The National Residency Matching Program (NRMP) uses a computer algorithm to match medical residents to hospitals, addressing timing problems in the medical labor market that historically led to increasingly early appointment dates and frenzied decision-making processes.
This report describes the National Residency Matching Program (NRMP), a computer algorithm-based system used to match medical residents to hospitals in the United States. The system was developed in response to historical timing problems in the medical labor market. Prior to 1945, hospitals competed by setting hire dates earlier than competitors, leading to appointment dates that 'unraveled' from the end of senior year to 2 years before graduation by 1944. Medical schools implemented embargoes on letters of reference in 1945 and gradually moved appointment dates back to senior year. However, new problems emerged with increasingly shortened decision deadlines - from 10 days in 1945 to 12 hours by 1949. This created a frenzied market with missed opportunities and hasty agreements that were sometimes not honored. The National Intern Matching Program, predecessor to the NRMP, was introduced in 1952 to address these market congestion issues. The report notes that similar timing problems and algorithmic solutions have been implemented in various other professional labor markets. The author, an economist, was asked by the NRMP to direct a study examining what kind of matching algorithm should be employed, suggesting there is ongoing disagreement about the optimal algorithmic approach.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI developers or state-like actors competing in an AI ‘race’ by rapidly developing, deploying, and applying AI systems to maximize strategic or economic advantage, increasing the risk they release unsafe and error-prone systems.
Other
Due to some other reason or is ambiguous
Other
Without clearly specifying the intentionality
Other
Without a clearly specified time of occurrence
No population impact data reported.