| This study investigated a proposed link between survivor guilt and HIV risk behavior in a sample of 321 men who have sex with men (MSM) who reported they were not HIV-positive; these men presented for psychological services at the University of California, San Francisco AIDS Health Project (AHP) between July 1, 1998 and June 30, 2002 and were divided into 4 year-cohorts (July 1, 1998--June 30, 1999; July 1, 1999--June 30, 2000; July 1, 2000--June 30, 2001; July 1, 2001--June 30, 2002). Participants were primarily European American, aged 21 to 62 years, with a mean monthly income of {dollar}2326. The majority had at least some college education.; Archival psychosocial data were utilized for this study. Participants had completed several measures created by AHP as well as the Profile of Mood States (POMS). New measures of survivor guilt and HIV risk behavior were developed and their psychometric properties were examined. The Survivor Guilt (SG) Scale consisted of 8 items that, on their face, represented elements of survivor guilt identified in the literature. Two scores were derived from this scale, SG Total Score and SG Number of Items Score. The former had good reliability (alpha = .70) and was used to test the relationship between survivor guilt and HIV risk behavior. The HIV Risk Behavior Score consisted of 6 items reflecting sexual risk behavior during the previous 3 months to 1 year. The reliability of this score was moderate (alpha = .66).; The hypothesis that the proportions of participants experiencing survivor guilt and the intensity of survivor guilt would remain stable across the four years of data was supported using between groups (4 year-cohorts) comparisons. Of the 37% of participants who experienced survivor guilt, the intensity of the experience was typically low, although for a small number it was higher. The hypothesis that survivor guilt contributes significantly to HIV risk behavior was supported (p < .01) for the total sample and for the first year-cohort after the effects of age, substance use, and depression and vigor as measured by T-scores on the POMS were statistically controlled during hierarchical multiple regression analyses. The relationships did not reach statistical significance for the last three year-cohorts. Fisher's z Tests demonstrated that the hypothesis that the relationship between survivor guilt and HIV risk behavior would remain stable across the four year-cohorts was supported.; The results of this study contribute to the limited empirical literature about the experience of MSM living in a community greatly affected by AIDS. Survivor guilt was shown to contribute to HIV risk behavior. Clinical implications are discussed. |
