Correspondence
Health Care in Canada and the United States
N Engl J Med 1993; 329:964-966September 23, 1993
- Article
To the Editor:
Two Special Articles in the March 18 issue1,2 provide detailed information on differences in health care between Canada and the United States. The authors of these articles fail to comment on two important issues, however. In “Hospital Expenditures in the United States and Canada,”1 Redelmeier and Fuchs note that visits to the emergency department per 1000 population are twice as frequent in Canada as in the United States. Furthermore, the patients making these visits appear to be ill, since the percentage of visits that result in admission is the same in both countries. It is shocking that the number of emergency admissions in Canada per 1000 appears to be twice as high as it is in the United States. The removal of financial barriers to access to care and the provision of a larger number of primary care physicians should have decreased the number of visits to the emergency department dramatically in Canada. Perhaps the large number of visits to the emergency department and resultant admissions is due to inadequate access to physicians in their offices.
In their article comparing management patterns after acute myocardial infarction in Canada and the United States, Rouleau et al.2 conclude, quite appropriately, that in Canada only patients with a clinical probability of myocardial infarction greater than 75 percent were admitted to the coronary care unit and that a more conservative diagnostic and therapeutic strategy was followed there than in the United States. They note no effect of this strategy on morbidity and mortality. However, Rouleau et al. fail to comment on the five times greater likelihood that a patient in the United States will initiate a malpractice suit if that person feels harmed by the failure of a physician or hospital to provide necessary care. When a person is injured, the legal system in the United States assumes that cost is not an issue and that an intervention might have benefited the patient. In addition, a physician could be charged with a substantial deficiency in quality of performance by a peer review organization if a patient who is at even low risk of having a myocardial infarction had such an event but was not admitted to a coronary care unit.
2 ReferencesJohn F. Schneider, M.D., Ph.D.
University of Chicago, Chicago, IL 606371
Redelmeier DA ,Fuchs VR . Hospital expenditures in the United States and Canada. N Engl J Med 1993;328:772-778
Full Text | Web of Science | Medline2
Rouleau JL ,Moye LA ,Pfeffer MA , et al. A comparison of management patterns after acute myocardial infarction in Canada and the United States. N Engl J Med 1993;328:779-784
Full Text | Web of Science | Medline
To the Editor:
The selection by Redelmeier and Fuchs of the individual hospitals to be used in their comparisons substantially affected the results, particularly with regard to length of stay. Whereas Wellesley Hospital in Toronto is indeed more community-oriented than most other Canadian teaching hospitals, it is still a full teaching facility and to a great extent a tertiary care referral center, with all the associated patterns of clinical practice. The article reinforces this role by comparing the use of lithotripsy between Wellesley and St. John's, rather than between Sunnybrook and Stanford.
If the intent was to compare two teaching hospitals and two community hospitals, selecting a true community hospital would have provided a more accurate picture of the hospital experience in Ontario.
The article used Canadian data for 1987-1988, citing a mean length of stay of 11.2 days for Canada, 10.6 days for Ontario, and 11.9 days for Wellesley Hospital. These figures were obtained from published data from the Hospital Medical Records Institute, the independent hospital data-base management organization to which Canadian hospitals report abstracted information on all discharges from acute care hospitals.
If one uses this same national data base, the mean length of stay in 1987-1988 for the nine community hospitals in Ontario with more than 400 beds was 7.9 days. This figure includes discharges of newborns as well as adults and children. (It is not clear from the article by Redelmeier and Fuchs whether newborns were included in their analysis.) If newborns were excluded, the mean stay in Ontario's large community hospitals would increase to approximately 8.2 days -- still well below the figure of 11.9 days for Wellesley Hospital. The difference in length of stay between St. John's Hospital in California (6.6 days) and a typical large community hospital in Ontario would then be either 1.3 or 1.6 days, depending on whether newborns were included, but not 5.3 days. This makes the factors related to administration and the use of centralized equipment less important in the cost comparisons.
