Comparison of Sexual Behavior Data Collected Using a Coital Diary and a Clinic-Based Interview During a Microbicide Pilot Study in Mwanza, Tanzania

Abstract

The performance of coital diaries (CDs) and clinic-based interviews to measure sexual behavior was compared during a pilot study for a Phase III microbicide trial. In Mwanza, 59 women were enrolled for 4 weeks and provided with 20 placebo gels. Weekly, women were given CDs to complete daily. At the final clinic visit, women attended a face-to-face interview (Clinic FFI) about their sexual behavior, and the gel use was accounted for (gel accountability (GA)). Comparisons were made between CD, Clinic FFI, and GA data. In-depth interviews following clinic visits elicited reasons for discrepancies in reports. Twice as many sex acts during 1 week were recorded in the CD (median 4) compared with the clinic FFI (median 2). At the clinic FFI, more women reported using the gel for each sex act (84% vs. 40%; P < 0.001) and vaginal washing for each sex act (98% vs. 56%; P < 0.001) compared with the CD. Over 4 weeks, 16.4% of women recorded sex during menstruation in CDs compared with 1.8% at the clinic visit (P = 0.01). The median number of gels used reported in the CDs was the same as the GA (10) with 59% agreement on the number used within +/-2 gels. Reasons for misreporting during clinic FFI were reported to have been poor recall, embarrassment, or misunderstanding. Inaccuracies in CDs were attributed to misunderstanding or poor recording. CDs elicited higher recording of sex acts and lower reporting of gel use than clinic FFIs, which has implications for measuring adherence during clinical trials. With clear instructions and support, coital dairies should be considered in future microbicide trial design.