Estimating the size of the cannabis using population to evaluate drug policies: A comparison between different metropolitan areas in Italy

Speaker: Flavia Mascioli (University Sapienza Rome, Faculty of Sciences, Italy).

Applying capture-recapture methods, we estimate the size of the population of cannabis users in different metropolitan areas in Italy who are at risk of being registered for personal use of drugs according to the law presently enforced ( Since 2006 the quantity of drugs possessed has determined the threshold between personal possession and trafficking. This is the first attempt in Europe to apply capture-recapture techniques to this particular sub-population and, consequently, not only improve the knowledge of the Italian situation, but to provide the first step in trying to evaluate the impact of drug policies, in order to prevent initiation, and to plan possible intervention policies, specifically secondary prevention. Most problem drug-users started their career using cannabis, and around 10% of young cannabis users proceed to become problem drug users after an average of 2.5 years. It is therefore crucial to try to prevent this step, and to know how many cannabis users are at risk of being apprehended by the authorities (mainly street users). A single source capture-recapture model, the truncated Poisson model, was used, and Chao and Zelterman’s estimators were obtained. Confidence intervals were obtained by the bootstrap approach. Males and females, divided by age and class, were treated separately, and the period 2005-2008 is covered. Studying the population through these years is of particular interest because of the change of law in 2006, which, from a statistical point of view, produced a modification in the “case definition”; Before the year 2006, possession of illicit drugs was prosecuted in accordance with the principle of mandatory prosecution, independently of the quantity of the substance. The hypothesis is that the new case definition produces a displacement of subjects from the consumers archive, to the dealers archive, and that this phenomenon is more pronounced in the main metropolitan areas. The data set has been provided by the Italian Ministry of the Interiors, and comprises several variables for each subject. The different subjects are univocally identified, so that it is possible to count the number of captures in a specified period. The areas considered are Rome, Milan, Turin, Genoa, Bologna, Naples, Bari, Palermo and Cagliari, which comprise (in 2005) about 36% of the population in the age group 15-64. According to the estimates obtained for these areas for the same period, the cannabis users in these areas are around 44% of the total estimated for Italy. In the conclusion we will also focus on some challenges in interpreting the results, taking into account the analogous estimates of dealers and traffickers obtained using a different archive, but the same methodology. We will also focus on the challenges and the opportunities provided by using different archives and surveys. In particular we will use the General Population Survey, the School Population Survey and the Survey Among Low Threshold Clients for more in-depth examination and comparison. One of the most promising applications of the various analyses is the possibility of estimating the recent incidence of cannabis-use among young people on the basis of the estimated prevalence and the distribution of age at first use, estimated from various databases, and the stopping rate, estimated from the School Population Survey Data Set.


CV: Flavia Mascioli – Associate Professor of Statistics, Department of Mathematics, University La Sapienza, Rome. Research interests: Applied statistics and statistical education.

Over the past years Professor Mascioli has focused her research on capture-recapture methods. Recently she published a paper together with Carla Rossi applying this technique based on a single source to estimate prevalence indicators in order to measure the size and dynamics of drug markets in Italy. Flavia Mascioli is the author of a number of publications on different topics, namely mathematical statistics, biostatistics, statistical education.

She teaches the undergraduate and postgraduate courses Statistics for Biologists, Topics in Applied Statistics, Statistical Theory and Methods, Clinical Biostatistics.