Estimating Size of Drug Users in Macau: an Open Population Capture-Recapture Model with Data Augmentation Using Public Registration Data
Successful drug intervention policies rely on timely assessment of size of drug users and close monitoring of trend regarding drug abuse. Taking advantage of data collected by the Central Registration System for Drug Abusers of Macau (CRSDAM), this study applied open population capture-recapture mod...
| Authors: | ; |
|---|---|
| Format: | Electronic Article |
| Language: | English |
| Published: |
2018
|
| In: |
Asian journal of criminology
Year: 2018, Volume: 13, Issue: 3, Pages: 193-206 |
| Online Access: |
Volltext (Resolving-System) |
| Journals Online & Print: | Drawer...
|
| Check availability: | HBZ Gateway |
| Keywords: |
| Summary: | Successful drug intervention policies rely on timely assessment of size of drug users and close monitoring of trend regarding drug abuse. Taking advantage of data collected by the Central Registration System for Drug Abusers of Macau (CRSDAM), this study applied open population capture-recapture models with data augmentation techniques to estimate the size of drug users and to identify influential factors on capture and survival probabilities from 2009 to 2014. In particular, the data augmentation technique was used to address missing data issues. The estimated size of drug users has been slowly declining from 2442 (95% Bayesian credible interval [BCI] 2042-2914) in 2009 to 807 (95% BCI 654-1000) in 2014 with a small fluctuation of 2251 (95% BIC 1950-2599) in 2011, and the estimated cumulative size of drug users is 6199 (95% BCI 5651-6873), which is correspondent to a prevalence rate of 1.20% of the current population aged 15 to 54. Although the estimated sizes of total, narcotic, and other drug users were declining, the size of stimulant users might be increasing. People who used narcotics as their first drug and who were reported by governmental agencies were more likely to stay in the registration system, while those who used needle injection were less likely to stay. Governmental agencies, higher education, and using needle injection were negatively associated with the probability of capture over time, while using narcotics as the first drug was positively associated with it. |
|---|---|
| ISSN: | 1871-014X |
| DOI: | 10.1007/s11417-018-9266-0 |