Assessing the effect of a guideline change on drug use prevalence by including the birth cohort dimension: the case of benzodiazepines

摘要: Purpose The aim of this study was to investigate whether including the birth cohort dimension in time series analysis leads a more accurate estimation (long-term) effect guideline change on trend benzodiazepine use. Methods We calculated age-specific (20-84years) and sex-specific prevalence use per 1000 population quarter year (1998 2008) using prescription database set Netherlands. studied over by age group within cohorts through interrupted analyses estimate 2001. Results From 1998 2008, overall age-standardized declined from similar 54 for men 107 women 45 85 women. relative increased significantly after 2001 both sexes majority groups. Within cohorts, with until leveled thereafter. age-period approach had worse model fit indicators than within-cohort predicted larger long-term effects approach. projection estimated 36% decline whereas birth-cohort 8% decline. Conclusion Explicitly following trajectories led models better fit; conventional stronger effect. This has important implications professional practice. Copyright (c) 2013 John Wiley & Sons, Ltd.