AHA Comment Letter on CMS’ Home Health Prospective Payment System Proposed Rule for CY 2024

American Hospital Association

August 29, 2023 2:16 pm

Submitted electronically.

August 28, 2023

The Honorable Chiquita Brooks-LaSure
Centers for Medicare & Medicaid Services
Hubert H. Humphrey Building
200 Independence Avenue, S.W.
Room 445-G
Washington, DC 20201

Re: Medicare Program; Calendar Year (CY) 2024 Home Health (HH) Prospective Payment System Rate Update; HH Quality Reporting Program Requirements; HH Value-Based Purchasing Expanded Model Requirements; Home Intravenous Immune Globulin Items and Services; Hospice Informal Dispute Resolution and Special Focus Program Requirements, Certain Requirements for Durable Medical Equipment Prosthetics and Orthotics Supplies; and Provider and Supplier Enrollment Requirements; 88 Fed. Reg. 43,654 (July 10, 2023)

Dear Administrator Brooks-LaSure:

On behalf of our nearly 5,000 member hospitals, health systems and other health care organizations, including approximately 1,000 hospital-based home health (HH) agencies, and our clinician partners – more than 270,000 affiliated physicians, 2 million nurses and other caregivers – and the 43,000 health care leaders who belong to our professional membership groups, the American Hospital Association (AHA) appreciates the opportunity to comment on the calendar year (CY) 2024 HH prospective payment system (PPS) proposed rule.

Our comments focus on CMS’ proposed budget neutrality adjustment related to the Patient-Driven Groupings Model (PDGM) under the HH PPS as well as its proposed market basket update. The AHA is extremely concerned that these policies, if finalized, would result in an overall net negative update to the HH PPS. We urge the agency to adequately resource HH providers as they are a critical part of the care continuum. We are particularly concerned about the substantial size of the agency’s proposed budget neutrality adjustment, a cut of 5.653%, and again call on CMS to withdraw it. Instead, we urge the agency to revise its methodology to more accurately account for changes in care delivery and payment dynamics due to the implementation of the PDGM. We also have concerns about the inadequacy of the proposed market basket update given the financial pressures facing HH agencies, including critical staffing shortages and rising supply costs. In addition, final data from CYs 2021 and 2022 indicate that the market basket forecasts underpaid HH agencies by a combined 5.1% for these years. This, combined with the difficult inflationary environment and the large budget neutrality adjustment proposed in this rule, risks putting HH agencies in serious financial peril. As such, we urge CMS to utilize its authority to provide a market basket adjustment to account for these extraordinary circumstances.


PDGM Behavioral Offset

AHA reiterates its staunch opposition to CMS’ proposal to continue to apply a PDGM budget neutrality adjustment based on a defective methodology. The agency specifically proposes to cut the standardized 30-day period payment rate by 5.653% in CY 2024. As described below, using this flawed methodology would erroneously lower overall payment levels to HH providers relative to the prior payment system. In addition, these proposed reductions have the potential to exacerbate capacity issues throughout the entire continuum of care, including for acute-care hospitals. Therefore, the AHA again calls on CMS to revise its methodology to more accurately account for changes in care delivery and payment due to the PDGM transition.

In compliance with the Balanced Budget Act (BBA) of 2018, CMS implemented the PDGM case-mix system together with a 30-day payment episode on Jan. 1, 2020. This law called for a budget neutral implementation that centered on the new 30-day episode of care. The adjustment includes two types of behavioral offsets:

The BBA did not specify a particular methodology for determining budget neutrality. Indeed, CMS’ budget neutrality adjustment for CY 2020 was set prospectively at -4.36% based on several assumptions regarding expected provider behavioral changes. Specifically, CMS assumed that HH agencies would alter their coding of primary and secondary diagnoses, both of which are key drivers of the PDGM payment setting process. In addition, CMS assumed that the number of low-volume cases, known as low-utilization payment adjustment (LUPA) cases, would decrease.

By law, the budget neutrality adjustment process will continue to occur each year through 2026, although the agency deferred action until CY 2023 in response to the COVID-19 pandemic. For CY 2023, CMS finalized a -7.85% permanent adjustment for purported overpayments in CYs 2020-21, to be applied over two years (CYs 2023 and CY 2024). Thus, in this CY 2024 rule, the agency is proposing to apply the second half of this reduction, a cut of 3.925%. However, it is also proposing an additional permanent adjustment to account for estimated CY 2022 overspending. In total, it proposes a cut of 5.653% for CY 2024. Finally, CMS says that while it has not yet proposed any temporary adjustments, it currently estimates it will need to recoup a total of approximately $3.4 billion in future years, which, for context, would be approximately a 20% reduction if applied in a single payment year at current net payment levels.

