Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

484 Abstracts 24222-24227

12th World AIDS Conference

596*/24222 Serious mental illness among HIV-infected
medicaid patients in New Jersey
Stephen Crystal, James Thomason Walkup, Usha Sambamoorthi. Rutgers Univ,
Ints for Health TH, 30 College New Brunswick, New Jersey 08901-1293, USA
Objective: Several studies have investigated HIV prevalence rates among seriously mentally ill patients in urban psychiatric treatment settings, but no study has
investigated serious mental illness in a statewide population of the HIV-infected.
This study aimed to identify seriously mentally ill patients among HIV-infected
Medicaid patients and to describe their characteristics.
Methods: We used claims histories from file matching between the New Jersey
Medicaid eligibility file and the state's HIV/AIDS Registry. The study population included all Registry-identified HIV-infected patients in New Jersey who participated
in Medicaid for at least 180 days. Based on ICD-9-CM psychiatric diagnoses,
we defined three groups: (a) those with serious mental illness (schizophrenia, affective psychoses), (b) other psychiatric diagnoses (excluding substance
abuse), (c) no psychiatric diagnoses. Demographic and clinical characteristics
were compared.
Results: Of 8,294 individuals in the study population, 1,040 or 12.5% were
classified as having serious mental illness. The seriously mentally ill were more
likely to be injection drug users, to have HIV rather than AIDS, and to have health
care encounters indicative of active substance abuse. Among the seriously mentally ill, those in the schizophrenia group were particularly likely to be black, to
have an injection drug use history, to have HIV rather than AIDS, and to have
substance abuse diagnoses. Fewer than one-third of schizophrenics with HIV
and injection drug use histories had health care claims for methadone treatment.
Conclusions: Those with serious mental illness are a significant and distinctive, but little recognized, subgroup of the HIV-infected. In HIV policy and
program planning, it is important to consider how standard practices may require
modification to meet the needs of this subgroup.
24223 | Depression and CD4 cell count as predictors of quality of
life in HIV-positive persons
Jose Maldonado1, D.M. Israelski3, S. Diamond2, A. Chapman2, C. Koopman2,
M. Gill2, D. Spiegel2. 1Dept. of Psychiatry, 401 Quarry Road, Stanford, CA;
2Stanford University Standford, CA; 3San Mateo CTY AIDS Program/Stanford,
San Mateo, CA, USA
Objective: This study examined the relationships between depression and CD4
cell count with quality of life (QOL) among HIV-positive patients.
Design: A cross-sectional research design was used to examine relationships
among two independent variables, depression and CD4 cell count, with the
dependent variable of QOL.
Method: 103 HIV-positive subjects, 73 males and 30 females, were recruited
into a randomized clinical trial on the effects of groups psychotherapy on health
risk behavior and QOL. The subjects' mean age was 40 years, 47% were low
income, 63% white, 25% African-American, and 12% other ethnicity. Half of the
subjects had been diagnosed with AIDS. Measures included the Center for Epidemiological Studies-Depression Scale (CES-D), current CD4 cell counts, and
the Medical Outcomes Study-HIV (MOS-HIV).
Results: In the multiple regression analysis, depression was found to be significantly related to perceived poor health (Beta = 0.44, adjusted R2 = 0.180,
p < 0.0001), pain experienced (Beta = 0.35, adjusted R2 = 0.106, p < 0.001),
pain interference with duties (Beta = 0.39, adjusted R2 = 0.133, p < 0.0001),
difficulty in reasoning and problem solving (Beta = 0.49, adjusted R2 = 0.227, p <
0.0001), forgetfulness (Beta = 0.40, adjusted R2 = 0.144, p < 0.0001), difficulty
in sustaining attention (Beta = 0.52, adjusted R2 = 0.272, p < 0.0001) difficulty in
concentrating and thinking (Beta = 0.48, adjusted R2 = 0.274, p < 0.0001), poor
health (Beta = 0.50, adjusted R2 = 0.238, p < 0.0001), and health limiting social
activities (Beta = 0.45, adjusted R2 = 0.216, p < 0.0001). In contrast, CD4 cell
count was significantly related only to health limiting social activities (Beta = 0.23,
adjusted R2 = 0.240, p < 0.014).
Conclusions: Depression more than CD4 cell count is related to most indices
of quality of life among HIV-positive adults. Thus, in HIV-infected persons, a
subject's experience of depressive symptoms is a better predictor of poor quality
of life than a low CD4 cell count. These results suggest that depression adversely
affects QOL, and that effective treatment of depression may dramatically improve
the QOL of HIV-infected persons.
24224 Life stress, anxiety, depression, and coping among HIV+
adolescents
Debra A. Murphy', L. Muenz2, A.B. Moscicki3, S. Vermund4. 1Health Risk
Reduction Projects, Suite 350, 10920 Wilshire Boulevard, Los Angeles, CA;
2WESTAT, Rockville, MD; 3University of California, San Francisco, CA;
4 University of Alabama, Birmingham, AL, USA
Objective: HIV-infected adolescents age 13-19 participating in a national study
of HIV infection (REACH) received an audio-computer assisted self interview at

