Predictors of vulnerability to AIDS risk behavior relapse.

Abstract

Sixty-eight gay men who earlier attended AIDS prevention sessions were longitudinally followed for 16 months and categorized as successful or unsuccessful in change maintenance. Psychological and behavioral data obtained prior to entry in the prevention program were used as variables to predict long-term change. Resumption of high-risk sexual practices was associated with younger age, earlier history of frequent unprotected receptive anal intercourse with multiple partners, greater number of past sex partners, reinforcement value levels of high-risk practices and condom use, intoxication preceding sex, lower scores on a depression measure, greater belief that HIV infection is largely determined by external factors such as chance or luck, and homosexuality "outness." Discriminant analysis revealed that 86% of Ss could be classified as relapsers or change-maintainers on the basis of these variables. Prevention implications are discussed.The longitudinal research study of 68 gay men from a Southern town in 1987 was able to successfully identify 86% of those who were likely to relapse in unprotected anal intercourse. The men were followed 16 months after receiving 12 AID's health risk reduction sessions. The mean age of the population was 32.7 years; 945 were white and 6% African-Americans or Hispanics. Multivariate discriminant analysis was used to identify the relative magnitude of each predictor variables contribution (minimum coefficient of + or - .30). Predictor variables included demographic data and sexual risk behaviors which the subject ranked by the degrees of pleasure on a 4 point scale. Also included were the role of intoxication in sexual behavior, the Beck Depression Inventory, the State Trait Anxiety Inventory, the Health Locus of Control Scale (HLOC), and the AID's Risk Behavior Knowledge Scale. Role plays involving each subject were rated by 2 judges on the overall skill in resisting coercive sexual pressures. After 16 months, 41 reported no occurrence of unprotected anal sex during the preceding 16 months and 27 reported some occurrence, of which the mean frequency was 5.9 occurrences in the preceding 4 months. The relapse population tended to be younger and more out about their homosexuality. The number of sexual partners, use of intoxicants with sex, and frequency of unprotected receptive anal intercourse were the strongest predictors. These predictors were the same for the relapsed men before the risk reduction sessions. These relapsed men also reported lower depression levels and higher scores on HLOC. Similar to cigarette smoking, resuming risk behavior was related to the strength, frequency, and reinforcement value level of past behavior. Outness may be related to frequency of sexual contact and HLOC to beliefs that luck, fate, or other determine risk. Serostatus data was not collected. The limitation of this study was that situational dimensions such as the environment, cognitive, affective, and interpersonal patterns were relevant but not included. Identification of such factors is a needed future research objective.