The GAO Calls Out Problems It Sees in the Hospital Value-Based Purchasing Program

July 11, 2017
In a report published at the end of June, the U.S. Government Accountability Office challenged HHS and CMS to improve the measurement methodology embedded in the Hospital Value-Based Purchasing Program

In a report published online late last month, the United States Government Accountability Office (GAO) criticized a number of features of the Hospital Value-Based Purchasing Program that has been evolving forward under the Medicare program. In the report, entitled “CMS Should Take Steps to Ensure Lower Quality Hospitals Do Not Qualify for Bonuses,” the agency took aim at some of the methodology around quality and financial outcomes at hospitals participating in the value-based purchasing program, calling for extensive changes in methodology in order to ensure that under-performing hospitals do not receive bonuses under the program.

As the GAO notes in its introduction, “The Hospital Value-Based Purchasing (HVBP) program aims to improve quality of care and efficiency by creating financial incentives for about 3,000 participating hospitals. From fiscal years 2013 through 2017, performance on quality and efficiency measures varied by hospital type. Safety net hospitals—those that serve a high proportion of low-income patients—generally lower in quality compared to all participating hospitals. In contrast, small rural and small urban hospitals—those with 100 or fewer acute care beds—scored higher on efficiency compared to all hospitals. Payment adjustments—bonuses or penalties, announced prior to each fiscal year —have varied over time for all hospitals. In four out of the five years of GAO’s analysis, small rural and small urban hospitals were more likely to receive a bonus compared to all participating hospitals, while safety net hospitals were more likely to receive a penalty. While a majority of all hospitals received a bonus or a penalty of less than 0.5 percent each year, the percentage of hospitals receiving a bonus greater than 0.5 percent increased from 4 percent to 29 percent from fiscal year 2013 to 2017. In dollar terms, most hospitals had a bonus or penalty of less than $100,000 in fiscal year 2017.”

What’s more, the GAO noted, “Some hospitals with high efficiency scores received bonuses, despite having relatively low quality scores, which contradicts the Centers for Medicare & Medicaid Service’s (CMS) stated intention to reward hospitals providing high-quality care at a lower cost. Further, among hospitals that were missing one or more quality scores, the efficiency score had a greater effect on the total performance score because of the methodology used by CMS. This methodology compensated for the missing scores by increasing the weights of all of the non-missing scores. Consequently, hospitals with missing scores were more likely to receive bonuses than hospitals with complete scores.”

Nor is any of this happening in a vacuum. As the GAO report noted, the relative weights given to the five “domains” within the Hospital Value-Based Purchasing (HVBP) Program have been designed to shift over time. This is how those values are changing: “clinical processes” accounted for 70 percent of the total weight of measurement in 2013; but have declined to 45 percent in 2014, 20 percent in 2015, 10 percent in 2016, and only 5 percent this year, in 2017. Meanwhile, though “patient experience” remains weighted relative steadily (at 30 percent in 2013-2015, and now 25 percent in 2016 and 2017), other domains are shifting. “Patient outcomes” were not included in 2013; then, that domain moved to 25 percent of the total weight of scores in 2014, 30 percent in 2015, and 40 percent in 2016, before falling back to 25 percent this year, in 2017. The “safety” domain is being measured for the first time this year, with a value of 20 percent of the total weight of measurement. And “efficiency,” not measured in 2013 or 2014, went from 20 percent in 2015, to 25 percent in 2016 and 2017.

As the GAO stated, “By law, the HVBP program is budget neutral, which means that the total amount of payment increases, or bonuses, awarded to hospitals deemed to provide higher quality of care must equal the total amount of payment reductions, or penalties, applied to hospitals deemed to provide lower quality of care. To fund the HVBP program, CMS first applies an initial fixed percentage reduction to the amount of each hospital’s Medicare reimbursements for its patients that fiscal year. The initial percentage reduction was 1 percent in fiscal year 2013 and has grown by 0.25 percent each year to the maximum of 2 percent for fiscal year 2017 and beyond, as specified in PPACA [the Affordable Care Act]. CMS determines each hospital’s payment adjustment based on the hospital’s total performance score relative to all participating hospitals.” Further, the agency wrote, “Hospitals with payment adjustments that exceed the initial reduction receive a net increase, or bonus. Hospitals with a payment adjustment less than the initial reduction have a net decrease, or a penalty…These payment adjustments are applied to the inpatient Medicare payment for each discharged patient throughout the upcoming fiscal year.”

Very importantly, the GAO wrote, “Safety net hospitals generally had lower median quality domain scores in comparison to all hospitals, while small rural and small urban hospitals generally scored higher on quality and efficiency domains during fiscal years 2013 through 2017. Median scores for each of the separate quality domains—clinical processes, patient experience, patient outcomes, and safety—were consistently lower for safety net hospitals and were generally higher for small rural and small urban hospitals than for hospitals overall during fiscal years 2013 through 2017.”

In its analysis, the GAO noted that “CMS decided to proportionally redistribute missing domain scores in order to maintain the relative weights of each remaining domain and reliably score hospitals on their performance. However, the issues we identified with the weighting formula—in that it results in some lower quality hospitals receiving bonuses—are exacerbated for hospitals with missing domain scores. As a result, hospitals with missing domain scores are more likely to get a bonus, and, in some cases, those bonuses are greater than median bonuses overall. Additionally, while CMS intended to keep the efficiency metric independent of quality, the effective weight of the efficiency measure depends on the extent to which hospitals report quality measures. As a result, the balance the agency tried to achieve in the total performance score—allocating 75 percent of the score to the quality domains and 25 percent of the score to the efficiency domain—is no longer achieved.”

And, in its conclusion, the agency warned that “The aim of the HVBP program is to improve hospital quality and efficiency by providing incentives for hospitals to improve their quality of care and to become more cost efficient. Throughout the 5 years of the program, CMS has made modifications to meet these goals by changing quality performance domains and domain weighting from year to year. With the addition of the efficiency domain in fiscal year 2015, CMS signaled the importance of hospitals’ providing care at a lower cost to Medicare, and, in its weighting formula, the agency tried to find balanced consideration for quality and cost. Rather than achieving this balance—which would have allowed the agency to identify and reward higher quality and lower cost hospitals—CMS’s weighting formula has resulted in bonuses for some lower quality hospitals, solely due to their cost efficiency. Because the program is budget neutral, bonuses for lower quality hospitals may result in smaller bonuses for hospitals that are performing well across all domains. The issue is especially stark for between 10 and 25 percent of the hospitals that were missing domain scores in fiscal years 2015 through 2017, which has also contributed to the awarding of bonuses to lower quality hospitals. If CMS continues to use the current formula, it will continue to reward hospitals that do not score well on quality and efficiency metrics.”

As a result, the agency concludes, “To ensure that the HVBP program accomplishes its goal to balance quality and efficiency and to ensure that it minimizes the payment of bonuses to hospitals with lower quality scores, we recommend that the Administrator of CMS take the following two actions: revise the formula for the calculation of hospitals’ total performance score or take other actions so that the efficiency score does not have a disproportionate effect on the total performance score; revise the practice of proportional redistribution used to correct for missing domain scores so that it no longer facilitates the awarding of bonuses to hospitals with lower quality scores.”

The report has been sent to “the appropriate congressional committees, the Secretary of Health and Human Services, the Administrator of CMS, and other interested parties.”