Special Report: Can this payment model work?

November 17, 2006

A new reimbursement method based on case rates seeks to go where P4P hasn't.

Tired of seeing your fee-for-service payments cut? Don't like capitation? Leery of pay for performance? Help may be on the way. A new reimbursement model called Prometheus Payment promises to pay doctors "evidence-based case rates" that reward them for efficiency and for following clinical guidelines based on the best scientific evidence.

Led by attorney Alice Gosfield, former chair of the National Committee for Quality Assurance, and by Francois de Brantes, national coordinator of Bridges to Excellence (an employer-funded P4P program), a group of health policy experts, healthcare leaders, employer and insurer representatives, and practicing physicians is developing the Prometheus model. Their goal, says Gosfield, "is to design a payment system that pays the right amount for the right science."

While she and de Brantes freely admit that their nascent model might not work, they and their colleagues have formed a nonprofit organization to move it forward, and they're planning to field pilots in up to four markets next year. They say they've already attracted interest from health plans and employer coalitions-which isn't surprising, considering the caliber of those involved.

Under the Prometheus model, physicians would be paid on the basis of negotiated case rates that would cover all necessary care in a particular case. They'd get 80 to 90 percent of the case rates in monthly payments. Based on their quality and efficiency scores on public report cards, physicians would receive pro rata payments from the pool of withheld money. Exemplary providers could earn bonuses above their negotiated rates.

Physician practices and hospitals would be able to form ad hoc associations to apply for case rates, and each provider would be paid for the portion of the practice guideline that it promised to deliver. But, although they'd cooperate clinically, their businesses could remain separate. To persuade unrelated providers to work together effectively, 30 percent of a provider's scores would depend on whether other team members also delivered high-quality, efficient care.

To make sure that providers don't avoid sicker patients, the case rates would be adjusted for severity, and payments would be higher for more-complex cases. Primary care physicians would also have a reason not to refer sicker patients to specialists right away, because they'd get a lower efficiency score if they did. Part of a doctor's efficiency rating would also hinge on how the patient fared after leaving the office: For example, did a patient with asthma stay out of the ED or the hospital?

The mix of incentives is designed to give physicians maximum flexibility to manage care as they see fit without motivating them to overuse resources, as fee for service does, or limit services, as capitation does.

Prometheus' "evidence-based case rates" (ECRs) also differ from capitation in that they represent payments for specific cases rather than a single monthly capitation rate for all of a patient's care or all of their primary care. And unlike capitation contracts with HMOs, emphasize Gosfield and de Brantes, Prometheus leaves the insurance risk with the health plans. If a case turns into a train wreck, the plan is on the hook, not the provider.

What happens to doctors who don't make the grade?

Participating physicians won't lose anything if they don't get any quality or efficiency rewards, says Meredith Rosenthal, an associate professor at Harvard School of Public Health and a Prometheus board member. That's because if they don't deliver some of the services that a guideline calls for, they'll spend less time on the case and have lower practice expenses, she says. Of course, for poor performers to receive as much as they would under fee for service, the case rate would have to be high enough to cover all of the prescribed services at the going market price.

The first pass at setting price ranges for the ECRs will entail matching the services specified in well-accepted guidelines to the costs of those services in a large, national claims database. The resultant case rates will then be adjusted upward to reflect what Gosfield calls "normal clinical variations." The initial ECRs will encompass some types of cancer, cardiac, and orthopedic care, as well as care for chronic diseases such as diabetes and depression.

A major unresolved question is how health plans, who'd administer the ECRs, would negotiate each piece of a case rate with the physician, group practice, or hospital system that bid on it. Gosfield merely says that a payer would have to "construct its pricing the way we say it has to be constructed to claim they're doing Prometheus Payment." She adds that plans would benefit from having fixed pricing and decreased administrative costs under the Prometheus model.

Putting providers together isn't going to be simple

Outside observers say that the biggest obstacle to Prometheus is the fragmentation of the healthcare delivery system. "Someone would have to be in charge of a care team that would integrate all the specialists and all the facilities," notes Uwe E. Reinhardt, a health economist at Princeton.

De Brantes acknowledges that. While the health plan would divide up the case rate among providers, he says, a "principal physician"-either a primary care physician or a specialist-would lead the clinical team. Rosenthal elaborates: "For this to work well, patients would need some sort of medical home-a provider who's taking on their care for one of these conditions-and they'd rely on that provider to take them through the system."

Rosenthal and de Brantes admit that care teams of the kind they're talking about are much easier to find in integrated delivery systems and large groups than in typical medical communities that are dominated by small practices and unaffiliated hospitals. But de Brantes says that Prometheus must find a way to work with small, independent practices to create "virtual" teams.

Reinhardt also wonders how many kinds of cases lend themselves to the Prometheus approach. "If doctors got together to do this, they'd have to specialize in a lot of cases that are very standard," he says. "Otherwise, imagine all the evening meetings they'd have to have on how to coordinate this care and split up the spoils." Factoring in the logistical difficulties and the lack of good evidence in many clinical areas, he estimates that no more than 15 to 20 percent of healthcare could be covered under ECRs.

Dennis Scanlon, a health policy professor at Penn State University, also views the Prometheus model with some skepticism. "It's challenging to get multiple healthcare providers to work together, because there are strong incentives for one party to not play along. Also, the commitment they have to make, and the costs of that commitment relative to the benefits, are uncertain."