I spent a couple of hours today discussing a topic that has become increasingly important in the world in which we live, and which would have completely mystified an earlier generation of physicians. The subject was “attribution.” Simply put, how should one decide which patients “belong” to which doctors? On a more technical level, what algorithms should be employed to connect patients, or episodes of care for those patients, or specific quality measures pertaining to those patients, to particular physicians?
Here's why this is a hot topic. The Centers for Medicare and Medicaid Services is moving rapidly to alternative payment models. Medicaid is transitioning to a capitated system. Commercial payers are entering into “risk” arrangements with providers. All around us, fee for service is losing sway and is being replaced by a spectrum of new ways to pay for care. In the “old world” of fee for service, whoever provided the service got the fee. There was no mystery about how the dollars should flow. In the “new world” all that changes. In many instances, payments are linked to quality measures. So, for example, physician groups or integrated health systems may be subject to penalties or earn bonuses depending on how “their” patients do. Too many readmissions? Penalty. Excellent blood pressure control? Bonus. Simple enough in theory but complicated in practice.
Leaving aside all of the limitations of measuring quality in this reductionist way, such arrangements demand that patients be attributed to physicians. Who is responsible for that readmission – the hospitalist who saw the patient, the surgeon who operated during the admission, or the primary care internist who cared for the patient prior to and after discharge? Who is responsible for that excellent blood pressure control? The primary care physician, the cardiologist, or the psychotherapist? How do you even figure out who the primary care physician is?
As the discussion wore on, it became clear that there is no one way to do this. In all but the simplest of circumstances (e.g., well baby care) the “attribution model” that makes sense depends on the question being asked. Simply put, “who's in charge” is no longer a simple question. And that's when it struck me that this conversation would make absolutely no sense to our predecessors. I don't know about you, but I remember that there was never any ambiguity about who the “responsible physician” was when I was in training. On the surgical services, there was no question that the operating surgeon was responsible. He or she often sought the input of medical colleagues, but maintained “control of the case” and took responsibility for the outcome. On the medical service, the “physician of record” likewise owned the care. In the outpatient arena, most (insured) patients had a readily identified doctor who provided longitudinal care. As residents, that sense of responsibility was continuously drummed into us, and used to justify the crazy hours we worked.
Sure, I get it. Good care often requires the expertise of many. And the old “captain of the ship” model of care has plenty of shortcomings. Still, a world in which it is hard to identify the responsible physician is a world in which something of great value has been lost.
What do you think?
Ira S. Nash, MD, FACP, is the senior vice president and executive director of the North Shore-LIJ Medical Group, and a professor of Cardiology and Population Health at Hofstra North Shore-LIJ School of Medicine. He is Board Certified in Internal Medicine and Cardiovascular Diseases and was in the private practice of cardiology before joining the full-time faculty of Massachusetts General Hospital. He then held a number of senior positions at Mount Sinai Medical Center prior to joining North Shore-LIJ. He is married with two daughters and enjoys cars, reading biographies and histories, and following his favorite baseball team, the New York Yankees, when not practicing medicine. This post originally appeared at his blog, Ausculation.