In settings where inconsistent, long-term use of the injectable and other short-term methods is common, the contraceptive prevalence rate calculated from responses to questions in cross-sectional surveys may overestimate the proportion of women protected against unintended pregnancy, according to “Contraceptive Dynamics in Rural Northern Malawi: A Prospective Longitudinal Study,” by Aisha Dasgupta et al., of the London School of Hygiene and Tropical Medicine. Contraceptive service data from a one-year prospective study in Karonga, a demographic surveillance site, indicated that only 51% of injectable users received their first reinjection on time, and just 15% received all their reinjections on time during the 12-month study period. In addition, the proportion of women in Karonga using the injectable at seven months after uptake, as estimated from service data (14%), was considerably lower than the proportion estimated from cross-sectional data from the Malawi Demographic and Health Survey (20%), suggesting that the former estimate picked up discontinuation and gaps in coverage not captured by women’s responses to questions about contraceptive use in the survey.