Impact of living arrangements on clinical outcomes among older patients with dementia or cognitive impairment admitted to the geriatric evaluation and management unit in Taiwan

Aim To evaluate the impact of living arrangements on mortality and functional decline among older patients with dementia or cognitive impairment after discharge from a geriatric evaluation and management unit (GEMU) in Taiwan. Methods The present retrospective cohort study used data from the Veteran Affairs Comprehensive Geriatric Assessment from January 2015 to May 2016 for analysis. Data of patients aged 65 years and older with dementia or cognitive impairment at admission to the GEMU of Taipei Veterans General Hospital during the study period were retried for study. The Veteran Affairs Comprehensive Geriatric Assessment included demographic characteristics, Clinical Frailty Scale, Braden Scale, St. Thomas's Risk Assessment Tool in Falling Elderly Inpatients Scale, Cumulative Illness Rating Scale for Geriatrics, Barthel Index, Instrumental Activities of Daily Living, Mini-Mental State Examination, Geriatric Depression Scale-5 and Mini-Nutritional Assessment – Short Form, as well as common geriatric syndromes. All patients were categorized into the home care group and institutional care group based on their living arrangement before GEMU admissions. Six-month mortality and decline in Barthel Index were defined as adverse clinical outcomes. Results Overall, data of 395 patients were used for analysis. The baseline comparisons showed that the institutional care group was more likely to be unmarried, have lower education, lower risk of falls and less polypharmacy, but more likely to experience functional decline at follow up than the home care group. Multivariate logistic regression showed that male (OR 3.59, 95% CI 1.04–12.38, P = 0.043) and higher Cumulative Illness Rating Scale for Geriatrics score (OR 4.08, 95% CI 1.49–11.19, P = 0.006) were associated with mortality, whereas the institutional care group (OR 0.30, 95% 0.09–0.99, P = 0.048) and lower Braden Scale (OR 0.80, 95% CI 0.67–0.94, P = 0.008) were protective against mortality. However, the institutional care group was independently associated with functional decline during the follow-up period (OR 2.19, 95% CI 1.12–4.29, P = 0.022). Conclusions Institutional care was associated with lower 6-month mortality risk for patients with dementia or cognitive impairment after discharge from the GEMU, but this group was more likely to experience functional decline. Further prospective study is required to clarify the clinical impact of living arrangements on long-term outcomes when people with dementia or cognitive impairment are admitted to acute hospitals. Geriatr Gerontol Int 2017: 17 (Suppl. 1): 44–49.