CMS’ Methodology Remains Flawed. The AHA reiterates its concerns that CMS overlooked numerous factors when calculating its PDGM-related budget neutrality adjustments. As discussed in our comment letters for CYs 2022 and 2023, the three behavioral assumptions that CMS made when calculating its initial adjustment of -4.36% did not match actual behavior by the field in CY 2020.

In addition, as AHA also has raised in its preceding year’s comment letters, CMS’ methodology for determining budget neutrality does not account for the drop in average per-episode therapy services under PDGM. This shift in care delivery means that CMS cannot simply reprice claims as it has done to date; instead, it must account for these additional dynamics to reach an accurate budget neutrality figure. Indeed, PDGM shifted incentives away from therapy visits resulting in a different unit of care with a new clinical and cost profile, which is incomparable to the pre-PDGM unit of care. Accounting for such shifts in care delivery when calculating budget neutrality is precisely what CMS did for the transition to the new PDPM under the SNF PPS. When determining a behavioral adjustment under that system, CMS determined that a drop in therapy utilization meant that comparison of data between the two payment systems would lead to a significant underestimation of what payments would have been under the prior system, thus leading to an inappropriately higher budget neutrality adjustment.1 However, for its PDGM calculations, CMS has refused to factor in changes in therapy delivery, which dropped by 29.7% in CY 2020, relative to CY 2019.

In addition, the PDGM framework – with 432 payment units, a 30-day episode of care and multiple case-mix levers – is significantly different, and therefore impossible to crosswalk to the prior payment system – with 153 payment groups, a 60-day episode of care and one dominant case-mix factor (therapy volume). Given these multiple, major differences, and consistent with CMS’ position noted above on the SNF PPS parity adjustment, PDGM-era claims from CY 2020 and 2021 cannot simply be recalculated using the prior payment system’s parameters to estimate what payments would have been under the prior model. Combining this significant shortcoming with a global pandemic that temporarily but fundamentally reshaped hospital and post-acute practices makes the repricing comparison done by CMS even more unsound.

We refer CMS to AHA’s comments in response to the CYs 2022 and 2023 HH PPS proposed rules for the entirety of our methodological objections to CMS’ budget neutrality approach. In addition, AHA reiterates its strong support for the analysis commissioned by the Partnership for Quality Home Healthcare, which has raised substantial concerns with the accuracy of the PDGM budget neutrality cuts. For these reasons, AHA continues to urge CMS to withdraw its proposed budget neutrality cuts and revise its methodology to more accurately account for changes in care delivery and payment dynamics due to the transition to PDGM.

Budget Neutrality Adjustments Jeopardize Access to Care and Disrupt the Care Continuum. CY 2023’s budget neutrality adjustment accelerated the strain on a HH field that already had been facing a two-pronged challenge of a critical labor-shortage and a particularly sharp inflationary environment. The imposition of additional cuts could seriously imperil Medicare beneficiaries’ access to HH services and contribute to additional limits in hospitals’ capacity to care for patients.

The most recent analysis from Kaufman Hall in its National Hospital Flash Report indicates that from 2020 to present, overall expenses have risen by 22% for hospitals.2 Although HH agencies have a different mix of goods and services, many of the key components of their expense profiles are the same, such as heavy reliance on nurses, nurse aides, therapists and other clinicians, as well as medical supplies. Indeed, much of the increase in expenses for health care providers have been driven by labor costs, including contract labor costs, which have risen 258% since 2019.3 As CMS knows, there has been a nationwide shortage of both clinical and non-clinical workers, including those needed to staff HH agencies. With reductions to payment rates, HH agencies struggle to retain and train staff to provide high-quality care for Medicare beneficiaries. Further, the Department of Health and Human Services (HHS) has found that health care workforce shortages will persist well into the future, meaning that CMS’ plan to continue to apply erroneous budget neutrality adjustments will continue to accelerate serious threats to the HH field’s ability to properly recruit and staff for their patients’ needs.4

In addition to labor costs, medical supply costs per patient have risen 18.5% from 2019 through 2022. Drugs make up a large portion of this increase, with an HHS study finding that many commonly used drugs have had their price increase by more than 30% in recent years.5

The threat to access to care is not hypothetical. Indeed, recent data highlight the ongoing reduction in access to HH services. Specifically, a 2023 report from Careport demonstrates that as the COVID-19 pandemic continued, patients were increasingly rejected from HH agencies as providers struggled to meet staffing demands.6 Notably, the rejection rate rose to 76% at the beginning of CY 2023, the highest levels ever recorded, which coincided with the imposition of the first post-implementation PDGM budget neutrality cuts. To emphasize, this means that three-fourths of all requests for HH agency care, most coming from hospitals for hospitalized patients, are rejected. This is very concerning and deserves examination.

