Institutional disclosure of adverse events at the Department of Veterans Affairs Healthcare System

Introduction: The importance of disclosing medical errors to patients has attracted considerable interest in recent years. Studies have shown that physicians and patients both believe that adverse events should be disclosed to patients. However, there is a gap between ideal and actual disclosure practices. Little is known about the rates of disclosure and the kind of adverse events leading to the disclosure. Methods: The Department of Veterans Affairs (VA) Healthcare System recognizes three types of adverse event disclosure: clinical disclosure, institutional disclosure, and large scale disclosure. Institutional disclosure is required when adverse events result in serious injury or death, or reasonably expected serious injury, or potential legal liability. It is the policy of the VA that disclosure of adverse events that cause harm to patients be a routine practice. As part of the quality assurance program, we collected and analyzed fiscal year 2010 institutional disclosure data from all 138 VA facilities. Results: A total of 330 adverse events were disclosed by 81 facilities, representing a rate of approximately 5.5 adverse events disclosed per 100,000 veteran patients treated per year. Delayed or missed diagnosis, procedure or surgical complications without infections, medication ordering or dispensing errors, patient falls, and misinterpretation of laboratory test results or radiographs constituted 64.5% of all disclosed adverse events. Approximately 63% of adverse events were disclosed within 90 days of occurrence; approximately 43% of the adverse events occurred while patients were hospitalized; and 27% of the affected patients died of the adverse events. Five specialties, i.e., general medicine, general surgery, primary care, radiology, and nursing, were responsible for more than half of the disclosed adverse events. Conclusions: The results provide a better understanding of the kind of adverse events leading to institutional disclosure in the VA patient population. As there are no comparable non-VA data available in the literature, whether similar results will be found in non-VA patient populations requires further investigation.