RESUMO
OBJECTIVE: To investigate the strength of association between impaired functional status and long-term (3-year) mortality in Chinese nursing home older adults. DESIGN: A 3-year prospective multicenter cohort study. SETTING: Nine nursing homes in Hong Kong. PARTICIPANTS: A total of 672 nursing home older adults (224 men; 448 women), mean age 85.0 ± 7.4. MEASUREMENTS: Functional statuses of participants were assessed by Barthel Index (BI) and participants were stratified into different groups according to their BI score: BI score 100, BI score 75-95, BI score 45-70, BI score 15-40, and BI score 0-10. Other covariates included age, sex, comorbidities, score of abbreviated mental test, serum albumin, serum creatinine, serum hemoglobin, and hospitalization in the preceding year. The outcome measures were the 1-year, 2-year, and 3-year all-cause mortality. RESULTS: Older adults with lower BI score had significantly higher all-cause mortality and this trend persisted in 1-year, 2-year, and 3-year mortality (P < .001). After multivariate analysis, there was a dose-response relationship in hazard ratio (HR) between BI score and 3-year all-cause mortality (compared with BI score 100; BI score 75-95: HR 1.38 [CI: 1.00-2.56; P < .05]; BI score 45-70: HR 1.80 [CI: 1.04-3.11; P < .001]; BI score 15-40: HR 2.12 [1.21-3.70; P < .001]; BI score 0-10: HR 3.13 [1.82-5.41; P < .001]; and trend test P < .05). Similar relationships were found in 1-year and 2-year mortality. CONCLUSION: Impaired functional status is associated with higher short-term and long-term mortality with a dose-response relationship in Chinese nursing home older adults.