In addition to overall disruptions in access to services, the data also indicate there is a disproportionate impact on historically marginalized communities. CareJourney’s analysis of Medicare fee-for-service claims data, shown below, found that these groups had a significantly lower conversion rate than white beneficiaries when referred for HH services following discharge from an acute-care hospital. The same holds true for Medicare-Medicaid dual eligibles when compared to non-dual eligible beneficiaries. This indicates that access challenges by the combination of factors discussed herein is having an inordinate impact on these communities, something that CMS should seek to address.

Source: CMS Virtual Research Data Center
Data: 2022 inpatient claim files filtered for STAC claims. Discharge data based on Q1-Q3 2022 data.
ADI: Area Deprivation Index (ADI) – from the University of Wisconsin Neighborhood Atlas Area Deprivation Index (ADI). It allows for the rankings of neighborhoods by socioeconomic disadvantage in a region of interest (https://www.neighborhoodatlas.medicine.wisc.edu/)

Further, CMS’ own data bears out how the recent economic conditions have strained access to HH services. As presented in Table B1 of the proposed rule, both the total 30-day periods of HH care and unique beneficiaries utilizing HH care was lower in CY 2022 than in CY 2020. In addition, the total visits per period of care was lower in CY 2022 than CY 2020 in every discipline other than social work, which remained flat. These figures are undoubtedly related to the difficult financial circumstances facing providers.

If addition to the growing impasse for HH services, CMS should be further troubled by the repercussions access restrictions have on acute-care hospitals. The same Careport analysis found that the average length of stay for patients discharged to HH has risen by 11% from 2019 to 2022. This is consistent with recent AHA findings of a 12.9% increase.7 As HH agencies lose capacity, acute-care hospitals proportionately lose their ability to discharge their patients. This, in turn, forces the acute-care hospitals to board patients who are ready for discharge, driving up costs and minimizing capacity to admit new patients. Indeed, CMS should consider adequately resourcing HH providers as a valuable tool that can be used to maximize acute-care hospital capacity and ensure the entire care continuum runs as smoothly as possible for Medicare beneficiaries.

Proposed CY 2024 HH Market Basket Update

The proposed market basket update of 2.7% (3% reduced by 0.3% productivity adjustment) for CY 2024 is inadequate to ensure Medicare beneficiaries have continued access to HH services. First, the update fails to account for missed forecasts from recent years. In addition, the proposed update continues to underappreciate the inflationary environment for health care providers. For these reasons AHA recommends CMS utilize its authority to revise the market basket upwards to meet its mission of ensuring access to HH services for traditional Medicare beneficiaries and to account for what the agency missed in the CY 2021 and CY 2022 market basket forecast. This is especially important if CMS proceeds with its plans to apply its budget neutrality adjustments as proposed.

In recent years, data indicate that the agency’s recent HH market basket updates were significantly lower than actual market basket growth. Specifically, CMS finalized updates of 2.3% and 3.1% in CYs 2021 and 2022, respectively; however, actual growth for these years were 4.2% and 6.2% respectively.8 This resulted in a substantial cumulative shortfall of 5.1%. Indeed, an analysis by Dobson DaVanzo & Associates, commissioned by the Partnership for Quality Home Healthcare, found that from 2021 through 2030, these forecast errors will result in nearly $11 billion in underpayments to HH agencies under the HH PPS – an astounding figure.

While final figures will not be available for several years, it is likely that the trend of under forecasting market basket cost growth has continued for CY 2023 and 2024. This is because, as AHA explained in last year’s comment letter, in atypical economic environments, the market basket approach used by CMS is much less adept at properly forecasting. And the atypical economic conditions are continuing. Specifically, as explained in the section above, health care costs, and especially labor and medical supplies, have seen historic increases which are slow to ameliorate.

