Updated

Some policymakers have suggested using the number of patients that form blood clots after surgery as a measure of a hospital's quality. But a new study questions that idea.

Researchers found high-quality hospitals and those that regularly check for the complications tend to have higher rates of blood clots than low-quality hospitals and those that don't look for clots as often.

"Usually, if you have a higher complication rate, you're not doing a good job, but in this case it's actually the opposite," Dr. Karl Bilimoria, the study's lead author from Northwestern University and Northwestern Memorial Hospital in Chicago, said.

Because hospitals and doctors that look for the blood clots tend to find more of them, a hospital's clot numbers may give a false impression of its quality, the researchers write in The Journal of the American Medical Association (JAMA).

Blood clots can form after surgery because the body is in a "clotting mindset" trying to stop all bleeding, Bilimoria said.

"Blood clots after surgery are an important complication," he said. "They can have severe consequences, and it can be fatal if the blood clot travels to the lung."

The blood clots - known as venous thromboembolisms or VTEs - are considered preventable if hospitals spot at-risk patients and put them on medications to stop blood from clumping together.

For the new study, Bilimoria and his colleagues examined data from 2,786 hospitals and 954,926 patients who were discharged after surgery between 2009 and 2010. All the patients were on Medicare, the government-run insurance program for the elderly and disabled.

In one analysis, the researchers compared hospitals they considered to be high quality with those they dubbed low quality. High-quality facilities, for example, were those that did a lot of surgeries and organ transplants and were teaching hospitals.

High-quality hospitals were more likely to comply with blood clot prevention methods, they found, but still had a higher proportion of surgery patients with clots.

Hospitals considered to be high quality reported about 6.4 blood clots for every 1,000 surgery patients, compared to 4.8 clots per 1,000 patients at low-quality hospitals.

What's more, hospitals that did the most tests looking for blood clots found and reported more than twice as many clots as hospitals that did the fewest tests.

In an editorial accompanying the study, Dr. Edward Livingston, a deputy editor at JAMA, said the results demonstrate that the blood clot rates reported on government websites "reflect how aggressively clinicians look for (blood clots) but probably are not directly related to quality of care."

Due to the U.S. government shutdown, the Agency for Healthcare Research and Quality, the government organization that first proposed using blood clot rates to judge hospital quality, did not respond to a request to comment.

"I wouldn't use that metric at all when trying to decide where to go for surgery," Bilimoria said. "If anything, you might want to use it in the other direction. That's sort of a suggestion from the data."

He suggested a better measure of quality would be how well hospitals follow the guidelines to prevent blood clots.