GAO on CMS’ Physician Feedback Reports

If you believe, and we do, that physicians deliver a fair amount of unnecessary or inappropriate care, the question becomes how to change that behavior.  First treatment patterns have to be measured for an individual doctor and then potentially compared to the care patterns of other doctors, as well as to evidence-based best practices.  Then the individual physician needs to get feedback on his or her practice patterns, and eventually incentives need to be created to discourage inappropriate care.  The Medicare program has been working on several aspects of this notion for years, and one area that has been challenging for it is the design, creation and implementation of feedback reports.  A new Government Accounting Office report reviews the status of CMS’ efforts.   (GAO Report)

The most troubling of GAO’s findings is that very few physicians are meeting the criteria for receiving a feedback report; out of a relatively small sample of physicians, only 18% or 1645, got a report.  Most physicians either did not have enough beneficiaries attributed to them for the resource measures or the quality measures and another set did not meet the requirement of having at least 30 physicians in the same specialty and geographic area for a peer group.  In addition, CMS is having difficulty actually distributing the report and it appears that few doctors review the reports when they get them and there are few incentives for them to do so.  Finally, core methodological issues like beneficiary attribution, peer group selection and minimum sample size determination continue to bedevil the agency.

One reason why this initiative is very important is that it has direct ties to the physician quality reporting program and the value modifier for physician payments to be implemented around 2015.  Getting feedback is critical to changing behavior, particularly because it creates a sentinel effect:  doctors know someone is watching their practice patterns.  Making the feedback credible is necessary or the recipient will quickly disregard it.  One of the best improvements CMS could make is to expand the feedback reports to be multi-payer; get Medicaid data and commercial payer data integrated to have a much bigger sample size, to be able to compare practice patterns for the same physician across payers and to lessen the burden of reviewing the reports  on the physician.  Many commercial insurers already produce these reports and they could help CMS with its methodological and distribution issues.  A unified physician practice pattern report would go a long way toward helping eliminate, or at least not pay for, inappropriate medicine.