Development and Pyschometric Validation of Measures for Vaginal Ring Adherence Survey

With funding from USAID, FHI 360 led an effort to create a coherent, validated, and feasible adherence measurement and support package for use in ARV-based vaginal ring trials. The package integrates evidence and best practices related to adherence measurement and support together with promising biomedical innovations. The first priority was to better understand the motivations for and against product adherence in the clinical trial context and to develop, test, and refine adherence support messages and other communications for use in future clinical trials to maximize ring adherence. Second, a vaginal ring psychometric adherence scale was developed and piloted with the goal of creating a tool that is inexpensive and easy to administer, with better predictive ability than current self-report adherence measures. Third, the project supported development of designs, models, and/or prototypes for innovative tools to measure ring adherence, using “smart” and biometric technologies to detect ring insertion and/or removal. This dataset is a product of the second priority, which developed a vaginal ring psychometric adherence scale. We used a tablet-based survey administered by same-sex interviewers with a total of 709 women who were former participants of vaginal ring or other HIV prevention clinical trials and similar trial-naïve women. This was a cross-sectional survey that included both closed-ended and possible brief open-ended responses. The primary objectives were to: 1. Identify a reduced set of items that measure: a. Ongoing barriers and facilitators to “enacted” adherence to a vaginal ring, and b. Propensity to adhere to vaginal ring use in a future trial; and 2. Assess content and construct validity of draft measures. The survey was divided into three sections. All women regardless of trial experience responded to 76 “screener” items measuring potential predictors of adherence; 26 items from the Crowne-Marlowe social desirability scale were also distributed in groups within this section. Only former trial participants responded to an additional set of 48 monitoring items assessing product-specific adherence attitudes and behaviors. The last section, administered to all participants, included additional demographic and psychosocial variables.