Pattern of Statin Use in Southern Italian Primary Care: Can Prescription Databases Be Used for Monitoring Long-Term Adherence to the Treatment?

Objectives: We sought to evaluate the prescribing pattern of statins according to national and regional health policy interventions and to assess specifically the adherence to the therapy in outpatient setting in Southern Italy. Methods: A population-based study was performed on persons >= 15 years old, living in the catchment area of Caserta (Southern Italy), and registered in Arianna database between 2004 and 2010. Prevalence and incidence of new treatments with statins were calculated for each year and stratified by drug. Adherence to therapy was measured by Medication Possession Ratio. Sub-analyses by individual compound and type of cardiovascular prevention were performed. Results: From 2004 to 2010, the one-year prevalence of statin use increased from 44.9/1,000 inhabitants to 79.8/1,000, respectively, consistently with the incidence of new use from 16.2/1,000 to 19.5/1,000, except a slight decrease after criteria reimbursement revision on 2005 (13.3/1,000). The incidence of new treatments decreased for atorvastatin, and increased for simvastatin over the study years. Overall, 43% of new users were still highly adherent to the treatment (MPR >= 80%) after six months, while 26% after 4-years of follow-up. As compared with highly adherent patients, the probability to be non-adherent (MPR <= 25%) at 4-years of follow-up was 26% higher for women than for men (full adj. odds ratio: 1.26; 95% CI: 1.10-1.45), and 64% higher in patients who started on primary rather than on secondary prevention (1.64; 1.29-2.07). Conclusions: Prevalence and incidence of statin use increased consistently with health policy interventions. Only one-fourth of patients who newly initiated a statin were adherent to the treatment after 4-year of follow-up. Since the benefits of statins in terms of cardiovascular outcome and costs are associated with their chronic use, the identification of patient-related predictors of non-adherence such as gender, primary prevention could be suitable for physicians to improve the patients' compliance.