And physicians have been buying them up. According to the study, between 2000 and 2007, there was a 50 percent increase in the number of Medicare-certified surgicenters, with doctors having a stake in 83 percent of them, and owning outright nearly half.

Hollingsworth says this rise in ownership tracks nicely with the declining real dollar income of doctors.


"If you're trying to maintain your level of income, you have a couple of options," he says. "See more patients, but another source of revenue would be nontraditional: investing in other technologies or other financial arrangements."

Caveats

Still, it's important to note that the current study doesn't say that any of the surgeons were performing unnecessary procedures. Because of the study limitations, the researchers weren't able to look at the actual appropriateness of the procedures ordered, all of which could have been medically required.

Plus, the researchers had to rely on physician identification numbers scrubbed clean of any real identifying data. The researchers therefore couldn't tell if a physician was actually performing on self-referred patients, although Hollingsworth says that aside from some of the GI procedures, most surgeries would have been done by the physician who ordered it.

More seriously, the researchers had to use a proxy for actual ownership. As the physician data they had was anonymous, they couldn't match doctors up with their practices. And even if they could, it might be hard to know if the physician was the owner without serious sleuthing, as current law doesn't require doctors to disclose interests in their outpatient centers, says Hollingsworth.

Instead, Hollingsworth and his colleagues dubbed as "owner" any physician who performed more than 30 percent of the surgeries at the surgicenter.

"We know from federal Safe Harbor [laws] that in order to not be in violation of the federal statute, they need to perform a minimum of one-third [of their surgeries at the center]," he says. "So we empirically defined an owner and who wasn't based on who concentrated their cases at a given surgical center."

Hollingsworth admits there's a possibility for error, which is why they also tested the data using more stringent cut-offs: doctors who performed at least 60 percent of their surgeries at one center were "owners," and those who performed less than 20 percent were "non-owners." And even with this, they were able to find good agreement with the data, Hollingsworth says.

Future studies

But Hollingsworth hopes to clear up these questions with future studies. He says he would like to look at a procedure -- such as surgery for breast cancer -- where there is less wiggle room for ordering, and compare its use by a physician-owner against an elective procedure.

More Industry Headlines