Prevalence and predictors of potentially inappropriate prescribing of central nervous system and psychotropic drugs among elderly patients: A national population study in Korea

Objectives: To investigate the prevalence of potentially inappropriate prescribing (PIP) of central nervous system and psychotropic (CNS-PS) drugs to the Korean elderly population, and to identify PIP-associated factors. Methods: Ambulatory care visits were identified from the 2013 National Aged Patient Sample (HIRA-APS-2013) data, composed of 20% random samples of all enrollees in the universal health security program aged >= 65 years. The CNS-PS section of Screening Tool of Older Person's potentially inappropriate Prescriptions (STOPP) criteria version 2 was used to identify PIP at these visits. Results: A total of 24,427,069 prescription claims records and 1,122,080 patients were included in the study; 10.73% of the claims and 53.64% of the patients satisfied at least one STOPP criterion in the prescription of CNS-PS drugs. The highest prevalence of PIP was observed for the criteria of first-generation antihistamines (FGAH), followed by tricyclic antidepressants (TCA) in patients with prostatism and TCA in patients with dementia. The generalized estimating equation logistic regression analysis showed that the PIP of FGAH was significantly associated with polypharmacy (5-9 drugs: odds ratio (OR) 4.965, 95% confidence interval (CI) 4.936-4.994; >= 10 drugs: OR 5.704, 95% CI 5.604-5.807), less severe health conditions (Charlson Comorbidity Index (CCI) = 2: OR 0.852, 95% CI 0.842-0.862; CCI = 1: OR 0.975, 95% CI 0.964-0.986), prescriptions from clinics (OR > 1.0), and outpatient care by general practitioners (OR > 1.0). Conclusions: Appropriate interventions to reduce PIP should be made, especially for the criteria that indicate a high PIP prevalence. Targeted strategies are necessary to modify the risk factors of PIP identified from this study.