Inappropriate prescribing is particularly common in older patients and is associated with adverse drug events (ADEs), hospitalization, and wasteful utilization of resources. We randomized 400 hospitalized patients aged >= 65 years to receive either the usual pharmaceutical care (control) or screening with STOP/START criteria followed up with recommendations to their attending physicians (intervention). The Medication Appropriateness Index (MAI) and Assessment of Underutilization (AOU) index were used to assess prescribing appropriateness, both at the time of discharge and for 6 months after discharge. Unnecessary polypharmacy, the use of drugs at incorrect doses, and potential drug-drug and drug-disease interactions were significantly lower in the intervention group at discharge (absolute risk reduction 35.7%, number needed to screen to yield improvement in MAI = 2.8 (95% confidence interval 2.2-3.8)). Underutilization of clinically indicated medications was also reduced (absolute risk reduction 21.2%, number needed to screen to yield reduction in AOU = 4.7 (95% confidence interval 3.4-7.5)). Significant improvements in prescribing appropriateness were sustained for 6 months after discharge.