Symptom experiences and HIV medication adherence: Modeling cognitive and affective pathways

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)



First Committee Member

Frank J. Penedo - Committee Chair


Purpose. Recent research in the area of HIV treatment has emphasized the need for near-perfect levels of medication adherence in order for treatment to be successful. This study sought to identify the extent to which beliefs about medications in general and HIV treatment medications in particular were related to HIV medication adherence. The potential influences of symptoms of HIV and negative mood states on medication beliefs were also evaluated.Method. This study evaluated these relationships among 325 HIV+ men and women prescribed Highly Active Antiretroviral Therapy (HAART). The Beliefs about Medicines Questionnaire (BMQ; Horne, Weinman, & Hankins, 1999), which was substantially revised using confirmatory factor analysis for the current study, was used to assess medication beliefs. Latent variable modeling, using multiple methods of adherence measurement, was used to increase the reliability of the measurement of HAART adherence and to limit the impact of random and measurement biases. Analyses examined negative mood states and beliefs about HAART medication as mediators of the relationship between symptoms and adherence within a structural equation modeling framework using a longitudinal design.Findings. Results showed that number of symptoms was negatively associated with medication adherence and that this relationship was partially mediated by increases in concerns about HAART. Contrary to expectations, negative mood states were not directly related to medication adherence. Negative mood states did show a significant indirect relationship to medication adherence through increased concerns about HAART. In the final model, concerns about HAART and general distrust of medications were each negatively predictive of poorer HAART adherence. Necessity beliefs about HAART and level of educational attainment were each positively predictive of adherence. Both symptom totals and negative mood state had significant negative indirect relationships to poorer HAART adherence in the final model.Implications. The results of this study suggest the framework for how patients understand their symptoms, their distrust about medications in general, and their specific concerns and necessity beliefs about HAART are important in influencing adherence behavior over time. These factors could be rewarding targets for future intervention studies.


Health Sciences, Medicine and Surgery; Psychology, Clinical

Link to Full Text


Link to Full Text