The human immunodeficiency virus (HIV) is spread largely through sexual contact with an infected individual. There is a clear need to prevent the spread of HIV through the increased use of prophylactics. There is evidence that condoms, particularly latex, prevent the transmission of HIV if they are used correctly and consistently. However, condoms are not used consistently by those engaging in behaviors which put them at risk for HIV infection. If we are to reduce the morbidity and mortality due to AIDS, it is necessary to design interventions to increase the use of condoms among those individuals. Such interventions should be based on a clear understanding of population specific factors associated with condom use. Several studies have shown that condom use among individuals who engage in risky behaviors is difficult to change or maintain. Promoting knowledge about AIDS does not appear sufficient to increase rates of condom use. A number of studies have identified factors associated with condom use, but none have systematically studied these factors using a theory of behavior and behavior change. No study has used one theoretical framework to compare these factors among different groups of individuals who are at risk for contracting and transmitting AIDS.
The specific aim of the proposed project is to apply attitude and behavior change theory to the identification and understanding of the factors affecting patterns of condom use in individuals who engage in risky behaviors. An expanded version of Fishbein's Theory of Reasoned Action will be used as the theoretical framework. Preliminary elicitation interviews will be conducted with four targeted groups of individuals: 1) homosexual or bisexual men, 2) IV drug users, 3) prostitutes and 4) multiple-partner heterosexuals, stratified by race/ethnicity status; White, Black, Latino. The most salient beliefs, social normative influences, experiences, and facilitating conditions associated with using or not using condoms will be identified. Based on this information, survey instruments will be designed to measure each of the model components as well as rate of condom use. The surveys will be pilot tested in small groups to assess reliability and validity. The surveys will then be administered to 900 individuals from the four target groups at two points in time, three months apart. Regression analyses will be conducted to identify the key beliefs, values, social normative influences, past experiences, and environmental conditions that best predict rates of condom use with each target group or racial/ethnic group. Further analyses will be conducted to compare those groups on the identified key variables and to assess changes in those variables over time.
