We aimed at investigating the relationship between socioeconomic (SES) and health status in the context of an observational multicenter study of elderly hospitalized patients. Our series consisted of 473 patients aged 70 years or more. K-means cluster analysis was used to generate 3 clusters on the basis of age, gender, education, perception of personal economic situation, difficulty to reach health services, need for formal or informal support, family arrangement, and population density of residence municipality. Logistic regression analysis was used to identify correlates of "negative" SES. Correlates of "negative" SES cluster were older age (odds ratio=OR=5.19, 95% Cl=2.28-11.8), cognitive impairment (OR=6.36, 95%CI=3.11-13.0), emergency hospital admission (OR=3.11; 95%CI=1.52-6.35), and dependency in at least 1 BADL (OR=4.36, 95%CI=1.53-12.4). In conclusion, "negative" SES is associated with age and selected indices of frailty in elderly hospitalized patients. The evaluation of socio-economic problems should be routinely addressed in elderly hospitalized patients in order to tailor appropriately post-discharge use of health care resources.