More on reporting medical errors.

To the Editor: Crone et al. (1 ) present a case study with much of their discussion about aspects of reporting suspected medical errors, particularly when there is a possibility of misconduct or violation of rules. Marx (2 ), who described a useful taxonomy of medical errors, recommends that no blame be given except for certain situations, one of which can be summarized as “failure to report safety incidents or risk exposures”. In the clinical laboratory case discussion of Crone et al. (1 ), the fact that the risk management team decided later that no error had occurred would not excuse a person from reporting an event thought to be an error. As described in the case report, it was up to the (risk management) team to classify the severity of reported events, including the classification that no error occurred. In discussing error reduction, Cone et al. focus on root cause analysis of sentinel events, omitting discussion of FMEA (Failure Mode Effects Analysis), which is designed to decrease the risk of potential errors. A successful FMEA will lead to fewer (and possibly no) sentinel events. FRACAS (Failure Review and Corrective Action System) is a tool to decrease the recurrence of observed events. In contrast to other industries, hospitals typically subject only sentinel events to root cause analysis. In a FRACAS, all errors are counted, and corrective actions are taken on those with the highest severity and frequency of occurrence. The process has a quantitative goal of reducing the combined error rate to an acceptable number, and the FRACAS process measures the combined error rate on a regular basis.