The gold standard method to assess hospital adverse events (AEs) is retrospective chart review. A cheaper alternative is incident reports (IRs) routinely used by hospitals to note adverse outcomes or accidents experienced by patients. Yet little is known about the value of IRs to detect AEs.
To assess the proportion of AEs measured through chart reviews that are detected by IRs and to describe the types of AEs documented by IRs.
A random sample of 2213 charts of adult patients hospitalized in 19 short-stay hospitals in the province of Quebec, Canada, were assessed. An AE was defined as an unintended injury or complication caused by clinical care. Research nurses assessed selected charts for screening criteria known to be associated with an AE. Physicians reviewed the positively screened charts to identify AEs.
Incident reports were present in only 15.5% of charts with AEs, and in 4.4% of charts without AEs. The presence of an IR in a chart varied, depending on the type of procedures or events associated with the AE and the consequence of the AE.
In their current form, IRs are not that useful to detect hospital AEs as defined in this study. For better quality assurance, the information collected using IRs should be expanded to cover a wider range of AEs. Given their routine use in hospitals and their relatively low cost, efforts should be made to make better use of IRs to improve patient care.