Tuesday, June 18, 2013

$10 billion per year to train the wrong physicians

Earlier this year, the physicians at my academic family medicine practice met with two senior officials from our parent health care organization to be oriented to its new initiatives and projects. Their presentation documented the organization's ongoing investments of many millions of dollars into renovating subspecialty care suites and purchasing new radiology equipment that was likely to be highly profitable, but provide dubious benefits to patients. Two of my colleagues asked why, given the expected influx of millions of newly insured patients into primary care starting in 2014, and an estimated shortfall of more than 50,000 primary care physicians by 2025, the organization had not identified expansion of primary care training as a financial priority. Where exactly did they expect to find family physicians to staff all of the new community offices they planned to open? An awkward silence ensued, followed by some polite hemming and hawing about how this was a complicated issue, and that supporting generalist training would likely require additional funding that was perhaps beyond the organization's limited resources.

Additional funding required? How about $9.5 billion? That's the approximate amount that that Medicare spends each year, with no strings attached, to subsidize the cost of training physicians in U.S. residency programs. Noting that the federal government doles out these dollars without requiring any particular outcomes from the institutions that benefit from them, some have called for Medicare to hold institutions more accountable for meeting America's physician workforce needs. If we have a surplus of radiologists and a shortage of general surgeons, why not tie funding to training more of the latter and fewer of the former? Given the decentralized nature of the U.S. health system, though, that has been easier said than done. In particular, it is challenging to follow the money trail and determine which institutions end up producing which types of doctors.

A new study in Academic Medicine by health services researchers at The George Washington University and the Robert Graham Center fills this information gap. Painstakingly assembling and cross-checking data from several sources on actively practicing physicians who completed their residency training from 2006 to 2008, they were able to identify residency-sponsoring institutions that were top producers of primary care physicians, that produced lower proportions relative to all physicians, and that produced none at all. Notably, they conclusively disproved "The Dean's Lie" that counts all internal medicine residents as going into primary care (when only 1 in 5 actually plan to do so), demonstrating that at some institutions fewer than 1 in 10 internists become primary care physicians. They also identified a large funding discrepancy between the top and bottom primary care producers.

The top 20 primary care producing sites graduated 1,658 primary care graduates out of a total of 4,044 graduates (41.0%) and received $292.1 million in total Medicare GME payments. The bottom 20 graduated 684 primary care graduates out of a total of 10,937 graduates (6.3%) and received $842.4 million.

In short, where physician production is concerned, you get what you pay for. In this case, Medicare pays a disproportionate amount of its nearly $10 billion per year in subsidies to institutions that produce mostly subspecialists, even in specialties where supplies are plentiful, at the expense of training sorely needed family physicians and other generalists whose presence has been shown time and again to deliver better health outcomes.

That's the big picture. Since all politics is local, policymakers who want to know what types of physicians their teaching hospital or health system is training can use the Graham Center's free GME Outcomes Mapper tool to find out. And if enough of them do so, maybe we can all have a serious national conversation about moving beyond guaranteed health insurance coverage to ensuring that the care (and the workforce) that coverage is paying for will actually help us to live longer or better.