On March 11, 2026, the Federal Trade Commission (FTC) published a report that provides a comprehensive analysis of how the available set of providers and physician practice structures evolved between 2015 and 2020 across 15 states and within different specialties. The report, “An Atlas of US Physician Practice Choice and Firm Structure: 2015–2020, for Selected States from the FTC Physician and Facility Merger Study,” utilizes a novel dataset of commercial insurance claims collected as part of the FTC’s Physician Study.
We interview one of the authors, Daniel Deibler, now an associate at Cornerstone Research, on some of his key takeaways from the report.
1. This report is referred to as an “Atlas.” What was the primary motivation behind this project, and how does it relate to the FTC’s Physician Study?
We sought to provide a systematic, data-driven analysis of how the organization and delivery of healthcare in the U.S. has evolved over the last 20 years. While previous research in this area often focused on inpatient hospital care or care provided by Medicare, this project expands that inquiry into physician services to understand how patients receive care, what types of firms are providing it, and the state of competition between those firms.
We sought to provide a systematic, data-driven analysis of how the organization and delivery of healthcare in the U.S. has evolved over the last 20 years.
This report is a direct product of the FTC’s Physician Group and Healthcare Facility Merger Study. It utilizes data specifically collected for that broader project—consisting of comprehensive claims from six major insurers across 15 states—to create a “geography of choice.” By mapping a consistent set of measures of competition and concentration common to antitrust practitioners and the enforcement agencies, we can visually describe the relationship between where people live and the types of healthcare choices available to them.
2. One of the key findings relates to “patient choice.” How does patient choice vary across different medical specialties?
The report finds that patients’ available provider options vary substantially by physician specialty. While primary care remains relatively fragmented, other specialties like cardiology and orthopedic surgery are significantly more concentrated.
The study finds that 18% of primary care is delivered in ZIP codes classified as “highly concentrated” under the 2023 Merger Guidelines. In contrast, the percent of physicians in highly concentrated ZIP codes is 42% for OB-GYN, 51% for orthopedic surgery, and 58% for cardiology.
3. There has been a lot of discussion regarding physicians joining large health systems. What did your data reveal about the extent to which care is provided by physicians who are part of an organization with greater “scope”?
Many physicians practice in multi-specialty practices (MSPs) and/or as part of a firm that is affiliated with a larger health system. For example, 76% of cardiology care is provided in MSPs, and 46% is delivered by physicians affiliated with health systems. Health system integration for primary care is particularly high in the Midwest and low in Texas. However, even in Texas, there are many ZIP codes with relatively high rates of primary care provided by a system.
This type of information is enormously useful to antitrust practitioners and will help us better understand the scope of options available to patients in local areas.
This type of information is enormously useful to antitrust practitioners and will help us better understand the scope of options available to patients in local areas. It will also give researchers the ability to investigate whether care being delivered as part of a health system or MSP impacted quality and whether there are tradeoffs between higher quality care and increased concentration.
4. Your research highlights a significant urban-rural divide. What should policymakers understand about how geography influences patient choice in healthcare?
Geography is a major factor in determining available choices. Measuring concentration with the commonly-used Hirschman-Herfindahl Index (HHI), we find that urban ZIP codes are generally less concentrated than rural ones. In non-urban areas, the median HHI is significantly higher, and patients are often twice as likely to face highly concentrated care. In my opinion, this could be related to a number of market factors, such as the supply of physicians in rural areas, demand (or lack thereof) for certain specialties, or the socioeconomic composition of the population.
That said, we also found significant variation across states that highlight the importance of local conditions. For example, nearly all OB-GYN care in Florida occurs in highly concentrated ZIP codes (including urban areas), whereas most ZIP codes in Texas (even many rural ZIP codes) are not highly concentrated for those same services.
This report is a direct product of the FTC’s Physician Group and Healthcare Facility Merger Study.
5. Finally, the study notes that the concentration measures aren’t moving in a single direction everywhere. What does this “dynamic” landscape mean for the future of antitrust and economic research?
Although the median change in HHI is positive across specialties, a substantial portion of ZIP codes actually experienced decreases in HHI between 2015 and 2020. This result has two main takeaways.
First, the organization of healthcare firms is highly specific and often idiosyncratic based on geography and specialty. Understanding how much choice patients currently have based on location and their diagnosis is an essential first step for analyzing the impact of concentration on how healthcare is delivered.
Second, the report offers key statistics for each geography and specialty, describing the overall shape of healthcare delivery. A key goal of this paper is to create a valuable public database for antitrust practitioners to better understand the current landscape of patient choice within a given locality and specialty.
5 Questions is a periodic feature produced by Cornerstone Research, which asks our affiliated experts and senior professionals to answer five questions related to their work. The views expressed here are those of the authors and do not necessarily represent the views of Cornerstone Research or the FTC.
Washington 
