There are hundreds, if not thousands, of ways to track the health of a population: the average blood pressure, the rate of mental illness and the average weight.
Epidemiologists have been collecting these data for years, but now, in Oregon, there is cold, hard cash riding on measurements like these.
The idea is to get doctors, nurses, hospitals and other caregivers to work together and get paid well to keep people healthy and to get rid of wasteful, unnecessary care.
But which measures should Oregon and the federal government look at to decide if the experiment succeeds? The federal dollars will go to the providers who do a good job, so first the state had to figure out what constitutes good coordinated care.
Sarah Bartelmann of the Oregon Health Authority's metrics and scoring committee says it was hard to boil down the list. "A lot of the discussion and the conflict was where the data would come from and if it was something the state could easily report on," she says.
Oregon decided on 33 measurements that cross the medical spectrum.
A look at one of them shows how quickly it can get complicated. Bartelmann explains the effort to measure how often doctors ask patients if they're abusing drugs or alcohol.
"So there are a couple of ways a practice could do this," she says. "They could start with a prescreen, that would be just one question that a doctor would ask. It could be part of your vital stats, when they're taking your blood pressure, weighing you in. They could also ask, 'Do you use alcohol? Do you use drugs?' And if the patient indicated that they did, to either one of those, that would then trigger a longer screening."
The goal is to get doctors to refer patients to a service patients don't actually have to go to be counted, but they have to be referred. But it's a measure that Oregon has created on its own, so there was no national standard.
To figure out where to start, Oregon Health and Science University studied how often doctors currently ask patients about drug or alcohol abuse not just whether they smoke or drink, but specifically about abuse of those substances. The answer was virtually never.
"This is a difficult subject for doctors to talk about with patients," says Lori Coyner of the Oregon Health Authority.
So, since doctors aren't asking the question now, the baseline was set at zero. The state then had to decide where to set the goal.
The university looked at one group of local doctors who started asking about substance abuse and found that after a couple of years, 44 percent had managed to set up a system to do it on a regular basis. So that's where Oregon set its goal.
"The committee wrestled with setting a benchmark that seemed so high 44 percent compared to approximately zero percent right now," says Bartelmann.
Not everyone is happy with the new measurement.
Take, for instance, Rick Hangartner, who runs an electronic health records company. He says it doesn't go far enough to help people battling drugs or alcohol.
"The big problem is finding a treatment program that can actually take them that works for them," Hangartner says. "If the whole point is improving health, we've got to show some follow-through."
The state is wrestling with questions like this for all 33 of its chosen metrics.
On top of that, Oregon has to worry about timing. For instance, rehab for drug abuse or alcoholism could save the state a lot of money in the long run by preventing diseases. But in the short term, rehab is expensive.
The federal government has given Oregon just five years to prove it can reduce increases in medical costs. If the state fails, it faces some substantial fines.
This story is part of a reporting partnership between NPR, Oregon Public Broadcasting and Kaiser Health News.