Interestingly, data from the Hospital Medical Records Institute for 1991-1992 indicate that the mean stay in large community hospitals in Ontario (with newborns included) had fallen to 7.1 days.
Sherryn Levinoff Roth, M.D., F.R.C.P.(C)
Catherine Cornell
Scarborough General Hospital, Scarborough, ON M1P 2V5, CanadaTo the Editor:
Redelmeier and Fuchs suggest that Canadian hospitals are more cost efficient than their American counterparts because of centralized technology and the establishment of waiting lists. Hidden costs, however, become apparent when one examines the application of this approach to dialysis therapy. This expensive therapy is hospital-based and funded through global hospital budgets.
In the United States, 87 percent of all patients undergoing dialysis are undergoing hemodialysis1. The average number of patients undergoing hemodialysis per dialysis station in a dialysis center is 3.9. The United States has more than 97.8 such stations per million population.
The situation is different in Canada, and particularly in Toronto, where there are 37.2 and 23.3 hemodialysis stations per million population, respectively2. In Canada, 62.3 percent of patients are treated with hemodialysis (49 percent in Toronto). Toronto treats 5.49 patients receiving hemodialysis per station. With an operation with three shifts a day, six days a week, the theoretical maximum is six patients per station. Toronto patients receive the more cost-effective peritoneal form of dialysis preferentially, usually without choice, and sometimes even in spite of frequent or severe peritonitis. Overcrowding has created queues for hemodialysis stations and has caused patients to be transferred to centers 100 miles (160 km) away to receive the treatment they need.
The absolute number of patients undergoing dialysis in North America is growing by 10 percent per year. The Canadian policy of limiting funding for the procedure allows this growth to be managed in a cost-effective way. Superficially, this seems to be a successful approach. Our political leaders can point out that in terms of crude mortality rates, Canadian patients undergoing dialysis fare better than their American counterparts3. However, in addition to lack of choice, there are other hidden costs. The incidence of therapy for end-stage renal disease is much lower in Canada than in the United States, even after adjustment for racial factors. There is no biologically plausible reason for this difference. There is concern that the difference in incidence can be attributed in part to indirect rationing of dialysis through lack of referrals by primary care physicians. Faulty principles of bureaucratic planning (and not input from consumers) are used to allocate resources. This should be a lesson for Americans.
3 ReferencesD.C. Mendelssohn, M.D., F.R.C.P.(C)
K.L. Skorecki, M.D., F.R.C.P.(C)
C.J. Cardella, M.D., F.R.C.P.(C)
University of Toronto, Toronto, ON M5G 2C4, Canada1
U.S. Renal Data System. USRDS 1991 annual data report. Bethesda, Md.: National Institutes of Health, August 1991.
2
Canadian Organ Replacement Register. 1990 Annual report. Don Mills, Ont.: Hospital Medical Records Institute, April 1992.
3
Hakim RM . Assessing the adequacy of dialysis. Kidney Int 1990;37:822-832
CrossRef | Web of Science | Medline
To the Editor:
Rouleau et al. imply that a conservative (Canadian) strategy is as effective as a more aggressive (American) approach in decreasing long-term mortality after myocardial infarction. This suggestion must be considered with extreme caution. Prospective, randomized trials have shown that in certain coronary subgroups, particularly patients with left main or triple vessel disease, severe symptoms, and left ventricular dysfunction, revascularization (“aggressive”) is superior to medical therapy (“conservative”) for survival and relief from angina1-4. Rouleau et al. state that their study “had limited power to detect differences between the two countries in mortality and recurrent myocardial infarction.” Nevertheless, these data may be used by health care planners to justify a “Canadian” strategy -- i.e., a less aggressive and less costly approach to coronary disease after myocardial infarction. Such an approach is not justified by the data.
