Association of Parkinsonism or Parkinson Disease with Polypharmacy in the Year Preceding Diagnosis: A Nested Case-Control Study in South Korea.

INTRODUCTION

Published studies on the association between polypharmacy and parkinsonism or Parkinson disease are very limited.

OBJECTIVE

The objective of this study was to investigate whether polypharmacy is associated with parkinsonism or Parkinson disease in elderly patients.

METHODS

From a South Korean national health insurance sample cohort database for 2002-2013, we matched parkinsonism cases (defined by diagnosis codes for parkinsonism/Parkinson disease) and Parkinson disease cases (patients who had records for both Parkinson disease diagnosis and anti-Parkinson disease drug prescriptions) with controls. Logistic regression analysis evaluated the associations of parkinsonism/Parkinson disease with polypharmacy (i.e., five or more prescribed daily drugs) during the year preceding parkinsonism/Parkinson disease diagnosis, medications potentially associated with parkinsonism, and comorbidity status (using the Charlson Comorbidity Index score and hospitalization records).

RESULTS

The study population included 6209 cases and 24,836 controls for parkinsonism and 1331 cases and 5324 controls for Parkinson disease. In univariate logistic regression, odds ratios for parkinsonism/Parkinson disease increased significantly with increased polypharmacy, medications potentially associated with parkinsonism, Charlson Comorbidity Index score, or prior hospitalizations. In multiple logistic regression, odds ratios for parkinsonism/Parkinson disease (adjusted for medications potentially associated with parkinsonism and comorbidities) also increased with increased polypharmacy. Odds ratios (95% confidence interval) for Parkinson disease were higher than those for parkinsonism with stronger statistical significance: 1.41 (1.28-1.55) and 2.17 (1.84-2.57) for parkinsonism and 2.87 (2.30-3.58) and 4.75 (3.39-6.66) for Parkinson disease for between five and ten prescribed daily drugs and ten or more drugs, respectively.

CONCLUSIONS

Polypharmacy in the year preceding diagnosis may be associated with an increased risk for parkinsonism/Parkinson disease. Medications potentially associated with parkinsonism were assumed to increase the risk for parkinsonism/Parkinson disease, but more studies are required to confirm this relationship.