Fall-Related Medications and Fall Risk as Predictors of Fall-Related Hospitalization Among Community-Dwelling Older Adults.

Abstract

Purpose: To examine fall-related medications and fall risk as predictors of fall-related hospitalizations among a cohort of community-dwelling older adults in the southern tier of New York State.

Method: Participants comprised 8,499 patients from 14 primary care practices within a large community hospital system. Bivariate and multivariate analyses were conducted to examine associations between fall-related medications and fall risk, and their interactions as predictors of fall-related hospitalizations.

Results: There was a statistically significant association between fall risk and Fall Risk Medication Score (FRMS) with a cut-point ≥6 (χ² = 85.145, p < .001), with a weak but significant positive correlation (φ = 0.100, p < .001). Older adults who had fall risk were 1.738 times more likely to have a FRMS ≥6 (95% confidence interval [CI] [1.544, 1.956]). Similarly, there was a statistically significant association between fall risk and FRMS with a cut-point ≥10 (χ² = 68.953, p < .001), with a weak but significant positive correlation (φ = 0.090, p < .001). Older adults who had fall risk were 2.348 times more likely to have a FRMS ≥10 (95% CI [1.909, 2.887]). In logistic regression analyses, the interaction between fall risk and FRMS was a significant predictor of hospitalization when controlling for age and sex (p < .001); however, FRMS did not significantly improve the model over fall risk alone.

Conclusion: Current tools for assessing and weighing the personal, environmental, and pharmacological factors that influence the likelihood of falls among older adults could be significantly improved through larger cohort studies using significant endpoints, such as fall-related hospitalizations.