"It suggests kind of a reverse disparity," said M. Kit Delgado, lead author of the new study and an emergency physician at the Hospital of the University of Pennsylvania, which is one of several top-level trauma centers in the Philadelphia region.
Some small, local studies in the past that examined what is known in the hospital industry as a "Wallet Biopsy" found a similar tendency to admit more insured patients, presumably helping the hospitals' bottom line. The study, published Wednesday in JAMA Surgery, was the first to use a new national database of emergency-room visits.
It found that patients with insurance were between 10 percent and 14 percent more likely than uninsured patients to be admitted rather than transferred.
"I don't think there is anything malicious going on," Delgado said. "Physicians are making decisions for the most part based on what they think would be best for the patient." The disparity likely shows up in cases that would be close calls, he said.
"There are times that if the patient doesn't have insurance, you might get a tap on the shoulder from a case manager in the emergency department," said Delgado, who said he had experienced such taps in the past when moonlighting at other hospitals.
Community hospitals, which clearly lack the resources to treat complex cases with severe injuries, were the least likely to hold on to those patients.
The greatest disparity - the most likely to admit insured patients rather than transferring them - was at large, metro-area hospitals that have trauma teams and specialists such as neurosurgeons available. They may have the expertise to care for these cases, Delgado said, but don't have in place some of the underlying protocols, such as the ability to immediately take patients to a trauma bay with a team in place.
A commentary with the study disagreed with its conclusions.
"Do financial factors have a role in how patients are treated and transferred about? Sure they do," wrote Charles D. Mabry, a surgeon at the University of Arkansas. "But, I believe that there were other factors in this study that had greater influence on where trauma patients were treated," he wrote, citing the type of injury and the capability of the admitting hospital.
The authors of the main study agreed that those factors played a role. Insured patients also may be more reluctant to go to trauma centers that often are big public hospitals.
The Penn team is doing three follow-up studies to better analyze its findings.
If the disparities persist, the authors said, the health system may need to consider "shared reimbursement schemes" that would pay both the initial hospital and the trauma center, lessening any incentive in the decision.
Patients may also need more education about the benefits of trauma centers because, Delgado said, "what we are finding is insured patients are at risk for worse outcomes."