Trends in Function and Postdischarge Mortality in a Medicine for the Elderly Rehabilitation Center Over a 10-Year Period

Witham MD, Ramage L, Burns SL, Gillespie ND, Hanslip J, Laidlaw S, Leslie CA, McMurdo ME. Trends in function and postdischarge mortality in a medicine for the elderly rehabilitation center over a 10-year period.ObjectivesTo ascertain trends in function and mortality after admission to a medicine for the elderly rehabilitation unit, and to analyze factors associated with these outcomes.DesignRetrospective cohort analysis of routinely collected clinical data during the period from January 1, 1999, to December 31, 2008.SettingHospital-based medicine for the elderly rehabilitation unit.ParticipantsPatients (N=4449) admitted for rehabilitation after medical and surgical illness, stroke, and fractured neck of the femur.InterventionsNot applicable.Main Outcome MeasuresAnalysis of routinely collected clinical data: admission and discharge Barthel scores; indices of cognitive impairment, mental illness, swallowing and feeding difficulties. Discharge diagnoses, place of discharge, date of death, and discharge medications were analyzed, along with length of stay. Regression analysis of factors associated with improvement in Barthel score, place of discharge, and postdischarge mortality.ResultsLength of stay and admission Barthel scores were unchanged over the study period, but discharge Barthel scores improved from 13.5 (maximum score, 20) in 2002 to 14.8 in 2008 (P=.002 for trend). Discharge to home increased from 290 (61%) of 472 patients in 2001 to 290 (76%) of 382 patients in 2007 (P<.001 for trend). Age, admission Barthel score, cognitive impairment, problems with understanding, and problems with expression were independent predictors of the change in Barthel score between admission and discharge. The adjusted hazard ratio for postdischarge mortality in 2007 to 2008 compared with 1999 to 2000 was .76 (95% confidence interval, .63–.93).ConclusionsFunctional and mortality outcomes improved over a 10-year period in this rehabilitation unit, despite similar Barthel scores on admission and equivalent lengths of stay.