OBJECTIVES: In 1993, the government of Pakistan started a new approach to the delivery of contraceptive services by training literate married women to provide doorstep advice and supplies in their own and neighboring communities. This report assesses whether this community-based approach is starting to have an impact on contraceptive use in rural areas. METHODS: A clustered nationally representative survey was used to collect data on contraceptive use and access to services in each cluster. Two-level logistic regression was applied to assess the effects of service access, after potential confounders were taken into account. RESULTS: Married women living within 5 km of 2 community-based workers were significantly more likely to be using a modern, reversible method of contraception than those with no access (odds ratio = 1.74; 95% confidence interval = 1.11, 2.71). CONCLUSIONS: After decades of failure, the managers of the family planning program have designed a way of presenting modern contraceptives that is appropriate to the conditions of rural Pakistan. The new community-based approach should be steadily expanded.