The missed forecasts and inadequate proposed updates by themselves would be problematic and create serious strain on HH agencies’ ability to meet the needs of their communities. However, when combined with the budget neutrality cuts proposed by CMS, they are simply unsustainable. As CMS knows, a 2.7% update, reduced by a 5.653% budget neutrality adjustment would result in net update of -2.4% overall. Given the past forecast errors of 5.1%, the more than 8% in budget neutrality cuts already applied, and the difficult economic environment, a -2.4% update for CY 2024 could grievously disrupt access to HH services for Medicare beneficiaries.

Due to the consistent under-forecasting of the HH market basket, the unusual economic circumstances and the reductions in payment due to PDGM implementation, AHA appeals to CMS to deviate from its typical methodology and apply a more robust update to the HH market basket for CY 2024. This one-time adjustment would ensure that access to HH services is preserved for Medicare beneficiaries. Further, and of special importance to AHA, it will help ensure hospitals are not limited in their capacity to care for their communities and safely discharge patients in need of HH care.

Productivity Adjustment. CMS proposes to apply a productivity cut of -0.3% to the HH PPS market basket update. Specifically, under the Affordable Care Act, the HH payment update is reduced annually by a productivity factor, which is equal to the 10-year moving average of changes in the annual economy-wide, private nonfarm business total factor productivity (TFP). This measure was intended to ensure payments more accurately reflect the true cost of providing patient care. The use of the TFP is meant to capture gains from new technologies, economies of scale, business acumen, managerial skills and changes in production. However, as CMS knows, services provided through the HH benefit are hands-on, labor-intensive services and do not lend themselves to the productivity gains realized in other sectors. CMS itself has acknowledged that health care providers, due to the nature of their service, lack the ability to add efficiencies in the way other sectors do.9

Given all the various circumstances putting financial pressures on HH agencies, AHA would again like to express its concern about the application the productivity adjustment to HH agencies. AHA respectfully asks CMS to use its authority to account for the lack of parity in this adjustment when considering its overall payment adjustment to HH providers.


HH Quality Reporting Program

In this rule, CMS proposes to adopt two new quality measures for the HH Quality Reporting Program (HH QRP); remove one measure and two items from the Outcome and Assessment Information Set (OASIS); and begin publicly displaying performance data on four quality measures.

Discharge Function Score Measure and Removal of Overlapping Discharge Function Measures. CMS proposes to adopt this assessment-based outcome measure that estimates the percentage of HH patients who meet or exceed an expected discharge score during the reporting period beginning with the CY 2025 HH QRP; in the FY 2024 final rules for their respective payment systems, the agency has finalized adoption of the same measure in the Inpatient Rehabilitation Facility (IRF), Skilled Nursing Facility (SNF) and Long-term Care Hospital (LTCH) QRPs. While this cross-setting discharge function score measure appears to fulfill requirements of the IMPACT Act better than the current, setting-specific self-care and mobility discharge score measures used in the various post-acute care quality reporting programs (which CMS proposes to remove in this same rule), we continue to doubt the cross-setting applicability of this measure considering the different patient populations served by the various post-acute care settings. We urge CMS to wait until this measure has undergone endorsement review by a consensus-based entity (CBE) and demonstrates that it gleans useful information for patients and providers before adopting it for use in the HH QRP.

Additionally, the measure uses information from Section GG items that appear on all four of the patient assessment instruments across the various post-acute care settings. While patients are assessed using the same or similar items, the capabilities and goals of patients differ widely by setting. The measure developer notes that the measure is risk adjusted and calculated individually by setting; then, the calculation for measure performance “rolls up” information from several items to calculate an overarching score. Risk adjustment takes many variables into account, and denominators vary by setting (for example, the denominator for the measure when calculated in the LTCH QRP includes all patients regardless of payer, while for the SNF QRP the denominator consists of patients/residents under Medicare fee-for-service).

While we appreciate the work the developer has done to attempt to consider the myriad of differences in patient populations across the various settings – including demographics, case mix, severity of illness, length of stay and comorbidities – at some point these variables alter the underlying calculation of the cross-setting measure and result in four different measures. In other words, discharge function is calculated in a way that is not truly standardized, as the IMPACT Act intended. It is at this point we ask whether it is necessary to force a measure that is “cross setting” in name only into CMS quality programs. Perhaps it is if testing of the measure demonstrates that this measure produces statistically meaningful information that can be used to inform improvements in care processes, but until we have that information from the endorsement review process by a CBE, the AHA has serious doubts about the utility of this measure.

