[Risk of repeated hospitalizations in elderly users of an academic health center].

The aim of this study was to stratify 305 elderly (> 65 years) treated at an academic health center (Botucatu, São Paulo, Brazil) according to risk of repeated hospitalization. Data collection used an instrument to evaluate risk of repeated hospital admissions, including eight health indicators: self-rated health, hospitalizations, doctor visits, diabetes, cardiovascular disease, gender, social support, and age. 56.4% of interviewees presented low probability of repeated hospitalization, as compared to 26.9% medium, 10.5% medium-high, and 6.2% high probability. Combining health indicators with the probability of repeated hospitalization, for elderly classified as medium, medium-high, and high risk compared to low, the relative risk was significant: medium or bad health (2.31); hospitalization (2.38); > 3 doctor visits (1.75); diabetes (2.10); cardiovascular disease (2.76); male gender (1.68); and > 75 years of age (1.62). The instrument was able to stratify elderly at risk of repeated hospitalization and is thus useful for reorganizing public health services.