BACKGROUND: Dementia is often considered a predictor of adverse hospitalization outcomes. However, the relative contributions of dementia and other risk factors remain unclear. OBJECTIVE: To assess, in a prospective study, the relative value of dementia for predicting hospitalization outcomes, taking into account comorbidity, functional and nutritional status in 435 inpatients (age 85.3 +/- 6.7; 207 cognitively normal, 48 with mild cognitive impairment and 180 demented) from the acute and rehabilitation geriatric hospital of Geneva. Hospitalization outcomes (death in hospital, length of stay, institutionalisation and formal home care needs) were predicted using logistic regression models with sociodemographic characteristics, cognitive status, comorbid Charlson index-CCI, functional and nutritional status as independent variables. RESULTS: Moderate and severe dementia and poor physical function strongly predicted longer hospital stay, institutionalization and greater home care needs in univariate analyses. CCI was the best single predictor, with a four-fold difference in mortality rates between the highest and lowest scores. In multivariate analysis, the best predictor of institutionalisation was dementia, whereas the best predictor of death in hospital or longer hospital stay was higher comorbidity score, regardless of cognitive status. Functional status was the best predictor of greater home care needs. CONCLUSIONS: Dementia in very old medically ill inpatients was predictive only of discharge to a nursing home. Higher levels of comorbidity and poor functional status were more predictive than dementia for the other three hospitalization outcomes. Thus, comorbid medical conditions, functional and nutritional status should be considered, together with cognitive assessment, when predicting hospitalization outcome.
BACKGROUND:Dementia is often considered a predictor of adverse hospitalization outcomes. However, the relative contributions of dementia and other risk factors remain unclear. OBJECTIVE: To assess, in a prospective study, the relative value of dementia for predicting hospitalization outcomes, taking into account comorbidity, functional and nutritional status in 435 inpatients (age 85.3 +/- 6.7; 207 cognitively normal, 48 with mild cognitive impairment and 180 demented) from the acute and rehabilitation geriatric hospital of Geneva. Hospitalization outcomes (death in hospital, length of stay, institutionalisation and formal home care needs) were predicted using logistic regression models with sociodemographic characteristics, cognitive status, comorbid Charlson index-CCI, functional and nutritional status as independent variables. RESULTS: Moderate and severe dementia and poor physical function strongly predicted longer hospital stay, institutionalization and greater home care needs in univariate analyses. CCI was the best single predictor, with a four-fold difference in mortality rates between the highest and lowest scores. In multivariate analysis, the best predictor of institutionalisation was dementia, whereas the best predictor of death in hospital or longer hospital stay was higher comorbidity score, regardless of cognitive status. Functional status was the best predictor of greater home care needs. CONCLUSIONS:Dementia in very old medically ill inpatients was predictive only of discharge to a nursing home. Higher levels of comorbidity and poor functional status were more predictive than dementia for the other three hospitalization outcomes. Thus, comorbid medical conditions, functional and nutritional status should be considered, together with cognitive assessment, when predicting hospitalization outcome.
Authors: Mark B Snowden; Lesley E Steinman; Lucinda L Bryant; Monique M Cherrier; Kurt J Greenlund; Katherine H Leith; Cari Levy; Rebecca G Logsdon; Catherine Copeland; Mia Vogel; Lynda A Anderson; David C Atkins; Janice F Bell; Annette L Fitzpatrick Journal: Int J Geriatr Psychiatry Date: 2017-02-01 Impact factor: 3.485
Authors: D Zekry; K H Krause; I Irminger-Finger; C E Graf; C Genet; A-M Vitale; J-P Michel; G Gold; F R Herrmann Journal: J Nutr Health Aging Date: 2012-03 Impact factor: 4.075
Authors: Daniel H J Davis; Sam T Creavin; Jennifer L Y Yip; Anna H Noel-Storr; Carol Brayne; Sarah Cullum Journal: Cochrane Database Syst Rev Date: 2015-10-29