In 2014, Maryland adopted an all-payer, global budget program for most of its hospitals, making a fixed payment to facilities for services provided in inpatient, outpatient and emergency departments. In its first two years, the concept contained costs, but didn’t appear to change utilization by patients.

In a JAMA Internal Medicine study led by University of Pittsburgh health policy professor Eric Roberts, PhD, researchers compared changes in hospital and primary care use between eight Maryland counties and 27 counties outside the state acting as the control group, using a 20 percent random sample of Medicare beneficiaries.

According to existing analyses by the Maryland Health Services Cost Review Commission, the all-payer model did slow growth in Medicare expenditures in 2014 (2.15 percent lower than national growth rate) and 2015 (2.22 percent lower), which saved a cumulative $251 million.

What Roberts and his coauthors found was on other metrics, the global budgeting didn’t seem to make an impact. It didn’t find consistent evidence the model reduced admissions, readmissions, observation stays, ED visits or outpatient use, or that it shifted patients to primary care. The same inconsistent results were found when comparing the Maryland counties to an alternative control group of Medicaid expansion states.

“Together, these findings provide no clear evidence that Maryland hospitals met their budgets by reducing hospital utilization or enhancing primary care beyond changes that would have been expected in the absence of global budgets,” Roberts and his coauthors wrote.

Several explanations as to why utilization didn’t change focus on the structure of the program itself. Under the model, payments to physicians aren’t included in the hospitals’ budget—unlike the risk doctors take on in an accountable care organization model—which Roberts and his coauthors guessed limited the influence the program had on physicians’ behavior. While hospitals assume all the risk, they’re still paid on a per-admission or per-service basis.

“Hospitals that lowered utilization were expected to raise their prices to receive their budgeted revenue, but since price increases were generally limited to less than 5%, the program’s structure may have dampened incentives to substantially reduce volume,” they wrote.

The third explanation is the least damaging to the all-payer model’s future prospects. Roberts and his coauthors said hospitals experienced some initial problems in aligning physicians and staff and implementing new care management programs in the first months of the program. Therefore, their analyses “may not encompass 2 full years of exposure to the program’s incentives,” and thus may miss its impact on utilization.

“Further monitoring is needed to assess how hospitals adapt to this payment model over time,” Roberts and his coauthors wrote.

Models which build on Maryland’s program may address these shortcomings. Vermont’s all-payer model is like a combination of the global budgeting with CMS’s Next Generation ACO model, with physicians being paid on a monthly basis for their Medicare, Medicaid and commercially insured patients. The model began its first performance year on Jan. 1 and is currently scheduled to run through 2022.