Eva Montané ^1,2* Yolanda Sanz ¹ Soraya Martin ¹ Clara Pérez-Mañá ^1,2 Esther Papaseit Fontanet ^1,2 Olga Hladun ¹ Georgina de la Rosa ¹ Magi Farré ^1,2

• ¹ Hospital Germans Trias i Pujol, Badalona, Spain
• ² Autonomous University of Barcelona, Barcelona, Catalonia, Spain

Objectives: We aimed to assess the characteristics of adverse drug reactions (ADRs) collected in a university hospital.Methods: A retrospective analysis of ADRs reported in the Hospital Pharmacovigilance Program database (RutiRAM) over a 13-year period was conducted. The analysis included a description of ADRs (System Organ Class [SOC]) and their seriousness, the drugs involved (level 1 of the Anatomical Therapeutic Chemical [ATC] Classification System), drug-drug interactions, medication errors, drugs 'under additional monitoring', positive rechallenge, and the 'pharmacovigilance interest' of ADRs. An ADR was considered of 'pharmacovigilance interest' when it was serious, and/or produced sequelae, and/or affected the paediatric population, and/or when the suspected drug was 'under additional monitoring'. Additionally, an exploratory analysis for bivariate associations through an automated method were was performed.Results: A total of 2,148 spontaneous ADRs were recorded registered in the RutiRAM database, with 92.5% recorded by medical doctors. The mean age of cases was 59.2 years (SD 20.9), range 1 day -99 years; 5.7% were pediatricpaediatric, 46.2% adults, and 48.1% elderly. The drugs most often involved were anti-infectives (ATC group J), mainly amoxicillin-clavulanic acid. 'Blood system disorders' were the most frequent SOC ADRs (classified by SOC), and skin rashes were the most frequent ADRs. The 63.2% of ADRs were considered of 'pharmacovigilance interest'. Almost half of ADRs were hospitalacquired, and these were related to medication error; serious ADRs were related to drug-drug interactions and elderly patients, and involved drugs 'under additional monitoring' were related to younger ones. This is the first study to overview of ADRs reported in an HPVP over more than a decade. Almost twothirds of the ADRs collected in the RutiRAM database are of sufficient quality to be classified as 'pharmacovigilance interest', and thus can contribute to signal detection and the issuing of drug alerts by pharmacovigilance systems. Analysing ADRs in hospitals contributes to patient safety by implementing relevant actions to prevent medication errors or ADRs, some of which can be applied to other centres.ADRs occurring in the hospital setting play a valuable role in monitoring internal patterns, which can be addressed through hospital policies, and for national pharmacovigilance systems in generating drug safety signals.