4 ReferencesBarry F. Uretsky, M.D.
University of Pittsburgh, Pittsburgh, PA 152131
European Coronary Surgery Study Group
Web of Science | Medline2
The Veterans Administration Coronary Artery Bypass Surgery Cooperative Study Group
Full Text | Web of Science | Medline3
Passamani E, Davis KB, Gillespie MJ, Killip T, CASS Principal Investigators and Their Associates
Full Text | Web of Science | Medline4
Varnauskas E, European Coronary Surgery Study Group
Full Text | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: Schneider calls attention to our observation that Canadians are more than twice as likely as Americans to visit an emergency department and be treated there. He suggests that perhaps Canadians have inadequate access to primary care physicians in their offices. But, other studies indicate that Canadians are more likely than Americans to contact a physician and receive outpatient care1-3. If there is a problem with access to primary care physicians in Canada, this problem is probably worse in the United States.
Mendelssohn et al. address services for patients with renal failure. As is true for other selected procedures (such as cardiac surgery, knee replacement, and cataract extraction), the average Canadian is less likely to receive renal-replacement therapy than the average American. We do not know, however, how much of the difference may reflect unnecessary care in the United States, differences in the amount of uncontrolled hypertension, diabetes, and glomerulonephritis in the two countries, or differences in patients' preferences in the two cultures.
Roth and Cornell focus on length of stay at selected community hospitals in Ontario. Their figures from the data base of the Hospital Medical Records Institute are underestimates, because the source excludes patients whose hospital stay exceeded a standard length. Our data for both the United States and Canada include all patients. The discrepancy between data sources highlights the importance of establishing comparable definitions and methods when international differences in health care are analyzed.
3 ReferencesDonald A. Redelmeier, M.D.
University of Toronto, Toronto, ON M4Y 1J3, CanadaVictor R. Fuchs, Ph.D.
Stanford University, Stanford, CA 943051
Katz SJ, Hofer TP. Comparing the use of outpatient physician visits in Canada and the United States: results from the first Ontario health survey. In: Proceedings of the 16th Annual National Meeting, Society of General Internal Medicine, Arlington, Va., April 29, 1993. Washington, D.C.: Society of General Internal Medicine, 1993. abstract.
2
Schieber GJ ,Poullier JP ,Greenwald LM . Health care systems in twenty-four countries. Health Aff (Millwood) 1991;10:22-38
CrossRef | Web of Science | Medline3
Fuchs VR ,Hahn JS . How does Canada do it? A comparison of expenditures for physicians' services in the United States and Canada. N Engl J Med 1990;323:884-890
Full Text | Web of Science | Medline
Author/Editor Response
We agree with Uretsky that revascularization procedures are indicated in a number of subgroups of patients after a myocardial infarction. We also agree that a randomized study evaluating a conservative as compared with an invasive approach to the investigation and treatment of patients with left ventricular dysfunction after an infarction is required to establish the best strategy. Nevertheless, our study found no significant difference in the rates of recurrent myocardial infarction and death between the United States and Canada, despite important differences in the frequency of use of revascularization procedures. This suggests that the frequency of use of these procedures could be safely reduced in this patient population. Our findings are consistent with those of phase IIB of the Thrombolysis in Myocardial Infarction trial. This trial suggested that a conservative strategy for the diagnosis and care of patients without overt myocardial ischemia and with negative exercise stress tests after an infarction does not significantly modify rates of reinfarction or mortality.1
We agree with Schneider that the threat of a malpractice suit is much greater in the United States than in Canada and that this may influence clinical management. This observation, however, does not invalidate the principal point of the paper; the frequency of adverse end points was similar in the Canadian and the U.S. patients, despite marked differences in the use of catheterization and revascularization.
1 ReferencesJean L. Rouleau, M.D.
Montreal Heart Institute, Montreal, QC H1T 1C8, CanadaEugene Braunwald, M.D.
Marc A. Pfeffer, M.D., Ph.D.
Brigham and Women's Hospital, Boston, MA 021151
The TIMI Study Group
Full Text | Web of Science | Medline
