Paragons’s Senior Policy Analyst Joe Albanese responds to Senator Sheldon Whitehouse and Senator Bill Cassidy’s request for information for their new legislation, the Pay PCPs Act.
Dear Chairman Whitehouse and Ranking Member Cassidy,
Thank you for your work on the important issue of physician payment reform in Medicare, including seeking out public feedback on your legislation, the Pay PCPs Act.
This comment letter responds to several of the topics you raised in your May 15, 2024, request for information and draws upon my analysis of Medicare payment policy in various published reports and testimonies before the House Energy and Commerce Committee.1
Hybrid Payments
It is commendable that you are exploring how to translate the experience of alternative payment models (APMs) tested by the Center for Medicare and Medicaid Innovation (CMMI) into permanent payment reforms. Moving away from a fee-for-service (FFS) payment methodology, particularly for ongoing care management, may be a sensible way to encourage greater accountability for the quality and cost of care among providers.
However, it is important to be mindful of the relatively limited body of evidence that currently exists for the success of hybrid payments for primary care. Table 12 lists CMMI models that utilize hybrid payments for primary care. None have led to net savings to the Medicare program, and several have not yet been evaluated or begun. Consequently, Congress should take care to identify what factors may allow hybrid payment models to succeed or not in a primary care environment. For example, burdensome quality reporting requirements may detract from the success of a model testing new approaches to payment.
Furthermore, a key limitation to incorporating hybrid or capitated payments for just one setting of care or payment system is that individual providers, even primary care providers who refer patients to specialists or undertake care management duties, are typically not responsible for all the care their patients receive. This is why capitated reimbursement at the payer level, as in Medicare Advantage (MA), is a conceptually simpler way to achieve accountability for patients’ total cost of care.
Risk Adjustment
Designing a new payment methodology also affords an opportunity to incorporate more risk adjustment into FFS payment systems. Congress’s goal should be to pursue a unified approach to risk adjustment between MA and traditional Medicare (TM) – including accountable care organization (ACO)-based payment models. This would align incentives for risk-bearing entities across Medicare and allow for more accurate comparison of health risk for the entire Medicare population by the Centers for Medicare and Medicaid Services (CMS). Two broad dimensions to consider are how to calculate risk and adjust payment.
Risk adjustment for MA plans and ACOs largely relies on the same risk model incorporating health care utilization data from both TM and MA patients. The prevalence of “unmanaged” FFS claims in the model (that is, TM claims that are not connected to an ACO or other risk-bearing entity) distorts the measurement of risk because FFS payment tends to incentivize more utilization than in MA or ACOs without similar diagnostic coding.3 Calculating risk from data sources besides claims and encounters could be a useful technical improvement.4
It will also be important to design risk-adjusted payment in a way that minimizes distortions on provider and payer activities. Coding intensity is one potential concern; compared to unmanaged FFS payments in TM, risk adjustment incentivizes more intensive diagnostic coding by MA plans and ACOs because it leads to higher payment. CMS combats this incentive in MA through a 5.9 percent payment adjustment and audits of medical records, while ACOs face limits on changes to risk scores depending on the model they are in.5
Incorporating more risk adjustment into TM would itself mitigate this concern since it would align incentives in both programs and thus “equalize” coding behavior. Other measures may address coding intensity that significantly differs between entities due to coding practices rather than differences in patient risk. For example, Congress could authorize tiered coding intensity adjustments within a statutory range based on risk score changes over time. Expanding risk adjustment data validation audits within and beyond MA, with extrapolated recollection of overpayments, would be another potential approach.6 Importantly, Congress should ensure that policies to combat coding intensity do not undermine the basic purpose of risk adjustment: ensuring that payers and providers do not avoid costlier patients.
Quality Measurement
There is some debate as to whether payment reforms are sufficient to ensure a high quality of care in addition to cost effectiveness. While quality information can potentially be useful, it should not come from CMS-developed metrics or a performance-based payment system.
Pay-for-performance in Medicare has failed to achieve higher outcomes across numerous payment systems. For clinicians, there has been voluminous evidence that the Merit-Based Incentive Payment System has increased provider burden without improving quality.7 Such failures have been due to both operational flaws – such as the development of excessively complex, numerous, and ineffective measures – as well as more fundamental shortcomings – such as the inherent fixation on “measurable” quality outcomes at the expense of less tangible ones and a limited ability to detect the underlying driver of those outcomes.
Furthermore, quality metrics prioritized by government agencies are unlikely to align with what patients value, since their preferences are diverse and ever-changing. Patients making well-informed decisions about where to receive care can be a better indicator of provider quality than measures designed through bureaucratic processes. Therefore, to the extent Congress articulates a role for CMS in quality measurement, it should be as a facilitator of more transparent information that patients and others can use to inform their care decisions. For example, CMS could collect and publish data about care outcomes and processes on platforms such as Medicare’s Care Compare or Plan Compare, or enable third parties to collect and present such data in a consumer-friendly manner.8
Technical Advisory Committee for Physician Fee Schedule Rates
You correctly identify the need to accurately determine payment rates in the Medicare Physician Fee Schedule (PFS). Misestimating the value of health care services reduces access to undervalued care, increases the cost of ineffective care to patients and taxpayers, and generally distorts the health care sector. You also rightly point out that CMS frequently defers to the American Medical Association Relative Value Scale Update Committee (RUC) despite evidence that its recommendations lead to such inaccuracies in the PFS.
Establishing a new process for determining relative values of PFS services separately from the RUC would address the perception that professional medical societies are influential in setting their own payment rates. However, enabling CMS to take a “more active role” in updating the PFS with the help of a new technical advisory committee would not address the methodological shortcomings inherent to centralized, administrative price-setting.
Other administrative FFS payment systems contain numerous instances of inaccurate pricing. In outpatient settings, Medicare payment policy has encouraged greater health care consolidation by paying hospitals at a significantly higher rate than physician offices for the same routine services.9 The Medicare Payment Advisory Commission (MedPAC) has also long recommended reducing the generous payment rates for post-acute care.10 There are certainly other, unknown cases of inaccurate payment rate determinations as well, and there is no reason to expect that CMS will be more capable to set PFS payment rates by virtue of a new committee. Since such a committee would not necessarily have better insights into the proper value of health care services than the RUC or another body, it seems more likely that it would serve as simply another forum for political or interest group considerations. An example of this dynamic is CMS recently mandating that MA plans appoint health equity experts to their (also newly required) utilization management committees.11
Prices are most accurately determined by the aggregate value that consumers and suppliers place on a good or service. This can be particularly true for routine, discretionary, and non-emergent health care services. The major flaw of the American health care system is that government policies exercise too strong an influence on private actors, which undermines the pressure to price services competitively and in accord with their actual value. Therefore, rather than adopting a new approach to administrative price setting, policymakers should instead explore ways to shift more financing to Medicare enrollees and enable more comparison of provider quality in order to empower them as consumers. Incorporating more data from the private health care market to inform the development of relative value units could also offer better empirical insights about actual consumer behavior versus administratively re-creating market prices from available data on input costs. Use of market data could also be designed in a way that encourages commercial payers to seek out greater efficiencies in their reimbursement of certain health care services rather than simply deferring to what Medicare has historically paid.
I appreciate your attention to these key issues in Medicare payment and the opportunity to offer my perspective.
Joe Albanese
Senior Policy Analyst
Paragon Health Institute
