A small area analysis estimating the prevalence of addiction to opioids in Barcelona, 1993

Abstract

Study objective-To determine the distribution of opioid use prevalence in small areas and its relation with socioeconomic indicators. Design-Capture-recapture was applied using data from the Barcelona Drug Information System for 1993 (treatment demands, hospital emergency room visits, deaths from heroin acute adverse reaction and pre-trial prison admissions). To avoid dependence between sources, a log-linear regression model with interactions was fitted. For small neighbourhoods, where capture-recapture estimates were not obtainable, the Heroin Problem Index (HPI) was used to predict prevalence rates from a regression model. The correlation between estimated opioid use prevalence by neighbourhoods and their socioeconomic level was computed. Main results-The city's estimated prevalence was 12.9 opioid addicts per 1000 inhabitants aged 15 to 44 years (95% CI:10.1, 17.2), which represents 9176 persons. The highest rate was found in the inner city neighbourhood. Comparing rates obtained for each neighbourhood with their unemployment rates, a high correlation coeYcient was obtained (r = 0.80, p < 0.001). Conclusion-The main contribution of this study is that of combining capturerecapture with the HPI to produce small area prevalence estimates, which would not have been possible using only one method. Areas with higher socioeconomic status showed proportionally low addiction prevalences, but in depressed areas, prevalences varied widely.

Evaluating the impact of methadone maintenance programmes on mortality due to overdose and aids in a cohort of heroin users in Spain

“…The discrepancy could be explained by the fact that users who had never been in treatment present higher prevalence of risk behaviours for overdose [11], as well as by the fact that the specific register collects deaths occurring in the city whether or not they were resident there. Furthermore, if we compare the 1993 overdose mortality rate in our cohort (2.3 per 100 person‐years) with the one calculated with the data from the 1993 specific mortality register and a city capture–recapture estimate of heroin users [40] (2.1 per 100 users), we can see the similarity of both rates, supporting the fitness of our cohort.…”

Estimating the likely true changes in rheumatic fever incidence using two data sources

“…Capture-recapture analyses have been used extensively in infectious disease research using methodologies similar to ours, and have proved useful in the public health field [30][31][32][33][34][35][36][37]. Different case numbers were noted for the period 1997-2011, compared with those identified in our original study [14].…”

Drugscapes and the role of place and space in injection drug use-related HIV risk environments

“…There is a long tradition of comparative research in geography, sociology and public health, in which units of analysis are geographic areas and dependent variables are either health outcomes or measures of health-promoting programs. A variety of studies examine relationships over time showing statistical associations of MSAs or neighbourhood-level characteristics like income inequality, poverty and mortality (Kaplan et al, 1996;Kennedy et al, 1996;Lynch et al, 1998), arrests related to illegal drugs (Hsing, 1996), drug use and initial cocaine usage (Bell et al, 1998;Petronis & Anthony, 2003), addiction to opioids (Brugal et al, 1993), tuberculosis, and AIDS (Wallace and Wallace, 1998), and HIV infection associated with drug use (Laktin et al, 2005;Maas et al, 2007;Ciccarone & Bourgois, 2003).…”