Estimating ten years prevalence trends of problematic cocaine use 1999-2008 in Barcelona, Spain. Problems encountered
METHODS: Data came from Emergency Room (ER) registers from Barcelona’s four major university hospitals. Individuals were categorized into two subgroups according to heroin use during the index year: cocaine plus heroin and cocaine without heroin. The full-year ER register was separated into three consecutive 4-month periods, considering each period as a different sample., Log-linear regression models were fitted to estimate the unknown population for each year and each drug subgroup. Models with all possible combinations of interactions were fitted to adjust for possible dependencies between samples. However, when dependence between the three capture periods was not rejected, the saturated model was preferred. 95% confidence intervals (95%CI) were calculated with a likelihood interval approach.
RESULTS: In 2008, mean age of detected problematic cocaine users was 34 years and 25% of them were women. The number of episodes per patient differed by drug subgroup. Estimates of the total number of problematic cocaine users increased from 6,028 (95%CI: 4,086-9,327) in 1999 to 22,640 (95%CI: 14,001-37,500) in 2006 and decreased thereafter. The consumption trend of cocaine plus heroin was stable over the period (0.10% in 1999, 0.15% in 2005 and 0.09% in 2008). Thus, two different trends were found depending on concomitant heroin use. However, changes in the information registered in clinical records, appear to have affected recent estimates.
CONCLUSIONS: Relying on secondary data to assess basic epidemiological indicators requires a thorough knowledge of how information is collected and possible changes in the procedures over time.