Giving health-care providers a lump sum payment for certain treatments – touted as a way to save money and improve coordination of care — yielded disappointing results for some major California hospitals and insurers, a study found.

The RAND Corp. study, funded by a $2.9-million federal grant,  looked at “bundled payments” for care of insured orthopedic patients under 65 at a handful of large hospitals and insurers in California.

Six of the state’s biggest insurers and eight hospitals started out in a pilot program in 2010, but only three insurers and two hospitals actually decided to enter contracts to adopt bundled payments. The others dropped out because they didn’t think bundled care, such procedures as total knee replacement surgery, would change the delivery of care significantly or lower costs, according to the study, published in the journal Health Affairs Monday.

The pilot project resulted in such a small number of hospital cases that it was hard to draw conclusions about how bundled payments affect health care quality or costs, which were the initial goals of the study, the researchers reported. Two ambulatory surgery centers managed to partner with an insurance company and had a higher volume of cases, but generally health plans have been slow to contract with them, the study found.

“That was unexpected,” said Susan Ridgely, the lead author of the study and a senior policy analyst at RAND, a Santa Monica, Calif.-based think tank. “They were a bit more flexible and also wanted the business, but hospitals began to see that it required too much time and effort or maybe that it was not in their best interest.”

Under bundled payments, doctors, hospitals and other health providers share a fixed payment that covers the average cost of a “bundle” of services – caring for a person with a hip replacement, for instance.

Hospital and payer incentives were sometimes at odds. It’s in the best interest of doctors and hospitals, for example, to limit bundles to the lowest-risk patients. Also, to keep costs down and maximize profits, they want to shorten the period during which the bundled payment covers treatment. Insurers, on the other hand, want providers to treat a diverse population of patients for as long — and for as little money — as possible.

“They’re solvable problems…We’re still quite enthusiastic about this approach,”  Ridgely said.   “What we’re hoping to do is bring to the forefront some of the issues these organizations are going to have to grapple with.”

Proponents say the “bundled payments” approach gives health care organizations more autonomy over how they spend their money and deliver care, and spurs collaboration between primary care doctors, specialists, nurses and outpatient facilities. In theory, it could make healthcare more comprehensive and less expensive, proponents say. Some also say that the approach could bring increased transparency to the health care system because patients and payers would know the costs up-front.

The model has drawn support from health-care experts including  Ezekiel Emmanuel, who helped draft the Affordable Care Act. Medicare has been experimenting with bundled payments, as well.

But because bundles aren’t replacing all fee-for-service claims, skeptics say their overall effect will be small.

“At best, it’s going to affect a relatively small percentage of spending,” says Bob Berenson, a policy fellow at the Urban Institute. “It’s one of the issues I’ve always been concerned about — if it’s only for a small number of conditions, is it worth all the administrative difficulty?”

In the California study, the few health-care organizations that embraced bundled payments chose to process bills manually because the custom-made software necessary to process bundled claims cost more than $1 million, Ridgely said.

For bundled payments to pay off, a health-care provider has to treat a minimum number of patients. Below that threshold, the administrative costs could be too much of a burden, Ridgely said.