This proposal is a competitive renewal of our NIMH-funded research on self-disclosure of diagnosis among individuals with HIV. To date, our investigation has generated important findings regarding the prevalence of unsafe sexual activity and nondisclosure to sexual partners among seropositive individuals, and psychosocial and medical correlates of disclosure to significant others (family members, friends, intimate lovers) and less significant others (employers, medical providers, landlords, religious leaders). Our work has resulted in three published articles. During the 1990-93 funding period, we accrued a large multi-ethnic, longitudinal data base of over 600 HIV seropositive men and women attending two outpatient clinics in Los Angeles to examine the representativeness of the findings from our first study (1987-90), which was conducted with a sample of Hispanic men. One purpose of the present proposal is to perform comprehensive statistical analyses of our new data and prepare additional manuscripts for publication addressing psychological, cultural, and public health issues. A second purpose is to propel our research to the next phase, in which we expand basic understanding of factors promoting high-risk sexual activities among those who are seropositive. This new direction will directly benefit counseling efforts and educational programs to control the spread of HIV, and advance the design of behavioral interventions. To this end, we propose to collect new questionnaire data from HIV- infected men to provide a pilot examination of the influence of psychological variables and substance use on engagement of high-risk sexual activities. The longitudinal Affective-Behavioral Model focuses on three psychological constructs measured at Time 1: 1) depressed affect/low self-acceptance, 2) anger, and 3) social isolation as direct and indirect predictors of nondisclosure to sexual partners and unsafe sexual practices measured three months later. We expect these psychological constructs to influence consumption of alcohol and other drugs, which will affect sexual practices, as suggested by our current data. We measure specific sexual encounters and the use of substances during the four hours preceding that encounter. The model is analyzed with multiple regression path analyses and direct tests of mediation. Medical information abstracted from subjects' medical charts are incorporated into the analysis.
Showing the most recent 10 out of 11 publications
