IMGCAP(1)]
(Reuters) NEW YORK | Thu May 19, 2011 8:56am EDT - Looking for hospital ratings on the Internet or in magazines may not serve patients' best interests, with so-called "50 Best Hospitals" not necessarily any better than other medical institutions, a
The study, published in the Archives of Surgery, looked at Internet-based HealthGrades and U.S. News & World Report, both of whose ratings it said are used by millions.
It showed that for three types of cancer surgery, "America's 50 Best Hospitals" as ranked by HealthGrades are no better than other hospitals once the number of patients they treat are taken into account.
Based on the U.S. News & World Report ratings, the top 50 hospitals trumped the rest on just one of the surgeries -- removal of the colon or part of it.
"Our findings show that both ratings fail to identify equally well-performing hospitals of similar volume," wrote Nicholas Osborne, of the University of Michigan in Ann Arbor, and colleagues.
"Because these ratings list only a selected number of best hospitals, these consumer aids may dissuade patients from seeking care at closer, high-volume, equal-quality hospitals."
The study examined how many patients died within 30 days of cancer surgery, comparing the top-50 hospitals from the two ratings to other hospitals across the nation.
The researchers made sure to adjust for differences between patients, including income and sickness. Their analysis is based on Medicare data, limiting it to people over 65, and they didn't look at other important outcomes, such as complications and cost.
At first glance, U.S. News & World Report's best hospitals led the pack on all three surgeries, cutting the risk of death by between 31% and 58%.
But when the researchers adjusted the numbers for how many patients were treated, something the ratings don't consider, the top-ranking hospitals had better results only in colon surgery.
HealthGrades' top-50 hospitals did better on just one type of surgery to begin with, but that advantage disappeared completely after taking patient volume into account.
HealthGrades, which bills itself as the country's "leading health care ratings organization," did not respond to an e-mailed request for comments.
In an editorial, David Linehan and David Jaques, both surgeons at Washington University School of Medicine in St. Louis, said the findings show the shortcomings of commercially available hospital ratings.
They gave an example of a colleague asking for a recommendation for a relative, which they were able to respond to since they knew where the patient lived and had first-hand knowledge of the hospital and surgeon without resorting to other means.
"That insider information is priceless and not easily available to the general public," they wrote, adding that current commercially-available ratings are "imperfect."
© 2010 Thomson Reuters.
