Reuters logo
Docs may cherry-pick cases after death rate reports
June 20, 2013 / 8:44 PM / 4 years ago

Docs may cherry-pick cases after death rate reports

NEW YORK (Reuters Health) - Doctors may avoid treating the sickest heart patients with a common procedure after their hospitals are marked as having high death rates, says a new study that points to a possible unintended consequence of transparency.

Researchers found the severity of cases treated at four Massachusetts hospitals was lower after they were labeled “outliers” for having high death rates after stenting between 2003 and 2010, compared to their counterparts with lower death rates.

“It really does highlight the idea that all the good that has come from public reporting and process improvement, we’ve traded that for another set of problems,” said Dr. Duane Pinto, who has written about public reporting but wasn’t involved in the new study.

During stenting, a balloon-tipped catheter is threaded through blood vessels in the wrist or thigh toward the heart. The balloon is inflated to clear narrowed vessels, then, a small mesh tube - the stent - is used to prop each vessel open.

Proponents of public reporting hope releasing death rates after certain procedures would give the public more information on the healthcare it’s receiving and encourage hospitals to adopt best practices.

Recent reports, however, have suggested that the data kept on deaths after stenting are unreliable and doctors end up avoiding the sickest patients (see Reuters Health articles of October 10, 2012 and May 10, 2013 here: and

For the new study, Dr. James McCabe, the study’s lead author from Brigham and Women’s Hospital in Boston, and colleagues used the data hospitals entered into their computers about patients before their surgery to estimate how many would likely die.

Higher expected death rates, according to the researchers, are a marker for sicker patients because they’re less likely to survive - even with stenting.

After analyzing data on 116,227 stenting patients at Massachusetts hospitals between 2003 and 2010, McCabe and colleagues found the hospitals that weren’t labeled as “outliers” had an expected death rate of about 1.6 percent.

That compared to about a 1.1 percent expected death rate among the four hospitals after they were labeled “outliers.” The difference implies that the patients being sent for stenting at “outliers” were not as sick as those at other hospitals.

“I think the unintended consequence here is if too much attention is given to this outlier status what happens is there could clearly be a high-risk patient who would benefit from having a high-risk procedure (who doesn’t get it),” said Dr. Gregory J. Dehmer, a professor of medicine at Texas A&M Health Science Center College of Medicine in Temple.


McCabe cautions that the study can’t prove doctors avoided stenting the sickest patients. It could be, for example, that those hospitals received healthier patients after they were identified as an “outlier.”

“We believe the patient populations presented to these outlier hospitals were the same before and after, but that’s something that can be debated,” he said.

Despite the study’s limitations, Dehmer writes in a commentary accompanying the new report in JACC: Cardiovascular Interventions that he believes the results.

“It’s completely understandable - the behavior of the physicians and the hospitals if they’re named an outlier,” Dehmer told Reuters Health.

Dehmer said the laws should be crafted with the help of doctors working in the field.

“What the American College of Cardiology and some of the other professional organizations are trying to do is cautiously move into this space and improve the quality of information available to the public,” he said.

Pinto, director of the cardiac intensive care unit at Boston’s Beth Israel Deaconess Medical Center, agreed.

“Mortality rates alone don’t tell the whole story. To understand the nuance of behavior and downstream consequences, you have to talk to the people who handle the patients,” he said.

SOURCE: JACC: Cardiovascular Interventions, online June 2013.

Our Standards:The Thomson Reuters Trust Principles.
0 : 0
  • narrow-browser-and-phone
  • medium-browser-and-portrait-tablet
  • landscape-tablet
  • medium-wide-browser
  • wide-browser-and-larger
  • medium-browser-and-landscape-tablet
  • medium-wide-browser-and-larger
  • above-phone
  • portrait-tablet-and-above
  • above-portrait-tablet
  • landscape-tablet-and-above
  • landscape-tablet-and-medium-wide-browser
  • portrait-tablet-and-below
  • landscape-tablet-and-below