By Melinda Young
Nearly a decade ago, Maryland experimented with a global budget payment model for rural hospitals. The plan was to give them a set amount of money, called Total Patient Revenue (TPR), to improve their efficiency.
But it did not quite work out.
“The hope was, and it was maybe naïve, that rural hospitals would concentrate on efficiency and keeping people out of the hospital,” says Bernard Black, MS, JD, professor of law and finance at Northwestern University at Evanston and Chicago.
The hospitals were told that if they provided more services, they would not receive additional reimbursement. If they performed fewer services, up to a point, they were rewarded for their efficiency, he explains.
“If they reduced costs by 5%, then they got an automatic price increase the next year,” Black says. “Did they have incentives to concentrate on efficiency? Sure.”
‘Be Really Careful With Incentives’
ED visits dropped 12%, and non-ED admissions declined 23%.1 But there was little incentive for the hospitals to collaborate with community providers to improve patients’ health. Instead, the hospitals just reduced overall services.
“You can’t just shut your emergency department,” he says. “But you can make it less convenient, not well-staffed, and that’s consistent with what we found — the emergency department visits dropped by a little bit.”
Hospitals have control over other services, including scheduled surgeries, direct hospital admissions, outpatient hospital clinic visits, and procedures in hospital-owned outpatient surgery settings. Researchers found that these services declined rapidly over the first few years of TPR’s implementation. Ambulatory surgery center visits declined 45%. Outpatient clinic visits fell 40%.1
The study also showed that Medicare spending increased overall in these regions, which suggested that people were discouraged from seeking services at the TPR hospitals, but sought healthcare services in non-TPR settings.1
“The message here is you have to be really careful with your incentives,” Black says. “A well-meaning, but blunderbuss, regulatory agency thought, ‘If we just give them incentives to be more efficient, they’ll be more efficient, and of course they wouldn’t turn patients away.’”
Yet, the TPR hospitals did turn patients away, and those patients still used Medicare dollars to receive help from other providers. “The big picture here is that people were not deprived of healthcare, and we did not reduce the need for healthcare — there’s no evidence of that,” Black explains. “The care was moving outside the capitation area of rural hospitals.”
Maryland’s quality measures, which could have served as a check and balance on the capitation model, only measured what the hospital was doing for patients inside the hospital. It did not look at how the patients were performing overall in their health and in other health settings, Black adds. This type of economic pressure can be anticipated when the payment model is poorly designed.
“Say the hospital is capitated, and everything else is fee-for-service,” Black says. “There’s incentive to push things out from under capitation to the fee-for-service world.”
Accountable care organizations (ACOs) operate differently because they cover their populations’ care in all settings. “ACOs are a hard way to make money on the provider side,” Black says. “If you have all of those patients, the tradeoff is that if you provide them with less care, you will make more money today, but get a bad reputation and have fewer patients tomorrow.”
Hospitals that are part of an ACO have a strong incentive to provide efficient and high-quality care, he adds.
The study’s takeaway message is that alternative payment models should not leave holes. “They have to be full capitation or be really smart and clever and on top of the incentives,” Black says. “If you’re fighting people’s incentives, the incentives will win.”
- Pines JM, Vats S, Zocchi M, Black B. Maryland’s experiment with capitated payments for rural hospitals: Large reductions in hospital-based care. Health Aff (Millwood) 2019;38:594-603.