Objective: To examine the relationship between administration of selected medications and falls experienced by hospitalized elderly patients. Benzodiazepines and other medications previously associated with falls in elderly patients residing in the community and nursing homes were the primary focus. Design: Retrospective case control. Setting: Private, not-for-profit, 575-bed acute care hospital. Participants: A total of 100 patients who had fallen and 100 control patients, aged at least 70 years, admitted during the same 17-month time period. Main Outcome Measures: We examined the relationship between falls and patient demographics, underlying disease states, number of concurrent disease states, and length of hospitalization. Possible associations between the administration of narcotics, benzodiazepines, antidepressants, antipsychotics, other sedating agents, antihypertensives, diuretics, nitrates, and digoxin 48 hours prior to the fall or reference day were explored. The relationships between benzodiazepine half-life, dosage, administration frequency, cumulative dose, and falls were also examined. Results: Demographically the groups were similar except that patients who had fallen were hospitalized significantly longer (mean 18.8 vs 12.2 d; p < 0.00001) than control patients. Benzodiazepines were received by more (40% vs 20%, odds ratio = 2.67) patients who had fallen than control patients. The use of long (>24 h) half-life benzodiazepines was similar in patients who had fallen (48%) and control patients (45%). Long half-life benzodiazepines were commonly administered (65%) to patients who had fallen in doses greater than that recommended for the elderly. Benzodiazepine use, expressed as milligrams of diazepam equivalents received during the 48-hour study, was higher in patients who had fallen than in control patients (15.00 ± 17.80 vs 9.73 ± 6.58 mg), but this was not statistically significant (p = 0.1030). Congestive heart failure (37% vs 24%), digoxin therapy (35% vs 22%), or administration of 3 or more psychoactive agents (17% vs 4%) were all more common in patients who had fallen than in control patients. Conclusions: Falls experienced by the elderly patients in our acute care institution were associated with the presence of congestive heart failure along with digoxin therapy, benzodiazepine use, or the use of at least 3 psychoactive agents.
Malnutrition and Left Ventricular Systolic Function in Hospitalized Elderly Patients with and without Heart Failure