In addition, the measure uses a statistical imputation approach to account for “missing” assessment elements when codes on the assessments note that the “activity was not attempted” (ANA). If an assessor codes an item as “not attempted,” the imputation approach inserts variables based on the values of other activities that were completed; in other words, the calculation makes assumptions about what the patient would have scored on that item if it had been attempted based on their performance on other, similar activities that were. CMS argues that this approach “increases precision and accuracy and reduces the bias in estimates of missing item values.” While we understand that scores would be influenced more heavily by individual assessment items if there are fewer included in the calculation, CMS errors in labeling items coded ANA as “missing.” When an activity is not attempted, it is likely because it would be clinical inappropriate or dangerous for a patient to attempt it; for example, it would be ill-advised (and painful) for a patient with a healing wound on one side to roll left to right. In such a case, making assumptions about the patient’s function based on other activities would, in fact, not improve the precision of the score.

We also question whether it is precise and accurate to generically apply an “expected” discharge score based on statistical regressions to unique patient populations, and whether the comparison of observed to “expected” function could wholly be attributed to the facility’s quality of care. The calculation approach for the “expected” discharge score is opaque, which makes it difficult for providers to know what they are working towards. In reality, providers strive to help each individual achieve his or her own specific goals related to function, independence and overall health. These goals are not based on statistical regressions.

The AHA understands the purpose of this measure and agrees that the discharge function measure currently in use in the HH QRP (Application of Percent of Long-term Care Hospital Patients with an Admission and Discharge Functional Assessment and a Care Plan that Addresses Function) does not meaningfully evaluate comparative performance across post-acute care settings. However, without further testing and review of the proposed Discharge Score measure by a CBE, we are not certain that this measure brings value to the QRP and thus cannot support it for adoption.

Percent of Patients/Residents who are Up to Date with COVID-19 Vaccination Measure. Beginning with the CY 2025 HH QRP, CMS proposes to adopt this assessment-based process measure that reports the percentage of stays in which HH patients are up to date with their COVID-19 vaccinations per the CDC’s latest guidance. The agency reasons that the measure would, when publicly reported, provide useful information for patients and their caregivers when choosing a facility, and “would be an indirect measure of HH agency action” since the agency would, according to CMS, have the opportunity to administer the vaccine to patients during their episode, coordinate a follow-up visit for the patient to obtain the vaccine at their physician’s office or local pharmacy, or educate the patient about the importance of staying up to date with vaccinations. CMS finalized the adoption of this measure for the SNF, LTCH and IRF QRPs in their respective FY 2024 proposed payment system final rules.

The AHA strongly supports the vaccination of health care providers and communities for COVID-19 and acknowledges the importance of up to date vaccinations. However, this measure has not been tested for validity and reliability and thus we cannot support it without knowing that it is, at minimum, feasible to report and likely to produce statistically meaningful information. Furthermore, we are not clear that the conceptual construction of the measure is the best way to encourage vaccination, especially in post-acute settings where care is delivered in episodic rather than longitudinal fashion. When reviewed by the National Quality Forum (NQF)’s Measure Applications Partnership (MAP) during the 2022-2023 review cycle, the Post-acute/Long-term Care Workgroup voted “Do Not Support” for this measure, meaning that a multi-stakeholder panel of experts representing providers, patients and payers do not support this measure for inclusion in the HH QRP as well.

Vaccination status among patients is subject to many patient-level factors outside of the control of providers. For post-acute facilities and providers, it may be infeasible or inappropriate to offer vaccination for patients due to length of episode, ability to manage side effects and medical contraindications, or other logistical challenges to gathering information from a patient who may have received care from multiple proximal providers. Even without these challenges, however, patients/residents may choose to forgo vaccination despite a provider’s best efforts. It is possible that a post-acute care provider could have a robust effort to encourage vaccination among their patients/residents, but still have a relatively low rate of vaccination. As the Health Equity subcommittee of the NQF MAP noted in its review of this measure, cultural norms often play a large role in vaccine confidence. While post-acute providers will always seek to counsel vaccination in a culturally sensitive way, they also want to honor the choice of their patients once they have offered their clinical advice.

We reiterate that we understand the importance of vaccination in protecting patients from the most serious outcomes of COVID-19. However, it is unclear whether the use of this measure will produce those results or if it is feasible for post-acute care facilities to collect and report the information necessary. The measure consists of a single yes or no item on the OASIS without any requirements for documentation or validation of vaccination status; while we acknowledge that additional documentation would be unduly burdensome for providers to collect, without it the measure is a mere checkmark in a box with no evidence that it leads to improved quality of care (since, as stated above, the measure has not been fully tested). For these reasons, we do not support the adoption of this measure in the HH QRP. CMS also may want to consider whether alternative measure constructions focused on the actions providers take in encouraging vaccination might be better suited to achieving the goal of higher vaccination rates.