baseline. Subjects reported: number of life events occurring in the past 3 months,
distress level in response to life events, anxiety and depression levels, satisfaction
with social support network, and coping strategies.
Results: Analyses were conducted on 135 HIV+ adolescents (104 females, 31
males). Higher distress response to life events is associated with higher levels
of depression (t [122] = 2.65, p =.009) and anxiety (t[122] = 3.76, p =.000).
Stressful life events reported within the previous three months included: family

financial problems (63%); parental arguing and fighting (23%); having medication
prescribed (75%); parental alcohol abuse (23%); and witnessing physical violence
(37%). Despite these stressors, most adolescents reported having a good support
network; only 16% were unsatisfied with their social support. The association of
anxiety and distress response to life events is stronger (OR = 5.0) when satisfaction with social support is lower than when social support satisfaction is higher
(OR = 2.5). When satisfaction with social support is low, the odds ratio is 3.60
for the association of depression and distress response to life events; when it is
high the odds ratio drops to 1.31. Despite these large drops in odds ratio, which
indicate that social support moderates the effect of life events on psychological
distress, these were not significant findings for anxiety or depression. However,
adaptive coping was significantly associated with lower levels of depression (p
=.003). Adjusted for other variables (age, gender, HIV status, and life distress
response), a 1-point increase in the coping score yielded a 2.67 drop in the
depression score (.23 standard deviations).
Conclusions: Stress is associated with high levels of depression and anxiety among HIV-infected adolescents. Primarily adolescent females comprise the
sample (77%); most live at home (70.6%). While this may provide many benefits,
the adolescents experience many family and parental-related stressors. Adaptive
coping strategies are associated with less depression, which has implications for
intervention. However, longitudinal studies of infected adolescents are needed.
24226 Utilization of mental health services by women living with
or at risk for HIV
Suzanne Ohmit', Paula Schuman2, J. Moore3, E. Schoenbaum4, R. Boland5,
A. Rompalo6, L. Solomon7. 14160 John R Suite, Detriot, Ml, Wayne State
University, Ml; 2Centers for Disease Control, Atlanta, GA; 3Montefiore Medical
Center, Bronx, NY; 4 The Miriam Hospital, Providence, RI; 6 Johns Hopkins
University, MD; 7Maryland dept. of Health, Baltimore, MD, USA
Objective: To identify factors associated with perceived need for and utilization
of mental health services (MHS) and use of psychotropic medications among
women participating in the HIV Epidemiology Research Study (HERS).
Methods: 871 HIV seropositive and 439 seronegative women were interviewed at baseline regarding demographics, psychosocial measures, health care
utilization, substance abuse, medical history and use of medications.
Results: 38% of HIV+ and 35% of HIV- women (p =.20) reported needing
MHS during the past 6 months. Those who reported needing MHS were more
likely to be older, White or Hispanic (compared to African American), high school
graduates, have less social support, report a stressful life event in the past year
and have CESD scores >15 (suggesting depression). Among women who reported needing MHS, 67% of HIV+ and 65% of HIV- women (p =.40) reported
receiving services. Women who were White or Hispanic, with social support and
a regular health care provider were more likely to receive the MHS they reported
needing. Recent injection drug use (IDU) was unrelated to either reported need
for or receiving needed MHS. Eighteen percent of HIV+ and 13% of HIV- women
(p =.07) reported current use of psychotropic medications, especially anti-depressants and anti-anxiety medications. Women who were older, White or Hispanic,
with higher CESD scores, some college education and health insurance were
more likely to report use of psychotropic medications.
Conclusion: Perceived need for and utilization of MHS including psychotropic
drug use, were unrelated to HIV serostatus or recent IDU, but were associated
with race, social support and higher CESD scores. The proportion of women perceiving a need for mental health services and the relatively low proportion actually
receiving services, suggest that additional steps should be taken to identify and
link women to the services they need.
124227 Predictors of depression in lower class injection drug
users: The SAIL study
Amy Knowlton, C.A. Latkin, D.D. Celentano, D.R. Hoover. 624 N. Broadway,
John Hopkins SHPH, HPM/SBS, Baltimore MD 21205, USA
Objectives: The SAIL Study (1994-present) examined individual and personal
network attributes as predictors of depression in injection drug users (IDUs).
Design: Prospective study.
Methods: Participants were 503 former and current IDUs recruited for an HIV
natural history study by street outreach. 306 participants remained in the study at
the third semi-annual followup (1996). Multiple logistic regression was used to assess the independent influence on depression (CES-D, cut-off 16), controlling for
depression at baseline, of the individual attributes of HIV status, current drug use,
gender, age, education level, physical functioning (Rand MOS), satisfaction with
support from family, and of the network attribute of proportion of current drug users
in support networks at baseline. Statistical significance was defined as p <.05.
Results: Controlling for all other variables in the model, individual (OR = 2.02)
and network current drug use (OR =.55) were independently positively associated
with depression, and perceived social support (OR =.52) was negatively associated with depression. Women (OR = 2.35) compared to men, and those with
poorer physical health were more likely to be depressed (OR =.65), controlling

for other variables in the model. Female gender was significantly associated with
depression in the multivariate analysis but not in the bivariate analysis. HIV status,
education level, age, and baseline depressive symptoms were not significant in
the model.
Conclusion: Results of the SAIL study indicate that current injection drug use
was a predictor of depression, as was being female, social support and physical
health status. This study highlights the independent roles of social affiliations