There is an article of faith among proponents of health care reform that expanding the primary care workforce, particularly at the expense of medical and surgical specialists, is a critical component of "bending the cost curve" and enhancing access to care. This approach is favored in several models of health care delivery, particularly using the concept of a "gatekeeper" physician who follows guidelines of care to prevent over- utilization of expensive care by specialists. Increased training of and reliance on primary care physicians is a stated goal of various health care proposals and has been encouraged in the past through specific funding mechanisms by the government. It has been one of the major sources of frustration with the managed care paradigm of control of utilization of medical services.
One optimistic view of how this gatekeeper model would work has the general practitioner with an orderly and happy group of patients waiting at the "gate" to find out whether they need anything more from the health care system and then gaining appropriate and easy access to advanced care when that test or procedure is indicated. Another view, perhaps a little more clear-eyed, has the primary care physician having a default position of avoiding any expensive treatment or test provided by specialists unless there is overwhelming evidence that the procedure or test will be highly likely to be effective, not just possibly effective. Which one is accurate? Are there data on the effectiveness of primary care physicians compared to specialists in the treatment of a variety of diseases? Is there not an intuitive problem with assuming a "generalist" can be as knowledgeable and as efficient and effective as a well-trained specialist given the ongoing explosion in medical knowledge and technical capabilities?
At one extreme of the medical spectrum, the case of Senator Edward Kennedy's tragic illness illustrates the dilemma posed by a health care system with enormously expensive therapies of often uncertain benefit but also offering great hope to patients with serious illnesses. After the initial diagnosis of the cause of his seizure at one of America's great hospitals in Boston, the senator and his family opted for further neurosurgical care at another of America's great hospitals in Durham, N.C. He then returned to Boston to receive chemotherapy and highly advanced radiation therapy. After what was assuredly an enormous cost, he died with his family at his side after about one year. Was all this proper and necessary? Did his need for hope and optimism about his illness outweigh the expense and effort of a treatment program that had little if any chance of a substantial prolongation of his life? Those who argue for major cost constraints on aggressive, high tech care will argue this example illustrates the need for a more rational and guideline-based care that does not opt for high cost but low-probability-of-success forms of care. Others would argue that the opportunity to receive innovative and advanced care should be the cornerstone of the American system. Clearly the senator and his family made their choice. No doubt it was based on the completely understandable impulse to seek out the best for ourselves and our families.
Yet, those who propose to use his career and his commitment to a type of health care reform that will count age and quality of life as criteria for the use of health care dollars clearly are in an ethical quandary if guideline-based care under the auspices of a gate-keeping generalist physician underlies any proposed plans--no guidelines will include required travel to the great academic medical centers of America as part of the protocol. No one disputes that under any proposed system, desperately ill patients will have access to specialty care. The question is whether that care will maintain the current high standards of advanced technology and innovative approaches to treatment if it is consistently underfunded or even implicitly discouraged.
Recent studies of the effectiveness of primary care based approaches to the management of more common illnesses suggests that even in our current system, limited access to specialty care has important negative consequences. Patients with an acute myocardial infarction were studied in 2004 at two hospitals in Norwich, England where guidelines dictated a set approach to care as determined by the National Health Service. While these guidelines included the rule that patients who suffered the heart attack would be seen by a cardiologist during their hospitalization, one group of nearly 300 patients did see the cardiologist but another quite similar group of patients never saw a cardiologist and were cared for by a primary care physician. Over the ensuing 18 months, the mortality rate was 75 percent higher in the group that did not see the specialist. The increased mortality was attributable to the failure of the primary care physicians to adhere to prescribing medications known to be effective in preventing late death after myocardial infarction. A similar set of observations have been made in the U.S.
Perhaps myocardial infarction care is too intense an example. Yet studies of the management of hypertension, a condition certainly highly familiar to primary care physicians, reveal a frustratingly similar set of results. In an assessment carried out by investigators at Johns Hopkins University School of Medicine in 2008, primary care doctors failed to increase antihypertensive medications when clearly indicated in 87 percent of over 1300 patient visits where a substantial increase in previous blood pressure levels were found.
How about management of well recognized risk factors for cardiac disease? After all, much more robust preventative medicine is one goal of health care reform. By using a standard case vignette and querying family physicians and general internists about how they would treat a patient with well-defined cardiovascular risk factors, researchers at the University of California at San Francisco found that two thirds of primary care physicians inappropriately would choose to prescribe low dose aspirin to patients who, according to standard guidelines, should not have been treated in this manner. Additionally, 40 percent did not follow a recommended guideline for lowering cholesterol level.
Other research shows similar deficiencies. In a study of the appropriate referral for colonoscopy in patients found to have blood in their stool after screening tests, 31 percent of the time, patients were not appropriately referred for the procedure in a study conducted at the University of Pennsylvania School of Medicine and Thomas Jefferson University.
Perhaps the most discouraging finding in my own field of nephrology is the failure to appropriately refer patients with advanced kidney disease to a specialist. It has been clearly shown that primary care physicians typically refer patients with advanced kidney disease to nephrologists less than 6 months before the need for dialysis or kidney transplantation leading to increased mortality rates. Researchers at Bay State Medical Center and Tufts University School of Medicine have found that 62 percent of patients with moderately severe kidney disease were not recognized to have kidney disease by primary care physicians and were therefore not receiving specialist referrals or the medications that might slow the progression of their kidney disease.
Many individual primary care physicians are highly capable and knowledgeable and can provide high quality preventive care. As a group, however, they apparently do not achieve those goals. In truth, they really cannot be expected to given the growth in new information in each field of medicine. They must rely on specialists to support them and to achieve the highest quality care. To rely on guideline-driven, primary care to solve the problems of the high cost of medical care is an example of the triumph of hope over reality.
Constraints on access to specialty care will be a key dilemma in any proposed reform of health care delivery. It is a myth, although quite widely accepted and advocated, that reducing expert but expensive specialist care will lead to better medical care. The myth is based on the idea that there is more uniformity than singularity among patients and that most care can be defined by guidelines. It may be true that for those with non-life threatening conditions that have only moderate intensity, that generalists could apply guidelines and deal with the problems in an efficient manner. But those with complex illnesses and advanced complications of highly prevalent diseases like Diabetes will not thrive in a generalist dominated world. The fear expressed by most Americans about the content of their medical care is not that their mild hypertension will be untreated but rather that if they develop a life threatening malignancy like that of Senator Kennedy, that they will receive every opportunity to avoid death and prolong meaningful life. This will not be achieved by a health care system that does not support an innovative and technologically adept specialist physician workforce. How to achieve wide access to specialists and contain costs is the conundrum that the current proposals do not begin to address. It just is not enough to simply cut the payments to physicians and hospitals. That will not generate more access to advanced care.
Stanley Goldfarb MD is associate dean of clinical education at the University of Pennsylvania School of Medicine and a nephrologist.