Frailty and hospitalization of long-term stay nursing home residents

OBJECTIVE: To determine the relationship between characteristics of older, long-term stay nursing home patients and hospitalization. DESIGN: A cohort study. SETTING: One hundred three proprietary nursing homes. PARTICIPANTS: All residents of these nursing homes aged 65 or older admitted between January 1, 1991, and December 30, 1993, who had no transitions out of the nursing home during the first 6 months of their stay. MEASUREMENTS AND MAIN RESULTS: Among the cohort of 3782 residents, 931 (25%) were hospitalized at least once during the second 6 months of their nursing home stay. In a logistic regression model, severe functional impairment (adjusted odds ratio (AOR) 1.20, 95% confidence interval (CI) 1.01, 1.43), worsening ADL self-performance (AOR 1.22, 95% CI 1.04, 1.43), presence of a decubitus ulcer (AOR 1.62, 95% CI 1.17, 2.24), presence of a feeding tube (AOR 2.03, 95% CI 1.54, 2.67), primary diagnosis of congestive heart failure (AOR 1.61, 95% CI 1.11, 2.34), and primary diagnosis of respiratory disease (AOR 1.77, 95% CI 1.24, 2.54) were associated with hospitalization. No form of advance directive was associated with a lower rate of hospitalization. CONCLUSIONS: Physically frail patients, who may be the least likely to benefit from hospitalization, are the most likely to be hospitalized. The lack of an association between ''Donot-hospitalize'' orders and lower rates of hospitalization suggests that there are substantial barriers to providing acute care in the nursing home. The association between recent functional decline, primary diagnoses of congestive heart failure and respiratory disease, and hospitalization indicates, however, that certain patient groups may be targeted successfully to reduce hospitalization rates.