Estimating Local and National Problem Drug Use Prevalence from Demographics

Abstract

Objective: To assess the applicability and validity of the Multiple Indicator Method for combining sparsely available local prevalence estimates of problem drug use into a national estimate. Method : Local estimates of problem drug use were available in seven out of the 27 addiction treatment regions in the Netherlands. Using demographic variables as predictors and under varying regression model specifications estimates were obtained for the remaining 20 regions and aggregated into national ones. These outcomes were compared with one obtained with the methodologically unrelated Treatment Multiplier Method. Results: All results point to a prevalence of 0.32% in the population of 15–55 years (or 0.27% in the population aged 15–64). Discussion: The applicability and validity of the Multiple Indicator Method – and adaptations thereof – for getting regional and national prevalence estimates from incomplete data remains subject of further research. However, the results found here and elsewhere more than justify an interest in future research efforts.

• Substance misuse
• Prevalence
• Epidemiology
• Regression models
• Multiple indicator method