The impact of COVID-19 further shows the need to modify federal pay-for-performance models, hospital advocate says

• Quality measures derived during the COVID-19 pandemic are not an accurate gauge of hospital performance in federal programs such as Value-Based Purchasing.
• Performance measures will be vulnerable to distortion even after the pandemic because of specific clinical challenges and broader operational issues that have arisen.
• The need to update the programs was evident even before the pandemic.

The metrics that form the foundation of federal pay-for-performance programs have been skewed by the COVID-19 pandemic, and greater accommodations are needed, according to a hospital industry leader.

“Measurement ultimately is the basis of value-based purchasing,” said Chip Kahn, president and CEO of the Federation of American Hospitals. “If you can’t measure, you can’t determine what the outcome is or what value is.”

But during the pandemic, “Our ways of measuring are not necessarily reflecting what a hospital is doing or how good the care it’s providing,” Kahn said during a presentation in September at the Virtual Collocated Value-Based Payments Summit.

Kahn primarily was speaking about the Hospital Value-Based Purchasing, Hospital Readmissions Reduction and Hospital-Acquired Condition Reduction programs. Combined, those programs put more than 5% of a hospital’s base MS-DRG payment at risk, although in Value-Based Purchasing, hospitals also are eligible for bonus payments.

CMS acknowledged the COVID-19 impact in the FY22 rule for the inpatient prospective payment system, Kahn noted. The agency said it would suppress specific 2020 data in the programs.

Although the adjustment is “unprecedented,” it may not go far enough.

“The industry appreciates the move that CMS has made,” he said. “But frankly, the rule only affects one year, and there will be a cascade effect of COVID throughout measurement.”

The pandemic’s impact on measurement

COVID-19 has affected performance on various measures that are part of the programs.

For example, hospitals have seen triple-digit increases in hospital-acquired infections (HAIs) such as central-line bloodstream infections, Kahn said. Rates of pressure ulcers have increased. Higher readmission and mortality rates have been seen after discharge for COVID-19 hospitalizations, “suggesting a period of heightened risk of clinical deterioration.”

Beyond specific clinical impacts, entire operations have been disrupted by shortages of staff and resources, including personal protective equipment. Transfer policies among health systems in various markets have left some hospitals with markedly higher COVID-19 caseloads, making the use of a single set of value-focused measures for hospitals all the more problematic.

It all adds up to a situation that is difficult to compare with validity to 2018 and 2019. “And frankly, the key to measurement is comparison,” Kahn said. “If you can’t compare to an earlier base year, you’re not going to have useful measurement.”

Bigger issues with P4P programs

Updates to the pay-for-performance programs were merited even before COVID-19, Kahn said.

“[As to] whether we had the right measures and whether those programs were geared to what’s necessary for judging what we think of as quality, many questions have been raised,” he said.

For example, Value-Based Purchasing Program scores have little correlation with mortality rates, among other outcomes. In addition, the HCAHPS patient experience survey used in that program needs to be digitized rather than remaining phone- and paper-based.

The readmissions reduction program may have run its course because hospitals seem to already have made progress to the extent feasible, he added, while HAI measures could be folded into the Value-Based Purchasing Program.

More fundamental changes should include stronger incentives, a streamlined set of measures and real-time availability of information “so that hospitals can actually use the metrics to improve,” Kahn said.

“One of the unfortunate aspects of the Medicare programs is that you’re sort of in an 18-month cycle,” he added. “By the time you get to real comparisons with other institutions, time has really passed [from] when those measures were gathered.”

The pay-for-performance programs represent a small segment of federal efforts to improve healthcare value. The Center for Medicare & Medicaid Innovation is looking at ways to expand accountable care and other potentially transformative programs, but Kahn thinks measures need to improve upon those that are seen in the Medicare Shared Savings Program and other models.

“My sense is that we now have available — or will in a short time — the kind of data and analytic ability to be constantly updating what we’re looking at and what we’re sharing in terms of information,” he said. Such metrics would go “way beyond the static measures that we’ve historically used because that was all we had.”