BACKGROUND: We examined the use and acceptability of a combination product (diaphragm and gel) compared to a single product (gel) during a 6-month safety trial in Zimbabwe. STUDY DESIGN: Women were randomized to the use of a diaphragm with gel or the use of gel alone, in addition to male condoms. Ever use and use of study product on the last act of sexual intercourse were assessed monthly by Audio Computer-Assisted Self-Interviewing. Acceptability, correct use and consistent use (use at every sexual act during the previous 3 months) were measured on the last visit by face-to-face interview. Predictors of consistent use were examined using multivariate logistic regression analyses. RESULTS: In this sample of 117 sexually active, monogamous, contracepting women, rates of consistent use were similar in both groups (59.7% for combination method vs. 56.4% for gel alone). Product acceptability was high, but was not independently associated with consistent use. Independent predictors of consistent use included age [adjusted odds ratio (AOR)=1.08; 95% confidence interval (95% CI)=1.01-1.16], consistent condom use (AOR=3.85; 95% CI=1.54-9.63) and having a partner who approves of product use (AOR=2.66; 95% CI=1.10-6.39). CONCLUSIONS: Despite high reported acceptability and few problems with the products, the participants reported only moderate product adherence levels. Consistent use of condoms and consistent use of products were strongly associated. If observed in other studies, this may bias the estimation of product effectiveness in future trials of female-controlled methods.