[Outcome and predictors of early geriatric rehabilitation in an acute care setting].

Background: Geriatric assessment and treatment including an interdisciplinary team is a well accepted method to improve functionality and mortality in the elderly. Nevertheless, only little evidence and limited data are available concerning the outcome of this approach in an acute care setting and early rehabilitation.
Patients and Methods: All patients included in a comprehensive geriatric program within 1 year were eligible for inclusion in a retrospective chart analysis. Admission to the program was allowed after counseling by an experienced geriatrician applying pre-defined criteria including significant deconditioning, preserved rehabilitation prognosis and/or ongoing instable medical condition requiring an acute care setting. Patients who refused therapy or died within the first week of treatment were excluded. A telephone interview was performed 6 months after patient discharge and data concerning mortality, functional course and quality of life (activities of daily life ADL) were retrieved. All patients were treated in an acute care setting in two geriatric and general internal medicine wards.
Results: A total of 138 patients were included in the program. In 128 cases data concerning the 6 month mortality could be retrieved and 92 patients agreed to participate in a complete telephone interview. Mortality within the early rehabilitation course was 4%, the 6 month mortality however was 28%. Functionality improved during early rehabilitation and a median increment of 10 points in the Barthel index was found (range 0-100). Analysis of individual increments showed high interindividual variability throughout the whole range of the ADL score at the beginning of the treatment. Concerning the 6 month course after hospital discharge functionality remained at a steady median value but showed a large range of both increasing and decreasing ADL values. Analysis of predictors by logistic regression revealed the ADL score on admission to the early rehabilitation program as the predominant predictor overriding other factors such as depression or cognitive impairment. An ADL value ≤15 on admission was a strong predictor for an unfavorable outcome in early rehabilitation defined as death or an ADL increment ≤10. An ADL value ≤15 on admission was also a strong predictor for overall mortality within 6 months. On the other hand an ADL value ≤15 on admission was a strong predictor for a steady or increasing functionality among the survivors of the 6 month postdischarge interval.
Conclusions: The mortality rate in the early rehabilitation course in this preliminary and observational study was relatively low and may be explained by both the exclusion of patients not surviving the first week of treatment and an effective inclusion algorithm. The high overall 6 month mortality rate may also be explained by the selection of patients focusing on those severely functionally impaired. This subpopulation of elderly inpatients in acute care units reveals a high vulnerability to critical incidents. Besides scale effects there is also a remarkable variability in the 6 month postdischarge course pointing to heterogeneity and different patterns concerning the time course of functionality.