Abstract
PURPOSE: A previous meta-analysis of randomized trials did not confirm findings from observational studies that suggested that statins reduce the risk of infection. However, animal experiments indicate that statins may be more effective in reducing the risk and/or the severity of infection among patients with diabetes. Hence, we evaluated the effect of statins on antibiotic prescriptions (a proxy for infections) among patients with drug-treated type 2 diabetes using two confounding-reducing observational designs. METHODS: We conducted a prescription sequence symmetry analysis and a cohort study using the IADB.nl pharmacy prescription database. For the prescription sequence symmetry analysis, a sequence ratio was calculated. The matched cohort study, comparing the time to first antibiotic prescription between periods that statins are initiated and non-use periods, was analyzed using stratified Cox regression. RESULTS: Prescription sequence symmetry analysis of 4684 patients with drug-treated type 2 diabetes resulted in an adjusted sequence ratio of 0.86 (95% confidence interval [CI]: 0.81 to 0.91). Corresponding figures for the cohort analysis comparing 9852 statin-initiation with 4928 non-use periods showed similar results (adjusted hazard ratio: 0.88, 95%CI: 0.83 to 0.95). CONCLUSIONS: These findings suggest that statins are associated with a reduced risk of infections among patients with drug-treated type 2 diabetes. © 2016 The Authors. Pharmacoepidemiology and Drug Safety Published by John Wiley & Sons Ltd.