The Arkansas Innovation

Ezekiel J. Emanuel

Ezekiel J. Emanuel on health policy and other topics.

Philadelphia

MENTION medical innovation, and you might think of the biotech corridor around Boston, or the profusion of companies developing wireless medical technologies in San Diego. But one of the most important hotbeds of new approaches to medicine is … you didn’t guess it: Arkansas.

The state has a vision for changing the way Arkansans pay for health care. It is moving toward ending “fee-for-service” payments, in which each procedure a patient undergoes for a single medical condition is billed separately. Instead, the costs of all the hospitalizations, office visits, tests and treatments will be rolled into one “episode-based” or “bundled” payment. “In three to five years,” John M. Selig, the head of Arkansas’s Department of Human Services, told me, “we aspire to have 90 to 95 percent of all our medical expenditures off fee-for-service.”

The change will encourage doctors and hospitals to work together to provide patients with the highest quality care, while at the same time lowering costs by eliminating unnecessary tests and treatments. It has been done before, in small-scale experimental pilot programs. But as the Arkansas officials make clear, this change will now be made in every corner of the state, for every hospital, and physicians in almost every specialty: surgeons, anesthesiologists, obstetricians, pediatricians, primary care physicians. For policy makers and the public, the Arkansas experiment is fascinating.

This is how it will work: Medicaid and private insurers will identify the doctor or hospital who is primarily responsible for the patient’s care — the “quarterback,” as Andrew Allison, the state’s Medicaid director, put it. The quarterback will be reimbursed for the total cost of an episode of care — a hip or knee replacement; treatment for an upper respiratory infection or congestive heart failure; or perinatal care (the baby’s delivery, as well as some care before and after).

The quarterbacks will also be responsible for the cost and quality of the services provided to their patients, and will receive quarterly reports on those metrics from the state (for Medicaid patients) or private insurers. If they have delivered good care based on agreed-upon standards, and if their billings come in lower than the agreed-upon level, they can keep a portion of the difference. If their billings come in above an acceptable level — usually because they have ordered too many unnecessary tests, office visits or inappropriate treatments — they will have to pay money back to the state or insurer.

Arkansas may seem an odd place for such a bold experiment. It has the sixth-highest poverty rate in the country, and ranks near the bottom in everything from the percentage of pregnant women getting prenatal care and the infant mortality rate to obesity, diabetes and life expectancy. It doesn’t have enough doctors; all but two of the state’s counties are designated as either entirely or partially medically underserved. And until recently, it was way behind on the adoption of electronic health records.

Yet Arkansas also has certain advantages. It has a governor who understands the issues very well. And it has doctors and hospitals who — faced with a State Legislature resistant to raising taxes, an imminent shortfall in state Medicaid funds and the threat of imposed managed care — agreed to support the scheme. Finally, it helps that Arkansas is a small state; when everyone knows everyone, it’s easier to work out implementation problems.

Still, it will be a challenge. Bundled payments for hip and knee replacements, which have similar costs for all patients, have been previously tested. But for other conditions, not every patient’s needs are the same. Some pregnant women are healthy while others have diabetes. The state and insurers will have to provide “risk adjustment” payments — in which providers are reimbursed more for treating sicker patients — and some patients with especially complicated illnesses may need to be excluded from the bundling system.

Even some low-cost conditions, like upper respiratory infections, are treated at widely varying costs, mainly because physicians prescribe different tests, numbers of office visits and medications (in 14 Arkansas counties, over 50 percent of patients with upper respiratory infections receive antibiotics, even though national guidelines say they should rarely be prescribed because most infections are viral).

But this is exactly what the new program will work to change, by providing standards for appropriate care linked to the costs of treatment and the quality of the doctor’s performance compared with that of other doctors.

Maybe Arkansas’s biggest challenge was getting the state’s insurers to work together. On that, it has succeeded. Arkansas’s two biggest private insurers, Blue Cross Blue Shield and QualChoice, are on board with Medicaid. But there is one big player missing: Medicare. To really make this innovation effective, the federal government should join in.

In the meantime, even as the state is working on implementing bundled treatments for a first round of medical conditions, it is gearing up with the second round. If Arkansas succeeds — even partly — it will show the way for the rest of the country.