The surgeons who replace more than a million knees and hips each year in the United States have a reputation for embracing every new implant. In the 1990s, for example, many abandoned the cement that holds an artificial hip in the bony socket in favor of a steel ball that articulated in a polyethylene cup. But after a few years, the polyethylene broke down, requiring a second surgery. Swedish surgeons, on the other hand, stuck with the cement and perfected the procedure.

It’s not that Swedish surgeons shy away from new technology. Rather, what they have—and U.S. surgeons lack—is a national registry that tracks the outcome of nearly all hip and knee implant surgeries. That is why the rate of revision surgery to repair a defective hip implant is 6% in Sweden compared with 17% in the United States, claims Henrik Malchau, an orthopedic surgeon at Massachusetts General Hospital who ran the Swedish registry from 1982 to 1992.

The need for a U.S. registry is urgent. In 2006 there were 77,000 hip and knee revisions in the United States at a cost of more than $3.2 billion. Even reducing the revision rate by 2% would yield an annual savings of $652 million, according to the American Association of Orthopedic Surgeons.

Fifteen countries have registries, but in the United States only a handful of local registries exist—one in Virginia created by the state’s orthopedic surgeons and a few run by such large health systems as the Mayo Clinic and Kaiser Permanente. That is about to change, though it’s not clear how. In June, New Jersey Democrat William Pascrell Jr. and Texas Democrat Lloyd Doggett introduced a bill in the House of Representatives that would create a government-run registry. But a month later, the AAOS incorporated its American Joint Replacement Registry and says it will begin collecting data in the next 12 to 18 months.

Several times last year, Pascrell’s staff met with the AAOS to discuss a public-private partnership. At the time, the AAOS didn’t have much to show for its 10 years of work to launch a registry, says Pascrell’s chief of staff, Ben Rich, and “we worried about financial conflicts of interest that might arise if surgeons ran the registry.” (Rich was referring to a $311 million settlement reached with four implant manufacturers in 2007 after the government claimed they had paid surgeons tens to hundreds of thousands of dollars each year to use their implants.)

Although the AAOS’s registry would welcome government funding—the surgeons have collected only half a million dollars of the registry’s estimated $25 million in start-up costs and $5 million to $10 million in annual operating expenses—surgeons aren’t likely to submit data to a government-run registry, says the AAOS. For starters, the congressmen’s bill has no provision to protect surgeons and device manufacturers from plaintiffs’ attorneys mining the registry.

Nor will surgeons buy into a registry if they aren’t in control of gathering and interpreting the data. “Surgeons are the best equipped to gather the most accurate information,” says Daniel J. Berry, chairman of the department of orthopedic surgery at the Mayo Clinic and second vice president of the AAOS. Nearly all foreign registries are run by orthopedic surgeons and funded by governments, notes Malchau, who is helping develop the AAOS registry.

Orthopedic surgeons would be heavily involved in a public registry, insists Rich. And whether the surgeons like it or not, other government-run registries are being proposed, he notes. “Maybe our legislation will spur the AAOS along,” says Rich. “If so, we’ll reassess the need for a public registry.”

The surgeons have a decade of preparation on their side, counters Berry. “It is one thing to create a registry in a bill; the challenge is to create a real-life registry that helps all parties, including government agencies, improve patient care.”