3 takeaways from a county-level exploration of the US radiologist workforce

Researchers performed a county-level analysis of the U.S. radiologist workforce, breaking down both the overall supply of radiologists in the nation as well as the subspecialization of those radiologists, and published their findings in the Journal of the American College of Radiology.

“For a variety of reasons, patient care may best be advanced through robust access to both generalist and subspecialist radiologists,” wrote lead author Andrew B. Rosenkrantz, MD, MPA, of the department of radiology at NYU Langone Medical Center in New York City, and colleagues. “Nonetheless, although geographic patterns of radiologist supply overall have received recent attention, geographic patterns of radiologist subspecialization are not well known. In this study, aiming to inform initiatives to right-size the US radiologist workforce, we explore associations between county-level measures of radiologist supply and subspecialization and county structural and health-related characteristics.”

The authors studied data from more than 32,000 radiologists participating in Medicare between 2012 and 2014, comparing the overall supply of specialists and subspecialization in more than 3,00 counties throughout the United States.

These are three key takeaways from their findings:

1. More than 58 percent of U.S. counties do not have a Medicare-participating radiologist. More than 78 percent of U.S. counties do not have a Medicare-participating subspecialist radiologist.

Rosenkrantz et al. noted that their findings indicate the radiology workforce might include more significant disparities than researchers previously believed. Studies that did not consider “geographic variation,” they add, “may have underestimated the degree of disparities in comprehensive access to radiological care at both the state and county level.”

“By focusing our work more granularly at the county level and further considering the substantial variation in individual radiologists’ unique practice characteristics, we show herein that those prior concerns of maldistribution of the radiologist workforce are likely further confounded by superimposed variation in the subspecialization of radiologists within given communities,” the authors wrote. “Although many counties had no Medicare-participating radiologists, a far larger number of counties lacked a radiologist with a majority subspecialty practice.”

2. Rural counties with smaller populations have fewer radiologists overall and fewer subspecialists.

“Counties’ radiologists per 100,000 population, generalist radiologists per 100,000 population, subspecialist radiologists per 100,000 population, and percent of radiologists subspecialized all showed moderate positive correlations with counties’ populations and moderate negative correlations with counties’ rural percentages,” the authors wrote.

To potentially help rural counties with smaller populations, the authors suggested that academic medical centers could develop partnerships with providers in those counties and “provide subspecialist radiologist support.” Teleradiology and “regional consolidation initiatives” were also listed as ways to improve care for patients from these areas.

3. Counties with at least one Medicare-participating radiologist had higher mammography screening rates, lower premature deaths.

In addition, the authors added, more radiologists overall “was weakly associated with higher mammography screening rates” while more subspecialization “was weakly associated with lower premature deaths.” What caused these differences? Rosenkrantz and colleagues aren’t sure, but they added that it warrants further research.

“The causes for the observed associations of radiologist workforce measures with mammography screening rates and premature death, without a change in costs, is unclear from our correlative analysis and likely relates to the impact of other variables for which this investigation does not account,” the authors wrote. “Although not the primary focus of this study, these suggested associations illustrate the considerable geographic disparities in patient access to local radiologist services and warrant further targeted research to better understand.”