In 2006, the Centers for Medicare & Medicaid Services (CMS) said it would only pay for bariatric surgery done at hospitals that had certain equipment and medical teams in place and were certified by the American College of Surgeons or the American Society for Metabolic and Bariatric Surgery (ASMBS).
"About a decade ago, there were very real concerns that there were safety problems with bariatric surgery," said Dr. Justin Dimick, who led the new study at the University of Michigan in Ann Arbor.
At some hospitals, he said, as many as nine percent of patients died during or after surgery. But that started changing before CMS stepped in with new restrictions, according to Dimick, in part due to less-invasive surgical techniques and better surgeon training.
He and his colleagues used billing codes to track bariatric surgery complications, such as leaks and bleeding, during about 300,000 procedures done between 2004 and 2009 both at centers of excellence and at other hospitals.
They found the proportion of patients on Medicare who had any procedure-related complication dropped from 12 percent before the policy change to eight percent afterward, according to findings published Tuesday in the Journal of the American Medical Association.
In non-Medicare patients -- who had no restrictions on where they received surgery -- complication rates also dropped, from between six and seven percent to below five percent.
More complications are to be expected among Medicare patients, the researchers said, due to their age.
"Bariatric surgery got a lot safer over this time period but it happened in both Medicare and non-Medicare patients," Dimick told Reuters Health.
"The policy was implemented in 2006, and outcomes were getting better well before that," he added. "The evidence shows that the policy itself had no benefit." Dr. Jaime Ponce, president of ASMBS, disagreed with that assessment and said the new study was limited by its use of billing data instead of more detailed patient records.
"It's very difficult to say that accreditation has not helped hospitals or bariatric surgery programs," said Ponce, who was not involved in the new research. Previous studies showed the CMS policy was linked to a reduction in procedure costs and in deaths after surgery, he noted.
Ponce said the policy may have encouraged both hospitals that did and didn't end up being considered centers of excellence to improve their patient care.
"What (the study) showed is that all of the hospitals improved over time," he told Reuters Health. According to ASMBS, about 200,000 people have weight loss surgery every year. Surgery is typically recommended for people with a body mass index -- a measure of weight in relation to height - of at least 40, or at least 35 if they also have co-occurring health problems such as diabetes or severe sleep apnea.
Dimick and his colleagues are pushing for a move away from the current certification system, toward one that provides more feedback for all hospitals that perform bariatric surgery. Ponce, however, said the Medicare policy is helping patients and should remain in place.
A representative from CMS said the agency is examining its bariatric surgery coverage, including the certification requirement, "and can't comment on it until we issue a final decision." That report is expected later this year.
One drawback of the current policy, Dimick said, is that some people on Medicare can't be treated at nearby hospitals that perform bariatric surgery but aren't certified.
"The harms here are that Medicare patients who needed surgery might not have been able to have it, because it required travel somewhere else," he said.