Using data from the 1982 National Survey of Family Growth (NSFG), this analysis reports differentials in contraceptive discontinuation among married women aged 15-44 years in the United States. The total discontinuation rate is broken down into change to no method (termination) or to a different method (a method switch), and rates are obtained for specific methods. In addition, sociodemographic differences in risks associated with each type of discontinuation are shown. Discontinuation rates are compared to use-failure rates to provide a more comprehensive understanding of the implications of discontinuation for contraceptive efficacy.