Blog: Who's in your ACO?

Hundreds of hospitals and medical groups have entered into contracts with insurers (notably Medicare) to deliver savings and quality gains. In return, hospitals and doctors may get a share of whatever they save. The contracts, called accountable care organizations, are popular but experimental and not without uncertainty.

One source of confusion is how best to determine which patients are included under the contract for the purposes of tallying up the savings and measuring the quality. Get it wrong and it could skew savings and quality results.

Now, researchers at Dartmouth University have crunched Medicare data to test two basic strategies for identifying patients. One picks out patients at the start of the contract and the other at the end of each year. The results were published in Health Affairs.

Hospitals and doctors like knowing ahead of time which patients fall under the contract, which is designed to include a specific set of patients for whom they primarily provide care. That potentially gives providers greater control over the cost and quality of patients' care.

But this turns out to be a little sloppy. Using Medicare data, the researchers found that 17% of patients identified before the contract got under way did not visit hospitals or doctors in the accountable care organization. Results were worse for primary-care doctors; nearly one-third of patients on the list did not see any of the ACOs' primary-care doctors.

As the authors point out, that leaves “organizations responsible for both the cost and quality of care of patients for whom they had little or no impact.”

That's not the case when patients are identified at the end of each year based on the care that they received. All patients received care from the accountable care organization, and 81% saw primary-care doctors in the ACO.

Researchers used Medicare data for 2008 and 2009 (but excluded patients in Medicare managed-care plans) to create mock accountable care organizations with at least 5,000 patients.

Knowing which patients fall under the contract before it starts would allow hospitals and doctors to tailor quality improvement and cost-saving measures to those patients. That could improve the likelihood that providers earn financial bonuses to reinvest, but could also lead to “two standards of care” should providers target only patients under the contract, researchers wrote.

The authors conclude retrospective, or after-the-fact, identification “may more fully and accurately reflect an ACO's patient population and may better position an ACO to achieved shared savings.”

You can follow Melanie Evans on Twitter: @MHmevans.