The objective of the review was to determine whether publicly releasing performance data changes the behaviour of healthcare consumers, professionals, providers and purchasers in a way that improves performance and quality of care. Although it is increasingly common to release information about performance in health care, its effectiveness in changing behaviour has not been determined. We searched the scientific literature for studies comparing the public release of performance data using different public release approaches. These approaches varied in type, content and presentation formats. The aim of the studies was to assess how and to what extent data release led to changes in care. We examined the effects of interventions by targeting the behaviour of healthcare consumers, professionals, providers and purchasers. We also examined the improvement effected by changes in care.
We found four relevant studies containing more than 35,000 consumers, and 1560 hospitals. Two were in health plan settings and the other two were in hospital settings. One cluster-randomised controlled trial and one cluster quasi-randomised trial reported no evidence that information from the Consumer Assessment of Healthcare Providers and System influenced the health plan choice in a Medicaid population. The interrupted time series study found a small positive effect of publishing data on patient volumes for coronary artery bypass grafting and low-complication outliers for lumbar discectomy, but these effects did not persist longer than two months after each public release. No effects on patient volumes for acute myocardial infarction were found. One cluster-randomised controlled trial studied performance and quality of care after the public release of performance data for patients with acute myocardial infarction and congestive heart failure. No effects were found for the composite process-of-care indicators for either condition, but there were some improvements in the individual indicators for acute myocardial infarction and congestive heart failure and in the acute myocardial infarction mortality rates. Further, more quality improvement activities were initiated as a result. No secondary outcomes (awareness, attitude, views and knowledge of performance data and costs) were reported. On the basis of this rather sparse evidence of low quality, we cannot draw any firm conclusions about the effects of the public release of performance data on behaviour or on improvements in the performance and quality of health care.