Maryland’s global budget model: Impact on overall healthcare utilization remains murky

April 10—Implementation of global budgets at some Maryland hospitals cut utilization for several services, according to recent research, but evidence suggests the care merely shifted to other hospitals and providers.

A key goal of the high-profile Maryland hospital global budget model, also called the Total Patient Revenue (TPR) model when it debuted in 2010, was to promote efficient resource use by replacing fee-for-service (FFS) payment for some rural hospitals with “global budgets.”

A new study in Health Affairs compared eight TPR hospitals to seven similar non-TPR Maryland hospitals and found inpatient admissions and outpatient services fell sharply at TPR hospitals.

Specific findings included:

• 12% decline in emergency department (ED) admission rates
• 23% decline in direct (non-ED) admissions
• 45% decline in ambulatory surgery center visits
• 40% decline in outpatient clinic visits and services

However, visits to all Maryland hospitals declined by smaller amounts and Medicare spending increased for residents of TPR counties. Those trends suggested to the study’s authors that some care moved outside of the global budget.

The authors were unable to assess the efficiency of patient moves outside the global budget, noting that care could have moved to more efficient locations.

TPR was replaced in 2014 by the Global Budget Revenue program, which expanded global budgets to all Maryland hospitals.

“The TPR program was a blunt tool that provided incentives to hospitals to provide less care, whether or not that care moved to a more efficient location,” the authors wrote.

Why the declines occurred

The authors were optimistic that the cumulative 12% decline in ED admissions could have stemmed from reversing hospitals’ prior incentives to admit marginal patients from the ED. They noted that prior research found a large variation in ED admission rates across hospitals, “which suggests that some of the admissions are discretionary.”

The study found no reduction in length of stay (LOS), which the authors said may have stemmed from pre-existing incentives for hospitals to limit LOS.

The authors also concluded that some of the declines they found in direct admissions could reflect greater efficiency (e.g., moving some surgeries from inpatient settings to outpatient settings), but evidence indicated most of the drop was offset by higher admissions in non-TPR Maryland hospitals.

Similarly, some outpatient services likely moved from TPR hospitals to nonparticipating hospitals, physician offices, or outpatient clinics, they wrote.

Service shifts differed by care type. For example, there was a sharp relative decline in mammography and a relative increase in chemotherapy, but the authors could not determine whether chemotherapy services were moved to more appropriate or lower-cost locations.

“We hypothesize that hospitals would have stronger incentives to reduce volume for services with lower margins (high cost relative to revenue),” the authors wrote.

Although the authors found declines in both inpatient and outpatient visits, the model was designed to respond by reducing hospital budgets. Without these budget reductions, the authors wrote, there might have been larger declines in outpatient services, even though the budget reductions were insufficient to counteract hospitals’ incentives to reduce outpatient services.

Medicare prices show signs of increasing

For FFS Medicare beneficiaries, the analysis suggested that the prices that Medicare paid may have increased under the model.

“This could be a direct outcome of the TPR program,” the authors wrote. “If services fell at TPR hospitals, the program allowed hospitals to charge more for the services they still provided.”

The study followed previous research on the first three years of Maryland’s Global Budget Revenue (GBR) program (2014–16). That study found fewer ED admissions, lower hospital costs, and lower total expenditures for the Medicare population, but no spending reductions for commercially insured patients.

A separate study of the first two years of GBR found no evidence that GBR affected hospital use or the frequency of primary care visits.

“If much of the care that is no longer provided within budget moves off budget, as our study suggests, a logical implication is that total healthcare costs could rise,” the authors of the latest study wrote.

They found evidence of higher overall healthcare spending for the FFS Medicare population.

Although the 2014 expansion of TPR to GBR, which brought all Maryland hospitals under global budgets, may limit the opportunities to shift care to other Maryland hospitals, “smaller hospitals will likely retain some ability to reduce direct admissions, thus pushing more patients to other hospitals, inside or outside Maryland.”

Implications for global budgets

Overall, the authors concluded that the incentives created by any global budget model will be “problematic” for providers because any one participant provides only some of a patient’s care.

“When top-down regulation, however well intentioned, competes with providers’ financial incentives, the incentives will often win — a problem known in the management literature as ‘the folly of rewarding A, while hoping for B,’” they wrote.

In response, they urged cautious implementation of alternative payment models, with close attention paid to the incentives they create and how effects should be evaluated, and recognition that the full effects are likely to be revealed over time.