We studied the association between the statin dosage and the risk of Parkinson disease (PD) in diabetic patients in Taiwan.

One million patients were randomly sampled from a National Health Insurance (NHI) database and followed from 2001 to 2008. Diabetic patients were screened by diagnosis of International Classification of Diseases, Ninth Revision, Clinical Modification (ICD‐9‐CM) codes, and statin dosage was determined according to the NHI pharmacy database. PD was diagnosed on the basis of ICD‐9‐CM codes and anti‐Parkinson medication use. Statin users was classified by statin dose‐duration‐day > 28 and matched with nonusers of statins using a coarsened exact matching method. There were 50,432 patients, and half of them were statin users. We examined the risk of PD between statin users and nonusers of statins and further tested the trends of the relative risk between the statin dosage and PD.

The PD incidence rate was lower in statin users than in nonusers of statins. The crude hazard ratio of PD incidence in statin users was 0.65 (95% confidence interval [CI] = 0.57–0.74) in females and 0.60 (95% CI = 0.51–0.69) in males compared with nonusers of statins. After Cox regression analysis, all statins except lovastatin exerted protective effects on PD incidence and had a significant dose‐dependent trend.

In Taiwanese diabetic patients, the risk of PD is lower in statin users than in nonusers of statins. Statin users, except lovastatin users, are dose‐dependently associated with a decreased incidence of PD compared with nonusers of statins. This finding provides a new indication for statin beyond lipid control and cardiovascular events in diabetic patients. Ann Neurol 2016;80:532–540