Integrated Medical Groups and Higher-Quality Medical Care

  1. Ateev Mehrotra, MD, MPH;
  2. Arnold M. Epstein, MD, MA; and
  3. Meredith B. Rosenthal, PhD
  1. From the University of Pittsburgh and RAND Pittsburgh, Pittsburgh, PA 15213, and Harvard School of Public Health, Boston, MA 02115.

    IN RESPONSE:

    We agree with Dr. Kumar that using claims data to measure quality has limitations (1), although claims data might be more accurate than medical charts for some quality measures we studied, such as diabetic eye screening or mammography, which are conducted outside the primary care office and are tied to reimbursement (2). Our goal was to compare the quality in different types of physician groups. We know of no evidence that any biases in claims data would vary systematically by type of physician group, and the performance data were audited to evaluate completeness and accuracy. In speculating about potential explanations for the quality differences we found, we conjectured that a potential reason was a systematic difference in the physicians who choose to work in one type of group versus another. This idea was based in part on previous research that documented differences in physicians who choose to work as employees versus independently (3). For example, younger physicians are more likely to work as employed physicians and generally have higher quality scores (4). We should emphasize that the relationship we observed does not belie the fact that there are high-quality physicians in independent practice associations and low-quality physicians in integrated medical groups. Finally, as noted in our paper and by Dr. Kumar, a limitation of our study is that the electronic medical records reported by groups were probably rudimentary compared with current standards.

    Ateev Mehrotra, MD, MPH

    University of Pittsburgh and RAND Pittsburgh

    Pittsburgh, PA 15213

    Arnold M. Epstein, MD, MA

    Meredith B. Rosenthal, PhD

    Harvard School of Public Health

    Boston, MA 02115

    Article and Author Information

    • Potential Financial Conflicts of Interest: None disclosed.

    References

    1. 1.
    2. 2.
    3. 3.
    4. 4.
    « Previous | Next Article »Table of Contents