Objective:

We performed an agnostic drug-wide association study (DWAS) in Norway estimating the association between all drugs prescribed in Norway and the subsequent incidence of Parkinson's disease (PD).

Background:

Several drugs including treatments for cardiovascular diseases have been associated with a reduced incidence of PD.

Design/Methods:

We employed data on drug use in the entire Norwegian population from 2004 to 2019 (4.6 million) registered in the Norwegian Prescription Registry, including more than 600 million individual prescriptions. Drug classes were screened according to ATC codes at level 2, corresponding to therapeutic subgroups. Individuals who developed PD during the 15 years of follow-up (n=15,849) were identified using an algorithm based on drugs prescribed to treat the disease. Cox regression models with exposure as time-dependent variable were specified, and all p-values were corrected for multiple testing using the false discovery rate.

Results:

We found several novel associations between drug classes and the incidence of PD including a decreased incidence of PD among users of drugs acting on the renin-angiotensin system (0.86, 95%CI: 0.83-0.89). We also found an inverse association between calcium channel blockers and PD incidence, but we could not confirm the previously reported association between lipid-modifying agents and PD. Further, beta blocking agents and drugs for cardiac therapy were associated with an increased PD incidence. We also found increased PD incidence for drugs used in the treatment of prodromal symptoms of PD, including urologicals, drugs for constipation, anti-psychotics, and antidepressants.

Conclusions:

We observed two large drug classes of anti-hypertensives being associated with a subsequent reduced incidence of PD. The finding of two other cardiovascular drug classes being associated with an increased PD incidence calls for more detailed epidemiological studies and experimental studies to elucidate possible mechanisms.