© 1998 by Oxford University Press
Poverty is hazardous to health. In the United States, studies have shown a link between cancer mortality and socioeconomic status (SES) -- the lower the SES, the higher the mortality rate. But recent studies led by Kevin Gorey, Ph.D., of the University of Windsor, Ont., show that residents of low-income areas in Canada may survive cancer longer than their U.S. counterparts.
Gorey hypothesized that Canada's comprehensive national health insurance program would result in greater survival for the economically disadvantaged. With equitable access to equivalent standards of care, Gorey felt, much of the survival disadvantage of poverty would disappear.
The U.S. General Accounting Office had looked at the cancer survival experience of the United States and Canada in 1994, comparing outcomes in lung, breast, and colon cancer, and Hodgkin's disease. The study found little difference in survival between the two countries. But the GAO study did not take into account socioeconomic differences, and Gorey suspected that it was in the low SES residents that the greater effect would be seen. He took it upon himself to look at SES, while comparing the cancer survival experience of the two countries in a controlled manner.
In order to test his hypothesis, Gorey compared Toronto, Ontario, and Detroit, Michigan, examining cancer survival after diagnosis for 15 common cancer sites. Patients were divided into low, middle, or high SES status according to census data on where they lived.
"I chose Toronto and Detroit because they were easy," Gorey said. Both cities have similar populations (between 3.5 and 4 million) and are within 200 miles of one another, both are on the Great Lakes, and both cities have similar weather and environmental factors.
The original study, published in the American Journal of Health in 1997, showed that SES was not linked to cancer survival after diagnosis for 12 of the 15 sites in Toronto, but was so linked in Detroit. Survival rates were comparable between the two cities for middle and upper class residents, but for the poorest third of residents, more Canadians survived than Americans for cancers of 13 of the 15 sites. For the five most common sites (lung, breast, prostate, colon, and bladder) there was anywhere from a 20% (prostate) to a nearly 100% (male lung) survival advantage for residents of Toronto. The two exceptions where no survival advantage existed were for rectal cancer and non-Hodgkin's lymphoma.
In a refinement of the study published earlier this year, breaking the population into five SES groups instead of three, the pattern still held. The Toronto survival advantage also held up when regions of Detroit with similar racial makeup were the only ones compared. And in data currently under peer review, Gorey made the same comparison with three other affluent cities in the United States, and saw the same survival advantage in lower SES residents in Toronto.
Hurdles to Analysis
Otis Brawley, M.D., director of Special Populations Research at the National Cancer Institute enumerated the many difficulties in conducting this kind of comparative research.
Brawley said it is difficult to determine SES in a rigorous way because comparing individuals is nearly impossible, and large studies are forced to assess SES indirectly. Most studies match up patients with census tract figures, he said, which determine the median household income for a given area.
But this method introduces an "ecological fallacy" because the range of SES of individuals in a neighborhood isn't immediately obvious, according to Brawley. For instance, if all the live-in maids in a high-income area come down with cancer, that cancer will be linked to a high-income region, even though everyone who has been affected is of low-income status.
Race also plays an issue. "You have to be comparing poor white regions with poor white regions, or you have another bias," Brawley warns. Access and treatment can be affected by prejudice and by cultural norms that may influence when a patient goes in for treatment or how he or she accepts the advice.
Gorey's argument that SES does not play a significant role in cancer survival in Canada is challenged by data from another study. Using the same cancer registry in Ontario, William Mackillop, M.D., of the Radiation Oncology Research Unit of Kingston General Hospital in Kingston, Ontario, found that across the province of Ontario, SES was linked to cancer survival, much as it is in the United States.
Mackillop's findings were that across the province, residents of low SES regions routinely survived a shorter time after a cancer diagnosis than residents of high SES regions. Mackillop concluded that despite Canada's attempt to provide a system that reduces the inequities in health caused by SES, the system does not in fact prevent a difference in cancer survival between rich and poor communities.
A Different Focus
Gorey said that the difference between the findings can be attributed to the different focus of the two studies. Gorey's study is confined to metropolitan Toronto, while Mackillop's covers the entire province. As a result, the size of the communities the census data groups together to assign an economic status to are vastly different. In some places in the fringes of the province, a single population group might range over hundreds of square kilometers.
An editorial accompanying Gorey's paper in the October, 1998 issue of Cancer Prevention and Control argued that there may be a significant regional effect. In Toronto, even if a person lives in a poor neighborhood, you are in easy travel distance of numerous high-quality hospitals. Out in the rural parts of Ontario the distances to a comparable high-quality treatment center can be enormous. It is the region's resources that produce the different result, the editorial argued. With one out of every three residents of Ontario living in the greater metropolitan Toronto area (one out of every seven Canadian residents), the pattern seen in Toronto is important.
Gorey also pointed to his study showing that while survival is unaffected by SES in Toronto, incidence is affected. Lower SES regions develop more cancers to begin with, due probably to a less healthy lifestyle.
-- - Laurent Castellucci
Economically Disadvantaged May Survive Cancer Longer in Canada
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||