"private" care. While the underlying reason for the debate is the same-relentless growth in the proportion of the gross domestic product allocated to health ...
John E. Wennberg, MD MPH
ill in the deleted expletive. South of the Canadian border, what springs to mind is "managed" care; north of the border, it is "private" care. While the underlying reason for the debate is the same-relentless growth in the proportion of the gross domestic product allocated to health care-the situations and solutions are quite different. Yet I fear that on both sides of the border the debate is leading us away from creative solutions-because in both nations we have failed to confront the conundrum of what constitutes medical necessity and therefore our ability to identify when valuable care is being rationed. Traditionally, medical necessity is dictated by what the doctor prescribes, and most policy analysts and patients have been raised to and assume that the basis for doctors' prescrip-
over [Expletive Deleted] Care
tions is medical science
and the physician's unerring judgment about what is best for the patient. Since, under this assumption, care is manifestly effica-
cious and of value to patients, care is perto rationed for ceived be when, whatever reason, limits are imposed on available resources. Participating in the Canadian version of the debate, DeCoster and Brownell ask in this issue of Public Health Reportsl whether the implementation of a two-tiered system would create a private care safety valve that releases pent-up demand, providing quicker, more efficient care to those who can afford it and making more care available to those who cannot. Their answer, at least for cataract surgery, is a resounding no. Using administrative databases, they show that in the first six years of this decade, demand for (utilization of) cataract surgery increased substantially in both private and public sector settings-in other words, that there was no substitution of private for public care. Overall costs went up, and poorer Canadians were almost as likely as richer Canadians to use the more expensive private system. The DeCoster and Brownell article comes close to
306 Public Health
the medical necessity/rationing of care conundrum by questioning the rationality of current patterns of surgical care. The authors cite the variations in practice among Canadian provinces to suggest that it is doubtful that disease incidence alone explains the demand for cataract services, noting that "in 1993-1994, age- and sexadjusted rates of cataract surgery across Canadian provinces ranged from 12.9 to 23.5 per 1000 population ages 50 and older, an 82% difference."l But DeCoster and Brownell fail to develop the implications of that variation-namely, that utilization rates might better reflect the flaws in the traditional process for allocating discretionary surgery than they do the needs and wants of patients. The treatment of common conditions such as cataracts, angina due to coronary artery disease, and urinary tract symptoms due to obstruction by the prostate gland all involve complex trade-offs between different treatment options, each of which has different costs, risks, and benefits. Early cataract surgery might be perceived as unnecessary by many patients if they were informed about the risks and benefits of the procedure. For those with coronary artery disease, dietary and drug therapies may be preferred over angioplasty or bypass surgery; for those with prostate conditions, watchful waiting or drugs may be preferred over surgery. The only way to know what treatment eligible patients want is to ask them and to do so in a setting that empowers them to choose according to their own preferences. Research on practice variation provides strong, albeit circumstantial, evidence that patient choice is not now determining the rates of use of surgery. The smaller the geographic unit of analysis, the wider the variation: data from Ontario show that the rates for discretionary surgery vary much more among the 33 District Health Councils in that province than they vary between provinces. In 1994-1995, lens extractions for cataracts ranged from a low of 5.3 to a high of 25.7 per 1000 residents 50 years and older, almost a fivefold range of variation. Coronary artery bypass procedures per 1000 residents 20 years and older ranged from a low of 0.4 to a high of 1.5, more than a threefold difference, and transurethral prostatectomies ranged from 0.6 to 1.3 per 1000 residents 50 years and older, differing by a factor of more than two. Research on shared decision-making, a strategy of clinical decision-making in which patients are fully informed about what is known and not known about the
July/August 1997 *Volume 1 12
Commentary on Rationing Care
outcomes of care (based on evidence-based technology assessments) and are invited to participate in the choice of treatment, provides evidence that demand for discretionary care changes when the patient is brought into the equation. Barry et al. report a 40% decline in utilization rates in two staff model health maintenance organizations (HMOs) once patients were offered shared decision-making.2 In a clinical trial in Toronto of shared decision-making about treatments for coronary artery disease, Detsky et al. found a 23% reduction in use of invasive treatment for coronary artery disease among patients randomized to shared decision making, compared to controls.3 '~ It is noteworthy that these studies were each conducted among groups .1 0 in which the rates of surgery have been low compared to most populations in North America. Under shared decision-making, the rates of transurethral prostatectomy in the two HMOs declined to a level seen in only two of the 306 hospital referral regions described in the Dartmouth Atlas ofHealth Care.4 Rates in Ontario are well below any of the other 301 hospital referral regions in the United States. The implication is that under shared decision-making, the demand for care may be substantially less than the amount now prescribed by physicians or approved by managed care companies. Quite apart from this consideration, however, the movement to shared decision-making offers the profession the opportunity to assure patients that the use of
July/August 1997 * Volume 112
discretionary care is in the best interest of the individual patient. If the bonus is reduced pressure on the expansion of resources devoted to elective surgery, so much the better for the beleaguered North American delivery systems, whether private or public. Dr. Wennberg is the Director of the Center for the Evaluative Clinical Sciences at Dartmouth Medical School. Address correspondence to Dr. Wennberg, 7251 Strasenburgh, Hanover NH 03 7553863; tel 603-650-1684; fax 603-650-1225. .
DeCoster CA, Brownell MD. Public S Health Rep 1997;112: 298-305. 2. Barry MJ, Fowler FJ, Mulley AG, Henderson JV, Wennberg JE. Patient reactions to a program designed to facilitate patient participation in treatment designs for benign prostatic hyperplasia. Med Care 1995;33: 771-82. 3. Morgan MW, Deber RB, Llewellyn-Thomas HA, Gladstone P, Cusimano RJ, O'Rourke K, Detsky AS. A randomized trial of the ischemic heart disease shared decision making program: an evaluation of a decision aid: the Toronto Hospital, University of Toronto, Ontario. J Gen Intem Med 1997,Apr Suppl 12:62. 4. Wennberg JE, Cooper MM, editors. The Dartmouth atlas of health care in the United States. Chicago: American Hospital Publishing, 1996. 1.
Public Health Reports