Improving the quality of care is an unceasing goal of clinical practice, achieved with a variety of methods. Analyzing one such method, Drs. Inra, Nayor, and colleagues contributed a study, “Comparison of colonoscopy quality measure across various practice settings and the impact of performance scorecards” to this issue of Digestive Diseases and Sciences [1], in which they compare physician performance before and after the distribution of individual scorecards for colonoscopic withdrawal times, cecal intubation rates, and adenoma detection rates (ADR) for endoscopists practicing in a tertiary-care hospital, a community-based private hospital, and a private practice group. They reported that, on average, the endoscopists in each practice venue performed high-quality screening examinations at baseline with no statistically significant improvements in performance after the distribution and review of the scorecards. Although the results of this study were negative, the authors’ careful analysis of an important and common process is informative and begins to fill a gap in the literature regarding endoscopy quality measurement.

Feedback and scorecards for improvement may not be apparently effective when the baseline performance is already quite favorable, as it is reported here. One can imagine a skilled endoscopist looking at his or her report and concluding that they perform rather well and no change or improvement is needed [2]. In contrast, one might expect that if there had been a significant unfavorable outlier, the endoscopist in question would have been alerted and motivated for a modification in his or her practice, particularly when compared with their peers, as the authors also noted. This same study might have been repeated in a group of less experienced or lower-volume endoscopists with a very different outcome. A study to show meaningful performance improvement would likely need a significant number of outliers. As suggested by the authors, there is experience to suggest that the simple knowledge that performance metrics are being monitored is motivating. In our own tertiary-care hospital, we have been providing scorecards to physicians for 10 years, with individual results compared to the practice and national standards. The year we announced the addition of ADR to our report, the overall rates increased by almost 10%.

Sometimes the information in the scorecard is sufficient to promote change, especially if the goals have face validity and are achievable [3]. Reports commonly begin confidentially in order to transmit information in a non-threatening fashion, but may be more effective if delivered regularly by a trusted colleague and as part of an overarching quality improvement structure. When an outlier needs to be alerted of their performance, mentoring or instruction in a skill that may be lacking should accompany the information. This second level of feedback for performance improvement is being adopted in most training programs. Concurrent specific feedback and instruction using simulation or video recording seems particularly effective [4, 5]. Moreover, the medical profession needs to learn how to use these data in order to most effectively promote active learning throughout one’s career.

The selection of the metrics is a key factor [6]. The performance metrics selected by the authors were standard well-established criteria for screening colonoscopy quality [7]. Since they are well understood, it is likely that the participating endoscopists were not naive to the criteria or their general performance. A greater impact might have been observed if there had been “stretch” criteria showing greater variation among the participants and ideally identifying a superior performer, who may have a best practice that can be shared among his or her colleagues. As one considers alternative or more discriminating criteria, one must be assured that the metrics can be measured simply and accurately. They should be reliable, reflect meaningful quality, and be conducted using a sample size large enough to provide confidence. Reports and measurements provided too frequently with small unstable numbers may undermine confidence in the process.

Performance feedback and scorecards have been widely used to assess and promote organizational performance improvement in many other settings [8]. In addition to performance improvement, scorecards may be used as a management communication tool, a quality-monitoring device, or as an educational intervention. The magnitude of the improvement, if any, may be quite variable depending on the specific situation, but when systematically studied, is often more effective than pure educational interventions [9]. Performance scorecards are also being used for purposes other than individual performance improvement. In addition to fellows in training whose performance is monitored as part of assessing training effectiveness and skill, more senior physicians are monitored for regulatory purposes, such as credentialing and Ongoing Provider Performance Evaluation, as mandated by The Joint Commission.

Performance-based physician scorecards also enable a level of accountability for quality of care and will become more widely used for in the coming years for public reporting. Ideally physicians will know their own results before regulators and outside agencies and will have an opportunity to learn from them and improve. The GI Quality Improvement Consortium (GIQuIC), a non-profit collaboration of the American College of Gastroenterology and the American Society for Gastrointestinal Endoscopy, is a quality benchmarking registry for benchmarking data on efficiency, effectiveness, and reliability that is probably underutilized. Organizations who do not participate in GIQuIC need to collect and analyze their own performance data as best they can in advance of public reporting. Additionally, academic journals need to be open to publishing studies of data-driven performance improvement initiatives even if negative. Independently, since insurers and the public have an appetite for these kinds of data, they will collect their own for use in contracting, payment modifications, or provider selection.

Although the specific nature of the move toward value-based purchasing seems uncertain, it is clear the direction toward greater accountability to demonstrate quality services will continue, if not accelerate. In order to meet this challenge, endoscopy practices of all sizes and across all settings will need more discriminating performance data, often adjusted for patient factors such as severity of illness, and the skills to use these data effectively. New criteria and continuing measurement of practitioner frontline performance is essential, and additional academic studies, such as this one, are needed to understand how to drive improvement and learning effectively, thereby delivering the high-value services that our patients expect and deserve.