Audit and feedback is a commonly used but variably effective technique designed to encourage health care providers to improve their practice. We conducted a systematic review of randomized trials of audit and feedback interventions to understand the kinds of feedback most commonly provided and whether the design of these interventions are consistent with theoretical predictions about how to maximize feedback effectiveness.
Method:
The 140 studies in the 2012 Cochrane update on audit and feedback interventions were utilized as our dataset. Thirty intervention variables were extracted, including: to whom the feedback was given, what specifically was given (content, group versus individual, individual versus aggregate, comparison, form), when (lag time), how (face-to-face, self-monitoring), how much (total and frequency), and why audit and feedback was chosen as the strategy (empirical, intuitive, or theoretical). Two extractors reviewed each study; disagreements were resolved by consensus.
Result:
The most common form of audit and feedback is a report of aggregate patient data (81%), sent to an individual (51%), about provider behaviour (79%), with a comparison to the performance of others within the group (50%). Key characteristics were often poorly reported: the number of times delivered and the nature of the comparison provided were unclear 25% and 37% of the time respectively. Some of the features were consistent with theoretical hypotheses - about own cases (57%), given to the target for change (92%), but audit and feedback intervention design was frequently suboptimal - delivered only once 33% of the time, face-to-face feedback <50% of the time, and self-monitoring in only 14% of the interventions. The reported rationale for using audit and feedback was rarely based on a theory of feedback or related constructs (9%); in 26% of trials, investigators reported no rationale for audit and feedback at all.
Conclusion:
To date, audit and feedback interventions are used without theoretical rationale and frequently designed in ways that are not in keeping with our understanding of how to make feedback most effective. While the optimal approach to using audit and feedback as a strategy to support providers in achieving quality care requires further theoretical and empirical investigation, intervention designers should improve the clarity of reporting, and evaluate interventions supported by theoretical hypotheses about audit and feedback effectiveness.