Abstract
The aim of this retrospective study was to evaluate risk factors for 3-years mortality after hospital discharge in all inpatients admitted to a general hospital in Milano, Italy. A total of 2580 consecutive patients admitted to Ospedale San Paolo, July 1 to December 31, 2012, for several classes of diseases (internal medicine, cancer, infectious diseases, trauma and surgery, pneumonia, and heart diseases) were studied. Age, total disease, type of admission, length of admission, age-adjusted Charlson index, prognostic nutritional index (PNI), and full blood count were evaluated. Univariate Cox models were used to evaluate the association between variables and death. Of the 2580 consecutive patients (age 66.8 ± 19.36 years, mean ± SD), 920 died within 3 years after discharge. At univariate analysis, all investigated variables, except sex and lymphocytes, were associated with patient death. Stepwise regression analyses revealed that the age-adjusted Charlson index or age plus total diseases, type of admission, number of admissions, and PNI were significant risk factors in the whole sample and in some classes of disease. Results were superimposable when considering death from date of admission instead of date of discharge, meaning that in-hospital death was not relevant to the total death count (115 out of 902). Seriousness of baseline conditions represents the major risk factor for mortality in most classes of disease, and possibly influences other predictors, such as type of admission and length of stay. This suggests that the current model of hospital admission might be improved, for instance, through comprehensive care at home, instead of hospital admission, or before admission.