Medication use and risk of falls among nursing home residents: a retrospective cohort study.

Background Geriatric falls are leading causes of hospital trauma admissions and injury-related deaths. Medication use is a crucial element among extrinsic risk factors for falls. To reduce fall risk and the prevalence of adverse drug reactions, potentially inappropriate medication (PIM) lists are widely used. Objective Our aim was to investigate the possible predictors of geriatric falls annualized over a 5-year-long period, as well as to evaluate the medication use of nursing home residents. Setting Nursing home residents were recruited from the same institution between 2010 and 2015 in Szeged, Hungary. Method A retrospective epidemiological study was performed. Patient data were analysed for the first 12 months of residency. Chi-squared test and Fisher's-test were applied to compare the categorical variables, Student's t test to compare the continuous variables between groups. Binary logistic regression analysis was carried out to determine the association of falls with other variables found significant in univariate analysis. Microsoft Excel, IBM SPSS Statistics (version 23) and R (3.2.2) programs were used for data analysis. Main outcome measure Falls affected by age, gender, number of chronic medications, polypharmacy, PIM meds. Results A total of 197 nursing home residents were included, 150 (76.2%) women and 47 (23.8%) men, 55 fallers (annual fall prevalence rate was 27.9%) and 142 non-fallers. Gender was not a predisposing factor for falls (prevalence in males: 23.4 vs 29.3% in females, p > 0.05). Fallers were older (mean years ± SD; 84.0 ± 7.0) than non-fallers (80.1 ± 9.3, p < 0.01). The age ≥80 years was a significant risk factor for falls (p < 0.001). The number of chronic medications was higher in male fallers (12.4 ± 4.0) than in non-fallers (6.9 ± 4.2, p < 0.001). Polypharmacy (taking four or more chronic medications) was a significant risk factor of falls (p < 0.01). Those PIMs carrying fall risk were taken by 70.9% of fallers and 75.3% of non-fallers (p > 0.05). Taking pantoprazole, vinpocetine or trimetazidine was a significant risk factor for falls. Conclusion Older age, polypharmacy and the independent use of pantoprazole, vinpocetine, and trimetazidine were found to be major risk factors for falls. Further real-life epidemiological studies are necessary to confirm the role of particular active agents, and to help professionals prescribe, evaluate and review geriatric medication use.