Q Shen1, D Cordato, D K Y Chan, W T Hung, M Karr. 1. Department of Aged Care and Rehabilitation, Bankstown-Lidcombe Hospital, Bankstown, New South Wales, Australia. qing.shen@swsahs.nsw.gov.au
Abstract
OBJECTIVES: To examine 12-month outcomes and develop predictive models for outcomes in elderly stroke patients. METHODS: Prospective study of 186 consecutive acute stroke patients aged > or = 65 years admitted to a local hospital between March 2002 and March 2003. Outcome measurements included mortality, functional independence measure (FIM) score and nursing home placement. Two predictive models, using multiple logistic regression analysis, were developed to identify the factors associated with (i) mortality, and (ii) being alive and independent (defined as mean FIM score > or = 90) at 12 months. RESULTS: One hundred and seventy two (92%) patients were followed up at 12 months post-stroke. Mortality rate was 31%, and was significantly higher in nursing home vs non-nursing home origin patients (68% [15/22] vs 25% [38/150]). Nursing home placement for non-nursing home origin survivors was 28% (31/112). Age > or = 85 years was associated with higher mortality (odds ratio = 5.3, 95% confidence interval = 1.8-15, P < 0.01) and lower FIM for patients living at home pre-stroke. Predictive models showed that age, not living at home pre-stroke, pre-stroke FIM < 108, inability to walk on admission, dysphasia, visual field loss and haemorrhagic stroke were associated with worse outcome. CONCLUSIONS: Predictive models--by developing new strategies to improve outcomes through identifying treatable predictive factors--may be clinically useful in elderly stroke patients.
OBJECTIVES: To examine 12-month outcomes and develop predictive models for outcomes in elderly strokepatients. METHODS: Prospective study of 186 consecutive acute strokepatients aged > or = 65 years admitted to a local hospital between March 2002 and March 2003. Outcome measurements included mortality, functional independence measure (FIM) score and nursing home placement. Two predictive models, using multiple logistic regression analysis, were developed to identify the factors associated with (i) mortality, and (ii) being alive and independent (defined as mean FIM score > or = 90) at 12 months. RESULTS: One hundred and seventy two (92%) patients were followed up at 12 months post-stroke. Mortality rate was 31%, and was significantly higher in nursing home vs non-nursing home origin patients (68% [15/22] vs 25% [38/150]). Nursing home placement for non-nursing home origin survivors was 28% (31/112). Age > or = 85 years was associated with higher mortality (odds ratio = 5.3, 95% confidence interval = 1.8-15, P < 0.01) and lower FIM for patients living at home pre-stroke. Predictive models showed that age, not living at home pre-stroke, pre-stroke FIM < 108, inability to walk on admission, dysphasia, visual field loss and haemorrhagic stroke were associated with worse outcome. CONCLUSIONS: Predictive models--by developing new strategies to improve outcomes through identifying treatable predictive factors--may be clinically useful in elderly strokepatients.
Authors: John Di Capua; Sulaiman Somani; Nahyr Lugo-Fagundo; Jun S Kim; Kevin Phan; Nathan J Lee; Parth Kothari; John Shin; Samuel K Cho Journal: Global Spine J Date: 2017-07-20
Authors: Fabrizio Carinci; Lorenzo Roti; Paolo Francesconi; Rosa Gini; Fabrizio Tediosi; Tania Di Iorio; Simone Bartolacci; Eva Buiatti Journal: BMC Health Serv Res Date: 2007-06-27 Impact factor: 2.655