Objectives To determine the incidence of drug-related deaths (DRD) in a university hospital in 2015, to describe their characteristics, and to discover risk factors of DRD. Methods An analytic and retrospective cohort study. Patients with a death diagnose predefined from a list of medical conditions potentially caused by drugs were the selected cases for further review. Causality assessment was evaluated by a local drug safety committee. Results Out of 1,135 inpatients deaths, 73 DRD were included (6 were hospital-acquired). The incidence of DRD of all hospital admissions was 0.34%, and the incidence of all deaths cases was 7%. Drugs were the cause of death in 38 patients (52%) and a contributive role in 35 (48%). The median age of DRD patients was 72 years (range 19 – 94) and 72.6% were men. The median hospital stay, Charlson score, and number of drugs were 5 days, 2 points, and 7 drugs respectively. The most frequent DRD were cerebral haemorrhages and infections in drug-immunosuppressed patients (32, 43.8%, each group). The most frequently involved drugs were antineoplastics and glucocorticosteroids (40% and 18%), and antithrombotics (33%); drug-drug interactions were present in 44% DRD. Sex, age, and number of drugs were risk factors of DRD. Conclusions Adverse drug reactions were a significant cause of death in hospitalized patients, mainly haemorrhages and infections precipitated by drug-drug interactions. Risk factors for DRD were sex, age and number of drugs. Preventable DRD and measures to avoid them should be accurately assessed in further studies.