Efforts in the early 20th century to improve the quality of medical education in the United States led to a steep decline in the number of medical schools and medical school graduates. In a new study, researchers examined the consequences of these medical school closures between 1900 and 1930 for the number of county-level physicians, nurses, and midwives, and for infant, non-infant, and total mortality. The closures led to a 4% reduction in physicians per capita and resulted in declines in infant mortality, non-infant mortality, and total mortality, they found.
The study, by researchers at Carnegie Mellon University, Stanford University, and Marquette University, appears as an NBER Working Paper.
"Our findings build on previous research on the historical supply and distribution of doctors and nurses in the United States, and offer new evidence on the consequences of medical school closures for nurses and midwives," says Karen Clay, professor of economics and public policy at Carnegie Mellon's Heinz College, who led the study.
Many U.S. medical schools opened in the late 19th century, and many were commercial and not affiliated with universities, leading to concern about their quality. The 1910 publication of Medical Education in the United States and Canada—written by Abraham Flexner at the request of the Carnegie Foundation and known as the Flexner Report—was among the most prominent reforms in the history of U.S. medicine. An evaluation of the quality of every medical school in the United States, the report called for closing most of them. Between 1905 and 1915, more than 40% of U.S. medical schools closed or were absorbed by other institutions.
In this study, researchers sought to determine the effects of Flexner-era medical school closures on mortality. Using a variety of data (e.g., U.S. Census data from 1900 to 1930, county-level vital statistics data on mortality), they constructed a measure of school closure intensity for all U.S. counties, combining variation in distance from closures, the historical number of graduates from closing schools, and the timing of the closures. Among the study's findings:
- Geographic areas near medical school closures were the most heavily affected: On average, in counties within 300 miles of a closed medical school, the number of physicians declined 4%.
- The effects came from a mix of decreases in young physicians per capita, the flow of which was reduced in places near closures, and old physicians per capita, who suffered reputational effects if the medical school where they trained closed.
- Market adjustment to these closures was significant: Physicians migrated to counties with higher school closure intensity values and physicians in counties with higher school closure intensity values postponed retirement. Both responses helped offset the decrease in new medical graduates.
- Medical school closures within 300 miles also led to increases in nurses per capita and had no effect on midwives per capita: On average, a county with average closure intensity experienced a 7% increase in nurses.
- Infant, non-infant, and total mortality declined as a result of school closures within 300 miles: On average, a county with average closure intensity saw a 8% reduction in infant mortality, a 4% reduction in non-infant mortality, and a 3% reduction in total mortality.
- These estimates suggest that 16,000 infant lives and 38,000 non-infant lives were saved annually by closing low-quality medical schools.
- The reductions were driven primarily by causes that were likely sensitive to physician quality, including reductions in infectious diseases and diseases of early infancy. Other factors (e.g., the presence of a county public health department; the number of county health department personnel; the number of hospitals and hospital beds; city-level public health spending on sewers, water, and refuse) were either unrelated or negatively related to medical school closures.
"Although we cannot fully isolate the role of physicians given other market adjustments in response to medical school closures, reducing the supply of poorly trained physicians appears to have reduced mortality," explains Margarita Portnykh, assistant teaching professor of business analytics and economics at Carnegie Mellon's Tepper School of Business, who coauthored the study.
The study was supported by Heinz College at Carnegie Mellon University.
