OBJECTIVE

To evaluate short- and long-term outcomes of elderly patients (>or=65 years) treated at an intermediate care unit (IMCU) and to identify outcome predictors.

DESIGN AND SETTING

Prospective observational study in the IMCU of a university teaching hospital.

PARTICIPANTS

We studied 412 patients over 8 months, classified into three groups: under 65years (control group, n=158), 65-80 (n=186), and >80 (n=68).

MEASUREMENTS

At admission: APACHE II, TISS-28 first day, Charlson Index, diagnosis, and prior Barthel Index.

OUTCOME MEASURES

in-hospital mortality, length of stay, discharge destination, and 2-year mortality and readmissions. Data analysis included multivariate logistic regression and receiver operating characteristics area under the curve (ROC AUC).

RESULTS

No statistically significant differences between groups were observed in hospital mortality (14.1%), discharge to a long-term facility (2.7%), or 2-year readmissions (1.2+/-2.1). However, hospital stay was longer in patients aged 65-80years (14 vs.10 days) and 2-year mortality was higher in those 65 or over (34% vs.10.6%). In the overall series in-hospital mortality was predicted by APACHE II, first-day TISS-28, and diagnosis (ROC AUC 0.81), and 2-year mortality by Charlson Index and age (ROC AUC 0.77). In the elderly patients 2-year mortality was predicted by Charlson and Barthel indices (ROC AUC 0.70).

CONCLUSIONS

Illness severity and therapeutic intervention at admission to IMCU were predictors of short-term mortality, whereas the strongest predictor of long-term mortality was comorbidity. Our results suggest that comprehensive assessment of elderly patients at admission to IMCUs may improve outcome prediction.