Characterising people that demanded drug treatment in Catalonia: a cluster analysis

We tried to characterise people admitted to treatment in 2017 in order to see whether this could help to improve drug care. With this aim, we performed a hierarchical cluster analysis using glower distance that allows us to use quantitative and categorical variables, using data from treatment demand indicator (including people admitted for alcohol and tobacco use disorder) and other data provided from the Catalan health system register. We used all cases with no missing values in selected variables (N = 8075) 59.9 % of people who started treatment in 2017. We identified four different profiles according to the results of cluster analysis. The first (n = 4491, 77% male) had 61% of alcohol and 26% of cocaine users and their mean age was 46. Sixty-five percent were working while 28% were pensioner. The second profile (n = 1472, 78% male) were 58% cannabis and 31% cocaine consumers. Their mean age was lower, 31 years, and 47% of people did not have a job, while 21% were working and 19% were studying. We also identified two more problematic drug profiles. One is similar to the first identified cluster (n = 1599, 74% male) but with worst socioeconomic indicators and a 3% of injectors. They were 72% alcohol and 16% cocaine users, and their mean age was 43 but all were unemployed. The last profile (n = 513, 86% male) is made up of 99% of heroin users and 46% of injectors. Their mean age was 43 and 22% were working, 36% were unemployed and 36% were in other situations. Identifying risk profiles could help public health officials to design better-suited programmes and practitioners to improve their interventions according to the complex needs of their clients.