The aim of the study was to investigate the association between socio-demographic variables and "at high risk of inappropriateness" of hospital admissions. We used hospital admissions data of Local Health Unit (LHU) Rome H (year 2004). We investigated the relationship between socio-demographic variables (sex, age, job activity, marital status, nationality, place of residence, educational level) and a high risk of inappropriate hospital stay. We computed univariate and multivariate analysis using chi2 test and logistic regression model. Out of 32,233 hospital admissions, 4685 (14.5%) resulted at high risk of inappropriateness. The following variables were associated with high risk of inappropriateness: age (for patients aged 0-15 and 46-65 OR: 1.83 (95% C.I.: 1.57-2.13) and 1.56 (95% C.I.: 1.42-1.72) respectively); job activity (for employed OR: 1.98 (95% C.I.: 1.81-2.17), for students OR: 1.34 (95% C.I.: 1.16-155)); marital status (for unmarried OR: 1.37 (95% C.I.: 1.23-1.51)); place of residence (for patients belonging to LHU Rome H OR:1.09 (95% C.I.: 1.02-1.78)); nationality (for foreign nationals OR: 0.71 (95% C.I.: 0.58-0.87)); educational level (for high school degree and graduated people OR: 0.89 (95% C.I.: 0.81-0.98)). Our study demonstrates that socio-demographic variables are related to the high risk of inappropriate hospital admissions. We believe that these variables could be considered as potential factors to modulate the offer of health services.