Integrating theories and methods in assessing longitudinal change in heterosexual HIV risk among women
Abstract
This study attempted to gain additional clarity into the relationships among five constructs central to the decision-making process in HIV/AIDS risk behavior assessment. Comparison of four theoretical approaches to HIV/AIDS risk assessment and prediction (Health Belief Model, Transtheoretical Model of Behavior Change, AIDS Risk Reduction Model, and Mulitifaceted Model of HIV Risk) indicated similar constructs operating in the decision-making processes. These constructs were: Perceived Risk for AIDS, Decisional Balance, Self-Efficacy for Condom Use, Lifetime Number of Sex Partners, and Behavior Risk. Data were analyzed from 527 heterosexual women from a New England Community who completed lifestyle surveys at three time points, six months apart, as part of a study designed to assess predictors of risk for heterosexual transmission of HIV in women. Results from cross-sectional and longitudinal statistical modeling procedures using structural equation modeling were compared (including longitudinal cross-section, fully cross-lagged longitudinal, and latent growth curve structural analysis). Results indicated that the five named constructs were important in HIV risk prediction. Relationships among constructs were consistent across methods. Higher Behavior Risk was associated with higher Perceived Risk for AIDS, lower Self-Efficacy for Condom Use, and lower Decisional Balance (lower perceived benefits relative to perceived costs of condom use). Lifetime Number of Sex Partners was negatively associated with Perceived Risk for AIDS. Cross-methods comparison indicated that Lifetime Number of Sex Partners had the strongest and most consistent association with Perceived Risk for AIDS. Differences in Perceived Risk for AIDS across stages were not discernable. This indicates that Perceived Risk is an important predictor, but once risk assessment is high, it remains at a fixed level. Potential causal relationships were indicated between Decisional Balance and Self-Efficacy for Condom Use and between Self-Efficacy for Condom Use and Behavior Risk. Self-Efficacy for Condom Use was determined to be an important outcome measure in the strength of relationships with all constructs in the model. Causal determinations were limited by the nature of the survey design (e.g., only three time points were assessed, and the design was not an intervention); replication of the analysis procedure is suggested for studies designed to change behavior. ^
Subject Area
Women's Studies|Health Sciences, Public Health|Psychology, Experimental|Psychology, Psychometrics
Recommended Citation
Gary J Burkholder,
"Integrating theories and methods in assessing longitudinal change in heterosexual HIV risk among women"
(2000).
Dissertations and Master's Theses (Campus Access).
Paper AAI9988099.
http://digitalcommons.uri.edu/dissertations/AAI9988099