Interactive Voice Response System: Data Considerations and Lessons Learned During a Rectal Microbicide Placebo Adherence Trial for Young Men Who Have Sex With Men

Background: Rectal microbicides, if proven effective, may aid in reducing human immunodeficiency virus (HIV) incidence; however, demonstration of efficacy and effectiveness is contingent on accurate measurement of product adherence. Delays in self-report, in particular, may affect the accuracy of behavioral data. Objective: The aim of this study was to capitalize on mobile phone use by young men who have sex with men (YMSM), and examine the use of an interactive voice response system (IVRS) by YMSM aged 18-30 years enrolled in a multisite, 12-week microbicide safety and acceptability trial. Methods: YMSM (N=95) enrolled across 3 sites (Boston, Pittsburgh, and San Juan) were asked to report their use of an applicator applied placebo rectal gel product during receptive anal intercourse (RAI) using the IVRS. IVRS was available in Spanish and English. After the 12-week trial, we examined whether IVRS problems were associated with YMSM's sociodemographic characteristics (eg, age, race and ethnicity, and education), sexual behavior, or recruitment site. We used a multinomial logistic regression to compare YMSM who experienced no IVRS problems (n=40) with those who reported one IVRS problem (n=25) or two or more IVRS problems (n=30). Results: We recorded 1494 IVRS calls over 12 weeks. Over half of the participants (55/95; 58%) experienced challenges using the IVRS during the 12-week trial. YMSM reporting greater RAI occasions during the trial were more likely to experience one (odds ratio [ OR]=1.08, 95% CI (1.02-1.14); P <=.01) or more (OR=1.10, 95% CI (1.03-1.16); P <=.001) IVRS challenges. Greater educational attainment was associated with multiple IVRS challenges (OR=7.08, 95% CI (1.6-31.6); P <=.01). Participants in the Puerto Rico site were most likely to report multiple IVRS problems. Conclusions: Although IVRS was a useful data collection technology in our trial, several challenges experienced by English and Spanish speaking YMSM diminish its overall acceptability. We discuss strategies to optimize future development of IVRS data quality protocols based on lessons learned.