Duration and type of statin use and long-term risk of type 2 diabetes among men and women with hypercholesterolaemia: findings from three prospective cohorts

Abstract

Aims/hypothesis

Findings from RCTs and observational studies indicate a positive association between statin use and risk of type 2 diabetes. Mendelian randomisation studies provide evidence to support that the effect is causal. However, little is known about the long-term effects, and data on different types of statins remain limited.

Methods

We analysed participants with hypercholesterolaemia from the Nurses’ Health Study (NHS; 30,510 participants), the Nurses’ Health Study II (NHSII; 21,547 participants) and the Health Professionals Follow-Up Study (HPFS; 9934 participants) who were free of diabetes, CVD and cancer at baseline. Statin use was assessed every 2 years starting in 2000 in the NHS and the HPFS and in 1999 in the NHSII. Incident cases of type 2 diabetes were confirmed by a validated supplementary questionnaire until the end of follow-up (31 January 2023).

Results

We documented 6762 incident type 2 diabetes cases during up to 23 years of follow-up. Compared with non-users, statin users had a significantly higher risk of type 2 diabetes after adjustment for BMI and other potential confounding variables (pooled HR 1.40; 95% CI 1.33, 1.48). Compared with non-use, durations of statin use of 1–5, 6–10, 11–15 and >15 years were associated with HRs of 1.36 (95% CI 1.27, 1.44), 1.41 (95% CI 1.31, 1.52), 1.60 (95% CI 1.44, 1.78) and 1.76 (95% CI 1.50, 2.06), respectively; significant linear trends were observed when the comparison included non-users and within statin users only (both p_trend<0.001). Compared with non-users, the HRs for type 2 diabetes associated with 10 year use of specific types of statins were 1.99 (95% CI 1.45, 2.73) for rosuvastatin, 1.66 (95% CI 1.12, 2.47) for lovastatin, 1.62 (95% CI 1.39, 1.89) for atorvastatin, 1.44 (95% CI 1.06, 1.97) for pravastatin and 1.37 (95% CI 1.13, 1.66) for simvastatin. Use of a low-potency statin for 10 years was associated with a 34% higher risk of type 2 diabetes (HR 1.34; 95% CI 1.15, 1.56), while use of a high-potency statin for 10 years was associated with a 72% higher risk (HR 1.72; 95% CI 1.46, 2.04). The difference in the 10 year cumulative risk of type 2 diabetes comparing statin users vs non-users was most pronounced in participants with the least healthy lifestyles (4.5% vs 3.1%), while the smallest risk differential was observed among participants who adhered to the healthiest lifestyles (1.0% vs 0.4%).

Conclusions/interpretation

The positive association between statin use and type 2 diabetes was more pronounced with a longer duration of use, and the association varied across different types of statins. Adopting and maintaining a healthy lifestyle can serve as a viable approach to diabetes prevention during statin treatment.

Graphical Abstract