Grant: Project Grant

Adverse events (AEs) are injuries caused by medical management rather than by the underlying condition of the patient. AEs are estimated to occur in 2.9% to 16.6% of all acute care hospitalizations, and studies have suggested that 30% to 58% of all AEs are preventable. Preventable AEs are a leading cause of death, resulting in 44,000 to 98,000 deaths each year. AEs are also very costly, with total annual costs estimated to range between USD $17 and $29 billion. For these reasons, preventing AEs is a high priority in acute care hospitals worldwide. An important requirement to evaluate the success of preventive measures is to have access to accurate, timely and efficient methods for monitoring AE rates. Moreover, with the growing emphasis on benchmarking and public reporting of AE data, these methods must also allow for valid inter-institutional comparisons. However, at present, there are no such methods. With the advent of electronic health records (EHR), and the development of automated methods for encoding and classifying EHR data, there is an exciting opportunity to identify potentially better methods of AE detection and monitoring. Taking advantage of this new opportunity, the objective of this research project is to assess the accuracy and the generalizability of using automated methods for detecting three of the most common AEs affecting hospitalized patients: 1) hospital-acquired pneumonia, 2) ventilator-associated events and, 3) central line-associated bloodstream infection. It is expected that this research project will result in more accurate and efficient measures of AEs. These measures could be used to assess the incidence rates of AEs, evaluate the success of interventions aimed at improving patient safety, and track progresses in patient safety through time. Lastly, automation has a strong potential to standardize AE monitoring, both within and across institutions, which is an important prerequisite to valid public reporting of AE data and benchmarking.