OBJECTIVE: Including the birth cohort dimension improves trend studies of mortality and health. We investigated the effect of including the birth cohort dimension in trend studies of prescription drug use by studying prevalence of statin use among adults. STUDY DESIGN AND SETTING: Data from a drug prescription database in the Netherlands (IADB.nl) were used to obtain the number of users of statin per 1,000 population (prevalence) in the age range 18-85 years from 1994 to 2008. We applied descriptive graphs and standard age-period-cohort (APC) models. RESULTS: From 1994 to 2008, the prevalence increased from ∼10 to ∼90 users per 1,000 population, with the peak in prevalence shifting from age 63 to 78 years. The APC model shows patterns that were masked in the age-period (AP) model. The prevalence rate ratio increased from the 1911 birth cohort to the 1930 birth cohort and then declined. Similar for both sexes, adding nonlinear period effects contributed ∼4.4% to reductions in deviance, whereas adding nonlinear birth cohort effects contributed ∼12.9%. CONCLUSION: Adding the birth cohort dimension to AP analysis is valuable for academic and professional practice as trends can be more accurately described and explained and it can help improve projections of future trends.