Background: Potentially inappropriate prescriptions (PIPs), associated with adverse drug reactions, hospitalization, and wasteful healthcare spending, are common in elderly patients with comorbidities and multiple drugs. Objective: Our objective was to develop and validate a new tool to reduce PIPs in a hospitalized geriatric population. Methods: This was an observational cohort study of two cohorts (development [n = 100 subjects] and validation [n = 449 subjects]) of consecutive patients aged ≥65 years admitted to geriatric wards from April to December 2012. In the development phase, data on clinical and demographic characteristics, Multidimensional Prognostic Index (MPI), and prescribed drugs before hospital admission were collected and processed using a tool that integrates the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions (STOPP) criteria and the Micromedex™ Drug-Reax System, a drug-management platform. This tool generated a report that was provided to the treating physicians. The weight of the PIPs, as defined by the Medication Appropriateness Index (MAI), was assessed on admission and on discharge. Similar procedures were followed in the validation cohort. Results: PIPs were independently associated with polypharmacy and with prescribing of antithrombotics, sedatives and antidepressants in both cohorts. The use of the tool led to a significant reduction of the MAI score, both in the development (median score 4 [interquartile range; IQR 1–5] vs. 2 [IQR 0–4], p < 0.001) and in the validation cohorts (median 3 [IQR 1–5] vs. 2 [IQR 0–4], p < 0.001). Conclusion: This tool significantly reduced PIPs, as defined by the MAI score, in a hospitalized geriatric population. This strategy might be useful to minimize inappropriate medication exposure in this group.