ANN ARBOR—After four years, a program designed to reward hospitals when they deliver high quality care has failed to show an impact on clinical process and patient experience measures, and mortality outcomes following hospitalization, according to a University of Michigan study.
The Hospital Value-Based Purchasing program, part of the Patient Protection and Affordable Care Act administered by the Centers for Medicaid and Medicare Services, offers incentive payments to acute care hospitals that meet certain quality measures when treating Medicare patients.
The study, published in the New England Journal of Medicine, looked at whether the program improved how care was delivered at participating hospitals when compared with control hospitals not exposed to HVBP.
Andrew Ryan, associate professor at the U-M School of Public Health, and colleagues at the U-M Medical School examined program data that measured seven clinical processes and eight patient experience indicators used to determine which hospitals received incentives. These include protocols for handling patients from the moment they enter hospital doors to discharge and beyond.
The team also studied death rates 30 days after leaving the hospital of patients treated for heart attack, heart failure and pneumonia.
"Hospital Value-Based Purchasing offered little patient benefit over four years," Ryan said. "Other than an improvement in mortality for pneumonia, there were no significant differences in the quality of care or reductions in deaths at the hospitals that participate in the program compared with those that do not."
The HVBP program has been evaluated at various times throughout the four years and has not shown an impact at any point in time.
"Four years should be long enough to see the impact of a change," said senior author Justin Dimick, the George D. Zuidema Professor of Surgery at the Medical School and professor of health management and policy at the School of Public Health. "Prior studies conducted within a year or two of a policy being implemented might be considered too early, but if a policy is having its intended effects, there should be some measurable change by four years."
A total of 2,842 hospitals were included in the clinical-process analysis, 3,247 in the patient-experience analysis, and 2,195, 3,256 and 3,525 for the heart attack, heart failure and pneumonia analyses, respectively.
The value-based program is funded by a reduction in the base rate of Medicare reimbursements to hospitals of 2 percent in what are called diagnostic related groups. DRGs are predetermined amounts paid to hospitals for common diagnoses based on the average cost of care for that treatment or procedure. The savings is then used to pay those hospitals that meet quality standards.
When the program started in 2013, the most hospitals could earn was 1 percent but today they can get up to 2 percent. This essentially gives those that perform well the opportunity to earn back what was taken away and penalizes those that do not perform.
Dimick said to make the program work, the government would need to rethink the quality metrics and the incentives.
"The measures were too complex and the incentives were likely too small," he said. "For a policy to have a positive impact, the measures need to be simple and understandable to clinicians. Clinician leaders must be able to interpret where they are falling short, so they can make changes. For a policy to work, the incentives must also be large enough to capture the attention of hospital leaders so they can deploy resources to do the improvement work."
Despite the lack of success of this program, Ryan said critics of health reform measures instituted in recent years should not be quick to call the value-based concept unsuccessful, saying it is about the design of this particular program.
He points to the recent finding about the Hospital Readmission Reduction Program that kept 2,400 people out of nearly 275,000 from heading back to the hospital, resulting in a savings of $32 million from reduced readmissions in 2015.
"There's not a clear blueprint for how value-based programs should be designed to improve care," Ryan said. "The fact that we're seeing success from other programs speaks to the importance of continuing to try."
Other U-M authors were Sam Krinsky and Kristin Maurer. Ryan and Dimick are members of the U-M Institute for Healthcare Policy and Innovation. The research was supported by grants from the National Institute on Aging.