In Bangladesh, family planning workers' visits reduce the costs of contraception and may increase the demand. If visits increase demand or if workers are targeting their visits, past visits by family planning workers should have a positive and significant effect on later probabilities of adopting contraceptive methods. Longitudinal data show that past visits are not significant in hazard models for adoption of contraceptive methods, whereas visits in the current round are significant. Therefore family planning workers' visits affect women's contraceptive behavior by decreasing the costs of contraception. Results of contraceptive discontinuation hazard models further support this hypothesis.