A Closer Look at the Quality of Cardiac Care in the United States
David S. Jones, MD, PhD, is the A. Bernard Ackerman Professor of the Culture of Medicine at Harvard Medical School's Department of Global Health & Social Medicine. He is a 2007 recipient of a Robert Wood Johnson Foundation (RWJF) Investigator Award in Health Policy Research, and the author of Broken Hearts: The Tangled History of Cardiac Care.
Every day, all over America, people visit their doctors with chest pain and other symptoms of coronary artery disease. Each year, more than a million of them choose to undergo bypass surgery or angioplasty. Are these decisions good ones? Even though modern medicine has committed itself to an ideal of evidence-based medicine, with its clinical trials, meta-analyses, and practice guidelines, the answer is not always clear. By looking closely at the history of these procedures, it is possible to understand some of the reasons why this is the case. With support from a RWJF Investigator Award in Health Policy Research, I looked at three specific questions: the role of evidence and intuition in medical decisions, the reasons why it can be so difficult to determine the risks of medical interventions, and the problem of “unwarranted variation” in medical practice.
Under the dictates of evidence-based medicine, decisions should be guided by the empirical outcomes of clinical research. But trial data has never monopolized medical decisions. Doctors and patients also pay attention to how treatments work. When an intervention directly addresses the perceived cause of a disease—as often happens with surgery—then doctors assume that it will work. Such thinking had a huge impact on bypass surgery and angioplasty. Belief that coronary artery disease was a plumbing problem, with the progressive growth of atherosclerotic plaques slowly plugging the arteries, motivated surgeons’ and cardiologists’ efforts to treat it by bypassing or compressing those plaques. The intuitive appeal of the procedures was so strong that doctors and patients maintained their faith even when clinical trials cast doubt on the procedures’ value. In principle there is nothing wrong with this balance between theory and empiricism. The question is whether the intuitions are correct.
Medical understanding of coronary artery disease has changed dramatically since the 1960s, especially with the advent of the plaque rupture hypothesis. In this model, some plaques become unstable and rupture, triggering a blood clot that plugs the artery and causes a heart attack. The problem is that it is difficult, maybe impossible, to predict in advance which plaques might rupture. Without knowing this, surgeons and cardiologists cannot choose their targets for bypass surgery or angioplasty. This leaves doctors in a situation in which the disease model has changed, undermining the rationale for two extremely popular interventions. Doctors and patients need to revise their intuitions, but history of bypass surgery and angioplasty demonstrates how complicated this process can be. The old intuitions still persist and leave patients and doctors with an exaggerated sense of the efficacy of revascularization.
Doctors have also under-estimated the risks of these procedures. When surgeons began performing open heart surgery in the 1950s and 1960s, they realized the heart-lung machines could cause strokes, seizures, delirium, and subtler forms of brain damage. Yet when surgeons introduced bypass surgery in the 1960s and 1970s, hardly anyone paid attention to its cerebral complications. When they finally did study the problem in the 1980s and 1990s, they often downplayed evidence of brain damage. Despite decades of research, the risk remains unclear today. If a patient, referred for bypass surgery, asks if there will be any complications, one surgeon could say “yes, we have to put you on a heart-lung machine; this is an imperfect substitute for the heart and introduces a risk of stroke, cognitive decline, and personality change.” Another surgeon could say “long-term cognitive outcomes are just as good in patients who have surgery as in those who do not, and sometimes even better.” How can this confusing situation persist? It is far easier to study the desired outcomes of an intervention than its expected or unexpected complications. As a result, doctors often end up with more thorough knowledge of a procedure’s efficacy than of its risks, an asymmetry that introduces a bias in favor of medical intervention.
To make matters worse, clinical decisions are not simply the result of risk-benefit analyses. Financial conflicts of interest, supplier-induced demand, local variations in medical judgment, and patient and doctor psychology all influence what decisions get made. As a result, medical practice, for bypass surgery, angioplasty, or almost any other medical treatment, varies between different hospitals, cities, states, and countries. This “unwarranted variation” remains poorly understood despite decades of research.
Because of these confounding factors, patients, doctors, and policy-makers need to think carefully about medical decisions. We need a healthy skepticism about claims of efficacy, whether they are grounded in clinical research or the intuitive logic of disease theories. We need to look past short-term surrogate markers and instead document all of the consequences of therapeutic interventions, both the good and the bad. And we need to be aware of the many non-clinical factors that impinge on clinical judgment. Paying closer attention to these ambiguities should result in more realistic expectations of efficacy. This, in turn, will improve both the quality of care and the satisfaction that patients and doctors take from it.
To learn more, read Broken Hearts: The Tangled History of Cardiac Care (Johns Hopkins University Press, 2013).