STUDY: PHYSICIAN PERFORMANCE CAN BE TIERED USING MEASURE COMPOSITES
Research published in Medical Care shows reliable means for differentiating quality of care

WASHINGTON— The performance of physician practices can be reliably measured by using a composite of quality measures, according to research led by Sherrie Kaplan, PhD, MPH of the University of California at Irvine and co-authored by National Committee for Quality Assurance Executive Vice President Greg Pawlson, M.D., M.P.H.

Findings published in the April 2009 issue of Medical Care used data from NCQA-Recognized physicians suggests that physician practices can be distributed into three levels of relative quality (high, average, low) when a cluster of measures, rather than a single measure, is used to differentiate performance.

Researchers used data submitted by physicians who were recognized in the NCQA Diabetes Physician Recognition Program (DPRP). Using a sample of 35 patients from each practice, the research team analyzed data from 11 diabetes measures. They found that combining performance on five to nine measures could reliably separate practices into at three levels of quality.

“This research suggests that evaluating physician practices for pay-for-performance or similar programs can be a sound process when based on selected measure composites,” said Pawlson. “As we move further into the widespread use of evaluation at the physician practice level, careful testing and evaluation of the evaluation methodology is critical for ensuring that the information is accurate and draws fair conclusions. Composite measures appear to offer a substantial advantage over individual or unrelated measures in this effort.” 

Evaluation programs need to use a carefully selected set of measures in order to be meaningful. "Our study underscores the importance of setting thresholds for the measures that reflect the degree of physician influence or impact on the measures,” said Kaplan. “Identifying thresholds that reflect physician impact in contrast to the influence of the other factors, like patient characteristics, is of critical importance for fair and reliable performance assessment."

Kaplan and Pawlson were joined by John L. Griffith, Ph.D. and Lori L. Price, M.S., of the Institute for Clinical Research and Health Policy Studies at Tufts-New England Medical Center; and Sheldon Greenfield, M.D, of the University of California at Irvine School of Medicine. “Improving the Reliability of Physician Performance Assessment” appears in the April 2009 issue of Medical Care.

NCQA is a private, non-profit organization dedicated to improving health care quality. NCQA accredits and certifies a wide range of health care organizations and recognizes physicians in key clinical areas. NCQA’s Healthcare Effectiveness Data and Information Set (HEDIS®) is the most widely used performance measurement tool in health care. NCQA is committed to providing health care quality information through the Web, media and data licensing agreements in order to help consumers, employers and others make more informed health care choices.

# # #