The Effects of Physician Practice Size and Composition on Health Care Use and Spending

Background:   The organization of physician practice has changed dramatically in recent years, with physicians increasingly practicing in both single and multispecialty groups.  Group practice, however, can have both positive and negative effects on consumers.  Large, multispecialty groups may benefit from economies of scale and scope that allow them to provide more coordinated, lower cost and higher quality care.  Larger groups, however, may also generate market power, resulting in higher prices for health care services.  While a growing body of evidence has documented that greater concentration in physician markets leads to higher prices, there is less evidence on the effects of physician group composition on the delivery of health care services.  In this paper, we examine the effects of physician group composition on health care use and spending.         

Data and Methods.  We examine how the organization of the practice of a patient’s primary care physician affects health care use by examining patients who change their primary care physician due to a change in their residence. Our analysis is based on 2004 to 2012 Medicare claims.  Following methods used to attribute Medicare patients to ACOs, we identify a patient’s primary care physician as the physician who provided the most primary care services to a patient in a given year. Using methods we have developed in our prior work, we identify the characteristics of the physician’s practice, including the number and specialties of the physicians in the group, based on the tax-identification number (TIN) on the claims the physician submits to Medicare.  We identify patients who moved based on the patient’s zip code of residence in the Medicare denominator file and then characterize patients who moved based on the practice type of their primary care physician in each years and compare compare changes in health care spending and use across the groups.  The models control for age, sex, race and year.  We will also control for market concentration in pre- and post-move region and instrument for practice type using the prevalence of practice type by geographic location.

Results:  We find the patients who change to a primary care physician who works in a large, multispecialty group practice, from either a solo practitioner or a primary care physician who works in primary care group practice, have significantly higher Medicare Part B expenditures after than before the change, relative to those who changed practices but did not change practice type.  Preliminary estimates indicate that, for patients leaving group primary care for large multispecialty practices, expenditures were almost 20% higher as compared to those who continued to receive care in group primary care practices.  The increase was even higher for those moving from solo practices to large multispecialty practices.

Conclusions.  The composition of a primary care physician’s practice affects the type of care patients receive with patients in multispecialty group practice receiving substantively more services than those seeing primary care physicians who are not part of multispecialty practices.