PURPOSE: The purpose of this study was to examine falls as an outcome measure at 12 months for two-group comparison (delirium cases and noncases) and five-group comparison (noncases, hypoactive, hyperactive, mixed delirium cases, and subsyndromal delirium cases).
DESIGN: Three hundred and twenty patients enrolled in the Delirium in Rural Long-Term Care Facilities Study, which examined subjects for delirium during a 28-day surveillance period, were followed longitudinally for fall events for 12 months. Fall events were recorded and data analyzed using date of "first fall" as the referent for statistical analysis.
METHODS: Fall reports were completed for all subjects for 12 months following delirium surveillance. Fall records were reviewed and the number of days until the first fall event was calculated. Data were censored for deaths that occurred during the 12-month period.
FINDINGS: Five group analysis of variance (noncases, hyperactive delirium, hypoactive delirium, mixed delirium, and subsyndromal delirium) showed significant differences in Functional Independence Measure scores (p = .001) and number of medications (p = .001). The percentage of patients who fell was higher in all delirium subtypes than in noncases at 12 months and was statistically significant for subsyndromal subjects. Two patients, one each from the subsyndromal and mixed delirium subtypes, died from injuries sustained during falls.
CONCLUSIONS: Delirium appears to increase the risk for falls.
CLINICAL RELEVANCE: Increased surveillance and fall prevention strategies are needed for patients who experience transient cognitive changes such as delirium and subsyndromal delirium, even after delirium resolution.