Performance measurement in congenital heart surgery: benefits and drawbacks.

Performance monitoring is playing an increasing role in modern congenital heart surgery. We present a number of examples to show the kinds of benefits that it can lead to. Conversely, we discuss some of the potential pitfalls including: errors in the collection of data; use of inappropriate data originally collected for a different purpose; assessment based on inadequate sample size; lack of confidence intervals in reported results; failure to include all members of the team in the assessment; and others. We emphasize that apparent changes in performance from year to year or between different units can be purely due to chance. However, despite these difficulties, we believe that performance monitoring, if carried out carefully, with clearly specified goals, can have many benefits in improving patient care. We suggest that statistical analysis of results should not be taken as a definitive indication of inadequate performance; confidence intervals for any performance measurements should always be clearly indicated; small groups of patients/operations should not be used as a basis for assessment; and that performance measurements should be assigned to the whole team, rather than individual members.