Topics
More on Quality and Safety

Health Affairs report pokes holes in hospital pay-for-performance programs

Authors suggest factoring hospital improvement into the scores.

"Would incentives for improvement be greater if all pay-for-performance programs specifically considered improvement as part of scoring hospital performance?" authors ask.

By 2017, Medicare officials want to have 85 percent of the program's hospital pays broadly tied to quality. But before that can happen, a new report says the agency and the industry need to fix three signature pay-for-performance programs.

This year, many hospitals find themselves being penalized under at least one of the Affordable Care Act's  programs: the Value- Based Purchasing Program (VBP),  the Hospital Readmissions Reduction Program (HRR) and the Hospital-Acquired Condition Reduction Program (HAC).

About 80 percent of about 3,400 U.S. hospitals subject to the programs will receive some penalty under one or more of the programs, including 30 percent of teaching hospitals, according to a new Health Affairs study.

By 2017, the same year federal health officials hope to have ACOs, bundled payment and other value-based payments covering much of Medicare's hospital reimbursement, 6 percent of total hospital payments will be based on performance in the HRR, VBP and HAC programs. To make sure these programs are working efficiently, Medicare should use their early experience to integrate them, said Charles Kahn III, president of the Federation of American Hospitals, along with researchers from Health Policy Alternatives.

[Also: See which hospitals will be hit with readmission penalties]

In the Health Affairs study, Khan and the Health Policy Alternatives set out to asses the three hospital pay-for-performance programs and whether they are achieving the goals outlined in the ACA. To at least some effect, they are.

Hospital performance across the three programs' metrics "is improving, although the extent to which improvement under the pay-for-performance programs exceeds what would be achieved as a result of public reporting alone is unknown," Khan and the co-authors wrote.

The hospital value-based purchasing program -- linking penalties or bonuses to measures of clinical processes, infections, safety, mortality and patient experience -- redistributed about $126 million in hospital payments for fiscal year 2015. Of 3,089 hospitals who received an adjustment, almost 45 percent were penalized with up to 1.5 percent of their base operating payments under the Inpatient Prospective Payment System. A full 2 percent of base operating pay will be at stake when the VBP is fully implemented in 2017.

The HAC reduction program covers some of the same hospital safety issues as the VBP, although it is run slightly differently. Hospitals are scored based on their rates of central line-associated bloodstream infection, catheter-associated urinary tract infection, and a composite measure of patient safety, and those in the worst-performing quartile receive a 1 percentage point reduction on their total IPPS payments. One-quarter of hospitals at the bottom quartile are subject to the 1 percentage-point reduction, regardless of industry-wide performance. Unlike the VBP, improvement is only recognized if the hospital moves out of the worst-performing 25 percent.

[Also: See which hospitals were fined for hospital-acquired conditions]

About 3,300 hospitals were subject to the HAC program this year, and about 293 shared $357 million in penalties. "The HAC Reduction Program payment adjustments for fiscal year 2015 show major teaching hospitals to be disproportionately penalized," Khan and co-authors wrote, noting that half of the penalties were applied to academic medical centers, even though they comprise only 7 percent of hospitals nationwide and 19 percent of hospitals in the bottom HAC quartile. They also found that many of the hospital scores on HACs were concentrated near the seventy-fifth percentile cut-off point that determines the penalty.

All three of the measures determining a hospital's score currently under the HAC reduction program are also included in the VBP, constituting about 30 percent of the total VBP score. Come 2017, this overlap will involve three other measures -- surgical site infections, Clostridium difficile infections, and methicillin-resistant Staphylococcus aureus infection.

For the hospital readmissions reduction program, up to 3 percent of base operating payments will be at stake when fully phased in, based on rates of preventable readmissions for patients with heart attack, heart failure and pneumonia. The HRRP, though, is perhaps the most contentious metric, based on what a number of researchers consider to be flaws in the adjustment for patient risk and factors outside of a hospital or doctor's control.

As Khan and co-authors write, readmissions were already declining when the penalty starting taking effect in 2013. In fiscal year 2015, about 75 of 3,478 hospitals subject to the HRRP were penalized, taking an average penalty of 0.5 percent of total operating payments and a collective $424 million.

Nearly 10 percent of hospitals accounted for almost half of the HRRP penalties, the study found, and that's raising concern that the measure may not be as meaningful as envisioned and disproportionately impacts hospitals serving particularly poor populations.

"While risk-adjusting for sociodemographic factors can mask poor quality of care in low-income areas, the concern is that in the context of the HRRP, penalties for hospitals serving vulnerable populations could exacerbate the problem and reduce access to high-quality care for low-income patients," Khan and co-authors wrote.

Like Healthcare Finance on Facebook

That issue is still ongoing, and a sociodemographic risk adjustment could always be possible. Either way, Khan and co-authors suggest that regulators take an integrated view of the programs as they try to improve.

"The ACA enacted the three new pay- for-performance programs in piecemeal fashion as separate programs layered on top of the IPPS with the intent to integrate payment policy and quality, defined broadly," they wrote. "The piecemeal adoption did not consider issues associated with appending quality programs to a payment system intended to cover the costs of an efficient provider of patient care."

With Medicare, Medicaid and private quality payment programs only growing, beyond the three ACA programs, "it is time to consider a more rational approach to better aligning payment policy with quality outcomes, given the experience to date and the questions raised when examining how these programs interact in combination," Khan and co-authors wrote.

Going forward, they think there should be several questions to guide the policies.

"First, are the measures appropriately risk-adjusted, and do they take into account the process of adjusting for sociodemographic factors recommended by the NQF?" Secondly, how should measures used in more than one program be used?

And, particularly in the case of the hospital-acquired condition program, they wonder if factoring in improvement would help move progress for the whole industry. "Would incentives for improvement be greater if all pay-for-performance programs specifically considered improvement as part of scoring hospital performance?"

Twitter: @AnthonyBrino