Consistency in women's reports of sensitive behavior in an interview mode experiment, São Paulo, Brazil.

Abstract

Inaccurate reporting of sexual behavior creates a misleading picture of individuals' risk for STI infection. Despite a substantial body of U.S. research on the consistency of self-reports of sensitive behavior, only a few such studies have been conducted in developing countries. Consistency in the reporting of sexual activity and other sensitive behaviors was assessed among 818 women aged 18-40 who enrolled in 2004 in a study examining STI screening and diagnosis in São Paulo, Brazil. Participants were randomized into face-to-face interview and audio computer-assisted self-interview (audio-CASI) groups, and a six-week follow-up interview was conducted using audio-CASI for all participants. Differences between groups were assessed using t tests, and logistic regression analyses were used to estimate the likelihood of inconsistency within the enrollment interview and between the enrollment and follow-up interviews. Consistency in reporting at the enrollment interview was higher in the face-to-face group than in the audio-CASI group, likely because interviewers prompted women to reconcile discrepant responses, whereas the audio-CASI program did not enforce logical consistency. However, consistency between enrollment and follow-up was significantly lower in the face-to-face group for abortion, marijuana use, transactional sex, coerced sex and number of lifetime sexual partners, because of increased reporting at follow-up using audio-CASI. Although the analysis of internal consistency at enrollment suggests that computerized interviewing may increase random measurement error, it appears to reduce social desirability bias and encourage higher reporting of sensitive behaviors.