HH Value-based Purchasing (VBP) Program

Replacement of OASIS-based Discharge to Community (DTC) Measure with Claims-based Version. CMS proposes to replace the measure that is currently used in the HH VBP program with a measure identical except for its data source (claims rather than OASIS assessments) and period (two years rather than one). The AHA does not oppose this proposal, as the OASIS-based measure has not been publicly reported since 2016 and thus does not appear to provide value to patients and providers. In addition, the claims-based measure is informed by two years of data, which is likely to improve the reliability of performance calculations.

Replacement of OASIS-based Function Measures with New Discharge Function Score. As we note above in our comments on CMS’ proposal to adopt this measure for the HH QRP, we have serious doubts about the utility of this measure if it were to undergo the evaluation by a CBE. We understand that CMS has already finalized the measure for adoption into the other post-acute care quality reporting programs and is thus likely to adopt it for the HH QRP as well; however, we strongly urge the agency to hold off on adopting this measure into a pay-for-performance program until we have better information about whether the measure gleans information as intended.

Replacement of Hospital Use Measures with Potentially Preventable Hospitalizations Measure. CMS currently uses two measures to assess HH patient use of hospital services during their HH episode: Acute Care Hospitalization During the First 60 Days of Home Health (ACH) and Emergency Department Use Without Hospitalization During the First 60 Days of Home Health (ED Use). In the CY 2022 HH PPS final rule, CMS replaced the ACH and ED Use measures in the HH QRP with the Potentially Preventable Hospitalizations (PPH), The PPH measure calculates a risk-adjusted rate of hospitalizations or observation stays for certain diagnoses that could have been prevented with proper management and care according to the Agency for Healthcare Research and Quality’s Prevention Quality Indicator Ambulatory Care Sensitive Conditions. CMS would remove the ACH and ED Use measures and adopt the PPH measure beginning CY 2025.

We agree that the all-cause hospital utilization measures proposed for replacement may not provide accurate assessments of provider performance or actionable information. HH providers may be unable to prevent ED visits due to factors outside of their control, including geographic factors like a lack of alternative (and lower intensity) sites of care or patient-level factors like a lack of family or community support. In addition, it is difficult to determine appropriate attribution for hospitalization between different providers and settings, especially because the current all-cause hospitalization measure does not require the reason for admission to be related to the reason the patient is receiving home health care. By removing these two measures in favor of one that is more likely to reflect whether HH agencies are providing proper management and care as well as clear discharge instructions and referrals, CMS can better assess quality of care for the purposes of the HH VBP program. Thus, we support removal of the ED Use and Hospitalization During the First 60 Days of Home Health measures.

However, when CMS proposed this replacement in the CY 2022 HH PPS rule, we voiced two concerns that we continue to urge CMS to consider regarding the PPH measure. First, the PPH measure uses the Agency for Healthcare Research & Quality’s Prevention Quality Indicators and Ambulatory Care Sensitive Conditions to inform the definition for hospitalizations that can be potentially prevented. These frameworks are not specific to post-acute care or to hospitalization, and HH care generally is not included in the definition of ambulatory care. While CMS shares its analyses that applied the conceptual definition of potentially preventable hospitalization to claims data, we encourage the agency to continue to monitor performance and hone this definition with the help of stakeholders and quality improvement experts.

Second, we recommend that CMS incorporate social determinants of health (SDOH) into its risk adjustment methodology. As proposed, the PPH measure is risk-adjusted for demographics including age, sex, enrollment status and activities of daily living scores, but not other SDOH that have a demonstrated impact on hospitalization. Several reports, including from the National Academies of Medicine, have shown that there are a number of plausible mechanisms by which sociodemographic information can be incorporated meaningfully into quality measurement. We urge CMS to incorporate social risk factors into the risk adjustment for the PPH measure and submit the measure for review by a CBE before its adoption.

We appreciate your consideration of these issues. Please contact me if you have questions or feel free to have a member of your team contact Jonathan Gold, AHA’s senior associate director for policy, at (202) 626-2368 or jgold@aha.org.



Ashley Thompson Senior Vice President,
Public Policy Analysis and Development