Coronary bypass patients have greater chance of dying in Ontario than most U.S. states
Author: Michael Walker, Maureen Hazel and Nadeem Esmail Date: 11 February 2009

Ontario patients undergoing coronary artery bypass graft surgery could have reduced their risk of dying by having it performed in any one of 27 U.S. states rather than Ontario, according to a new peer-reviewed study from independent research organization the Fraser Institute.

The average risk of dying in hospital for patients undergoing coronary artery bypass graft surgery (CABG) in Ontario in 2004 was almost twice as high as Minnesota and Massachusetts and considerably higher than Colorado, Michigan, Maryland and Arizona, says the study, A Comparative Analysis of Mortality Rates Associated with Coronary Artery Bypass Graft (CABG) Surgery in Ontario and Select U.S. States. The complete report is available at www.fraserinstitute.org

“For a patient choosing where to have surgery, the main objective is to minimize risk. Minimizing risk means avoiding hospitals and jurisdictions that have high rates of risk-adjusted mortality. Since the estimates that are calculated have a range, avoiding risk means choosing the hospital or jurisdiction that has the lowest maximum probable mortality rate,” said Nadeem Esmail, Fraser Institute director of health system performance studies and study co-author.

An Ontario patient who had the surgery performed at an average Minnesota or Massachusetts hospital would have reduced their probable upper limit risk of mortality by about 41 per cent. Among the 32 US states whose mortality rates were compared to Ontario’s, only Vermont, Arkansas, New Hampshire, Utah, and Oregon had higher upper limit risk-adjusted mortality rates than Ontario in 2004.

“By simply crossing the border to either Michigan or New York, an Ontario patient could have reduced their probable upper limit mortality rate by 39 or 36 per cent,” Esmail said.

The coronary artery bypass graft (CABG) surgery mortality rate is a widely used health outcome measure since the surgery is performed in high numbers, requires complex surgical and perioperative care, and has easily measurable rates of adverse events. The methodology used for calculating bypass surgery mortality rates in a standard way was developed by the US Agency for Healthcare Research and Quality (AHRQ) with Stanford University. This measure has been shown to reflect the quality of care in hospitals where better processes of care may lead to lower mortality rates.

The study also examined average risk-adjusted mortality rates, and found that 20 of the 32 U.S. states for which data were available had statistically significant lower average risk-adjusted mortality rates than Ontario. This was also true for the United States as a whole. Only Arkansas had a statistically significant higher average mortality rate following CABG surgery than Ontario in 2004.

“Just as troubling as Ontario’s relative performance to that of the majority of U.S. states in 2004 is the fact that the gap between the performance of hospitals in Ontario and of those in many U.S. jurisdictions widened markedly between 2003 and 2004. For example, Minnesota’s average mortality rate as a proportion of Ontario’s went from 69 per cent to 52 per cent between 2003 and 2004. The average for the United States fell from 89 per cent of Ontario’s risk-adjusted average mortality rate to 71 per cent between 2003 and 2004,” Esmail added.

The study, which used publicly available outcomes data from the peer-reviewed Fraser Institute Report Card on Ontario Hospitals, also found that the risk of death for patients undergoing coronary artery bypass graft surgery varied widely from hospital to hospital within Ontario . The variation was large enough to make a material difference in the likelihood of mortality for a patient of a given risk-adjusted health status.

From 2002/03 to 2004/05, patients experienced the lowest risk of dying by having coronary artery bypass graft surgery at the University of Ottawa Heart Institute.

On the other hand, patients undergoing coronary artery bypass graft surgery had the greatest chance of dying if the procedure was done at Sunnybrook and Women’s College Health Sciences Centre in Toronto or Anonymous Hospital 104 (not all hospitals agreed to be named in The Fraser Institute’s Hospital Report Card).

The University of Ottawa Heart Institute, which had a maximum probable risk-adjusted mortality rate of 2.82 per cent over the three-year period, was overall the least risky hospital for CABG surgery. Hospital 10, St. Mary’s General, and Hospital 50, which had risk-adjusted mortality rates of 3.51 per cent, 3.61 per cent and 3.65 per cent, respectively, were also less risky. Sunnybrook and Women’s College Health Sciences Centre and Hospital 104, with probable mortality rates as high as 6.23 per cent and 7.36 per cent respectively, were the most risky.

“While we can’t explain why hospitals have different mortality rates, it is clear that changing hospitals or jurisdictions can have a significant impact on the likelihood of whether a patient lives or dies,” Esmail said.

“Though it may be possible for a patient to have a less risky procedure at a higher-risk institution due to the variance in mortality rates among surgeons, it is nevertheless true that an average patient of average risk would be better off in some Ontario hospitals than others, and would be better off in the majority of U.S. states than they would be in Ontario.” 

Authors of the study: Michael Walker, Maureen Hazel and Nadeem Esmail of the Fraser Institute. 

For more information please contact:

Eustace Davie, Director

Tel: 011 884 0270, Fax: 011 884 5672, Email: hpu@mweb.co.za

Website: www.healthpolicyunit.org