. Women and Health 1993; 20:15–34[CrossRef][Medline]
-
Russell JM, Smith K: HIV infected women and women's services. Health Care for Women International 1998; 19:131–139[CrossRef][Medline]
-
Spitzer RL, Williams JB, Gibbon M, et al: Structured Clinical Interview for DSM-III-R. Washington, DC, American Psychiatric Association, 1990
-
Spiegel D, Bloom J, Kraemer H, et al: Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet 1989; 2:888–891[CrossRef][Medline]
-
Berkman LF, Syme SL: Social networks, host resistance and mortality: a nine-year follow-up study of Alameda County residents. Am J Epidemiol 1982; 109:186–204[Abstract/Free Full Text]
-
Fawzy FI, Cousins N, Fawzy NW, et al: A structured psychiatric intervention for cancer patients, II: changes over time in immunologic measures. Arch Gen Psychiatry 1990; 47:729–735[Abstract]
-
Ostrow D, Monjan A, Joseph J, et al: HIV-related symptoms and psychological functioning in a cohort of homosexual men. Am J Psychiatry 1989; 146:737–742[Abstract/Free Full Text]
-
Zich J, Temoshok L: Perceptions of social support in men with AIDS and ARC: relationships with distress and hardiness. Journal Applied Social Psychology 1987; 17:193–215
-
Jenkins S, Coons H: Psychosocial stress and adaptation processes for women coping with HIV/AID, in Women and AIDS, Coping and Care. Edited by O'Leary A, Jemmott LS. New York, Penguin, 1996, pp 33–71
This article has been cited by other articles:
|
|
|
|
 
D. E. Brashers, S. M. Haas, J. L. Neidig, and L. S. Rintamaki
Social Activism, Self-Advocacy, and Coping with Hiv Illness
Journal of Social and Personal Relationships,
February 1, 2002;
19(1):
113 - 133.
[Abstract]
[PDF]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 2000
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
