Whether it is American senior citizens driving into Canada in order to buy cheap prescription drugs or Canadians coming to the U.S. for surgery in order to avoid long wait times, the relative merits of these two nations' health care systems are often cast in terms of anecdotes. Both systems are beset by ballooning costs and, especially with a presidential election on the horizon, calls for reform, but a recent study could put ammunition in the hands of people who believe it is time the U.S. ceased to be the only developed nation without universal health coverage.
Gordon H. Guyatt, a professor of epidemiology and biostatistics at McMaster University in Hamilton, Ontario, who coined the term "evidence-based medicine," collaborated with 16 of his colleagues in an exhaustive survey of existing studies on the outcomes of various medical procedures in both the U.S. and Canada. Their work appears in the inaugural issue of the new Canadian journal Open Medicine, and comes at a time when many in Canada are debating whether or not to move that country's single-payer system toward for-profit delivery of care. The ultimate conclusion of the study is that the Canadian medical system is as good as the U.S. version, at least when measured by a single metric—the rate at which patients in either system died.
"Other people knew that Canadians live two to two and a half years longer than Americans," says Steffie Woolhandler, an author on the paper and an associate professor of medicine at Harvard Medical School, citing a phenomenon that many attribute to differences in lifestyle between the two countries. "But what was not known was once you got sick, was the quality of care equivalent in the two countries."
Americans Less Likely to Survive Treatment
According to Woolhandler, by looking at already ill patients, the researchers eliminated any Canadian lifestyle advantage and just examined the degree to which the two systems affected patient deaths. (Mortality was the one kind of data they could extract from a disparate pool of 38 papers examining everything from kidney failure to rheumatoid arthritis.)
Overall, the results favored Canadians, who were 5 percent less likely than Americans to die in the course of treatment. Some disorders, such as kidney failure, favored Canadians more strongly than Americans, whereas others, such as hip fracture, had slightly better outcomes in the U.S. than in Canada. Of the 38 studies the authors surveyed, which were winnowed down from a pool of thousands, 14 favored Canada, five the U.S., and 19 yielded mixed results.
Mortality Isn't the Only Measure That Matters
Not all experts agree with the implication that the Canadian system is better than the U.S. system, however, or with the researchers' methodology. Vivian Ho, who is the James A. Baker III Institute for Public Policy chair in health economics at Rice University in Houston and has spent time living and conducting research in both the U.S. and Canada, argues that the study's focus on mortality could be misleading.
"When we look at health systems we look at other things than death," Ho explains. In her own research on hip fracture, which was cited in Guyatt's study, she found that the time a patient had to wait before surgery—which was significantly longer in Canada than the U.S. because of a shortage of operating rooms—made only a 1 percent difference in terms of mortality.
"But certainly if you ask people waiting in the hospital," Ho notes, "They're going to say I'd rather have the U.S. system . Waiting means there's a significant amount of distress for an elderly patient, and also higher complications for pneumonia because you have the patient immobile for so long."
Patti Groome, an epidemiologist at Queens University Cancer Research Institute in Kingston, Ontario, said she believes that overall the paper was balanced. "But when you get into [the] meat of [the] paper they can't sort out what's going on . There's way too much heterogeneity in these studies to come to a conclusion about these systems." In meta-analyses such as this one, "heterogeneity" in results corresponds to variations in the size of an effect across the studies being reviewed.