Incidence indicators for policy-making: models, estimation and implications
- Autores: L. Ravà, M.G. Calvani, S. Heisterkamp, Lambert Wiesing, C. Rossi
- Localización: Boletín de estupefacientes, ISSN 0251-7086, Vol. 53, Nº. 1-2, 2001
- Idioma: inglés
- Enlaces
-
Resumen
Modelling the incidence of first opiate use can be a useful tool for understanding the process of diffusion of drug use in space and time. Such information is important in evaluating current and future needs for and effects of services and interventions.
Back calculation can be used in the analysis of spatial and time trends of incidence of drug users who will eventually seek treatment. This is done using incidence data of drug users entering first treatment and an external estimate of the latency period distribution, that is, the time lapse between first use and first treatment. The authors of the present article studied the heroin epidemic in Italy using Empirical Bayesian Back Calculation, at the national level and for eight regions of the country, using data from public treatment services for the period 1986-1998, which are routinely collected by the Ministry of Health. After a period of rapid increase during the 1980s, the nationwide aggregated incidence curve seemed to stabilize at the beginning of the 1990s. While that trend might possibly indicate some success in prevention efforts, it could well be due to the fact that the natio nal epidemic is the combination of several local epidemics that emerge, saturate and move geographically. That hypothesis is confirmed by the regional analysis, which provides a picture of several sub-epidemics showing peaks of different size, location and time. In particular, Liguria, Piedmont and Emilia Romagna peaked more or less simultaneously before the other regions, followed by Tuscany, Latium and Sardinia (which show the highest incidence per million inhabitants), while Trentino and Friuli never peaked.
The results of the Empirical Bayesian Back Calculation form an important evidence base for planning, monitoring and evaluating preventive interventions and for forecasting treatment needs in the near future, at both the regional and the national level.
