Abstract
INTRODUCTION: Adverse events (AEs) involve a safety problem and compromise the quality of care, but evidence on how they affect the prognosis of patients is limited. OBJECTIVE: In this study, the relationship between the presence of AEs and in-hospital mortality among patients is analysed, and their characteristics are compared. MATERIAL AND METHODS: An observational study with a cross-sectional design was conducted in 32 hospitals of varying complexity in the Community of Madrid. The clinical history of 9,111 patients was analysed. Patients who were in the emergency room and those admitted to psychiatric units or centres were excluded. All hospitalisations were reviewed using the Harvard Medical Practice Study methodology for the detection and characterisation of AEs. The association between in-hospital mortality and the number of AEs was analysed with two multivariate models via logistic regression: 1) an explanatory model adjusted for confounders and 2) a predictive model of in-hospital mortality. A descriptive analysis of the differential characteristics of the AE was performed for the patients who died. RESULTS: In-hospital mortality was 5%, with a higher incidence of AEs in patients who died (29.8% versus 11.9%; p < 0.005). The presence of 1 AE (OR [95% CI]: 2.1 [1.6 to 2.7]) or ≥3 AEs (2.4 [1.1 to 5.1]) significantly increased the odds of mortality. In addition, the increase in the number of AEs was a predictor of mortality without a dose-response effect. The AEs that were most associated with in-hospital mortality were those related to care (15.5%; p = 0.047), and 15.3% of the AEs that occurred during ward care contributed to in-hospital mortality. CONCLUSION: There is an association between AEs and in-hospital mortality. The presence of at least 1 AE implies a critical event in the patient's prognosis without a dose-response effect. Reducing AEs related to care in patients with comorbidities is positioned as an efficient strategy for improving health outcomes.