Abstract P151: Duration and Types of Statin Use and Long-Term Risk of Type 2 Diabetes Among Men and Women with Hypercholesterolemia: Findings From 3 Prospective Cohorts
Introduction: Although effective for cardiovascular disease (CVD) prevention, statin use has been associated with an elevated risk of type 2 diabetes (T2D). However, data on duration of use and types of statins are still limited.
Hypothesis: We hypothesized that longer duration of statin use was associated greater risk of T2D and the association varied with different types of statin.
Methods: We analyzed participants with self-reported hypercholesterolemia from the Nurses’ Health Study (NHS, 30 480 participants), the Nurses’ Health Study II (NHS II, 21 773 participants) and the Health Professionals Follow-up Study (HPFS, 12 064 participants) who were free of diabetes, CVD and cancer at baseline. Information on statin use was self-reported by participants every 2 years from 2000 for NHS and HPFS, and 1999 for NHS II. Incident cases of T2D were first self-reported on the biennial questionnaires and then confirmed by a validated supplementary questionnaire through 2010 or 2011. Participants also reported their serum cholesterol level every two years. Multivariable Cox models were used to estimate the hazard ratios (HRs) of developing T2D and their 95% confidence intervals (CIs) by statin use status.
Results: We documented a total of 4 555 incident T2D cases during the follow-up. Compared to nonusers, statin users had a significantly higher risk of T2D after adjustment for potential lifestyle confounding variables, hypertension medication and self-reported serum cholesterol levels (pooled HR, 1.38, 95% CI, 1.30-1.48). Compared to non-use, the durations of statin use of 1-3, 4-6 and ≥7 years were associated with HRs of 1.30 (95% CI, 1.19, 1.42), 1.41(95% CI, 1.29, 1.54) and 1.47 (95% CI, 1.33, 1.63), respectively, with significant linear trends in both the entire study population (P trend < 0.001) and statin users (P trend =0.05). The HRs for T2D were 1.10 (95% CI, 0.84, 1.44) for lovastatin, 1.17 (95% CI, 0.99, 1.39) for pravastatin, 1.30 (95% CI, 1.14, 1.49) for simvastatin, 1.36 (95% CI, 1.22, 1.52) for atorvastatin and 1.56 (95% CI, 1.30, 1.86) for rosuvastatin. In a series of sensitivity analyses by excluding T2D cases in the early follow-up, using propensity-score adjustment for potential confounders and restricting analysis to only new statin users, the results did not change appreciably.
Conclusions: The positive association between statin use and T2D was more pronounced with longer duration of use and the association varied with different types of statins.
Author Disclosures: D.D. Wang: None. Y. Li: None. A. Pan: None. E. Giovannucci: None. W. Willett: None. J. Mason: None. F. Hu: None.
- © 2015 by American Heart Association, Inc.