Incidence of adverse events in an integrated US healthcare system: A retrospective observational study of 82,784 surgical hospitalizations

Background: Many health care facilities have developed electronic reporting systems for identifying and reporting adverse events (AEs), so that measures can be taken to improve patient safety. Although several studies have examined AEs in surgical settings, there has not previously been a systematic assessment of the variations in adverse event rates among different types of surgery, nor an identification of the particular types of AEs that are most common within each surgical category. Additionally, this study will identify the AE severity level associated with each of the AE category types.Methods: This retrospective observational study was conducted at three Midwestern hospitals that are part of a large integrated healthcare system. Data from 2006 through 2009 were analyzed to determine the rates of reported adverse events (per 1,000 hospitalizations involving a surgical procedure) for 96 categories of surgery as classified according to the ICD-9-CM procedural coding system. Univariate and bivariate summary statistics were compiled for AEs by type, severity, and patient age.Results: During the four-year study period, there was a total of 82,784 distinct hospitalizations involving at least one surgical procedure at these three hospitals. At least one adverse event was reported at 5,368 (6.5%) of those hospitalizations. The mean rate of AEs among all surgical procedure groups was 82.8 AEs per 1,000 hospitalizations. Adverse event rates varied widely among surgical categories with a high of 556.7 AEs per 1,000 hospitalizations for operations on the heart and pericardium. The most common type of adverse event involved care management, followed by medication events and events related to invasive procedures.Conclusions: Detecting variations in AEs among surgical categories can be useful for surgeons and for hospital quality assurance personnel. Documenting the specific AE incidence rates among the most common types of surgical categories, and determining AE severity and age distributions within surgical categories will enable officials to better identify specific patient safety needs and develop appropriately targeted interventions for improvement. © 2014 Zeeshan et al.; licensee BioMed Central Ltd.