Estimates of contraceptive prevalence based on service statistics and surveys in Gujarat State, India.

This report demonstrates that the estimates of contraceptive prevalence based on surveys are substantially lower than those based on the service statistics generated by the Indian family planning program. The reasons for this discrepancy were examined by contacting a subsample of acceptors recorded by female health workers as users in their registers. This inquiry indicated that the health workers themselves knew that 15-39 percent of the women who had been recorded as users of reversible methods were not really using them. About 19-27 percent of the recorded users of IUDs and only 3-4 percent of the recorded condom users confirmed use of the devices. Overall, the nonusers formed 59 and 64 percent of the recorded users contacted in Bharuch and Panchmahals districts, respectively. The estimates of nonuse of contraceptives in the follow-up survey are high enough to reconcile most of the observed discrepancy between the two sets of estimates of contraceptive prevalence.