Value-Based Payment, the Safety Net, and Rethinking Risk

Am J Manag Care. 2022;28(8):e280-e281. https://doi.org/10.37765/ajmc.2022.89200

Takeaway Points

• Safety-net providers need alternative payment models (APMs) that take into account the full set of risks that they face.
• Without considering the unique circumstances of safety-net providers, Medicaid managed care organizations may adopt APMs that inadvertently destabilize the very providers on which Medicaid members and states often depend.
• Population-based, prospective payments that are condition specific or defined scope of practice offer providers strong revenue stability and flexibility, combined with lower insurance, technical, and performance risks.
• Demonstrations of condition-specific and defined-scope-of-practice APMs can serve as stepping stones to safety-net providers successfully adopting APMs with greater risk.

In September 2020, CMS issued guidance designed to accelerate the adoption of value-based payment arrangements under Medicaid.¹ This guidance followed a decade of demonstration projects and a 2016 rule that explicitly allowed states to require their Medicaid managed care organizations (MCOs) to direct a portion of Medicaid payments to providers using alternative payment models (APMs). Some states had already begun using APMs in their managed care programs before this rule, and since its enactment more states have introduced these reimbursement models in their Medicaid programs. Currently, 26 states mandate some type of APM in their Medicaid managed care contracts, and 18 states have established specific APM spending targets for their Medicaid MCOs.²

More recently, the Biden administration announced a commitment to 3 goals for value-based payment (advancing health equity, improving quality, lowering costs)³ and strategies⁴ to extend more APMs to safety-net providers.CMS also expressed a strategic goal of increasing value-based models’ clarity and transparency.⁴ These objectives are laudable, but including more safety-net providers in APMs and promoting equity will require understanding the full range of financial and operational challenges that shape providers’ risk exposure. Payers and MCOs should more clearly understand the promise and the risks of different APMs as they engage safety-net providers in experimental payment reforms.

Because calculating the risk inherent in different types of APMs is a multidimensional task, many MCOs may not know precisely how much financial risk they are passing on through these contracts. Safety-net providers have slim operating margins and may be hesitant or unwilling to participate in APMs that involve financial risk. Moreover, APMs expose providers to insurance, technical, utilization, and performance risks.⁵ Without a way to consider the full range of vulnerabilities that affect Medicaid and especially safety-net providers, MCOs may adopt APMs that inadvertently destabilize the very providers on which Medicaid members and states often depend.

How might specific types of risk be relevant for safety-net providers?

Insurance risk refers to normal variation in utilization of medical services and differences across subpopulations and speaks to the numbers of members needed to smooth this risk. There is increasing recognition that social factors, such as housing status or income level, can influence the demand for medical services, leading to calls for payment models that account for social as well as clinical risks.⁶

Technical risk refers to how many structural elements a model has and the complexity of calculating allowed amounts (eg, based on index events, time periods, or exclusions). Models with high technical risk, such as shared-savings APMs, are complex to design, implement, and monitor. Accounting for social risk factors in payment models, although crucial, would add technical risk. CMS has recently called for making APM parameters and details more transparent and easily understandable to help reduce complexity and encourage adoption.⁴

Utilization risk refers to how changes in service utilization affect provider costs and profitability in a payment model. For example, in a fee-for-service (FFS) model, assuming payment rates cover costs, increased utilization leads to higher profit margins. But traditional FFS models are vulnerable to utilization risk, as many providers discovered when patients avoided or delayed care during the COVID-19 pandemic.

Performance risk refers to the potential for inefficiency, suboptimal quality, and high cost of care—criticisms that have been leveled at FFS medicine. Lack of technological savvy or the capacity to address social factors influencing patient health have also emerged as performance risks and may be exacerbated under population-based payments. Many safety-net providers are underresourced, making them especially vulnerable to these risks. Contracts with financial penalties for poor performance could further diminish quality among these Medicaid providers.

The Health Care Payment Learning & Action Network (LAN) APM framework outlines 4 major payment categories: (1) diagnosis-related groups and FFS payments with no link to quality and value; (2) FFS payments linked to quality and value; (3) retrospective performance-based payments built on FFS architecture; and (4) population-based, prospective payments.⁷ These last, so-called LAN category 4 APMs, offer payments that are up front, flexible, and predictable—qualities that support provider viability—but the individual APMs within the category differ in important ways. Some category 4 models, such as comprehensive population-based payments (category 4B), tend to have high risk across the board; integrated finance and delivery systems (category 4C) tend to have high insurance and utilization risk.

Our research suggests that only 1 APM can be designed to involve low to medium risk across all risk categories and provide strong provider flexibility and revenue stability: defined-scope-of-practiceor condition-specific payments (category 4A). These payments provide less risk across all categories than other population-based payment models. Examples of category 4A APMs include bundled or per-member per-month payments for primary care, mental health, oncology services, and substance use disorder treatment.⁷ In 2018, only 23% of Medicaid payments were funneled through APMs, and only 1.9% of Medicaid dollars were disbursed in the form of defined-scope-of-practice or condition-specific payments.^8,9 Despite being relatively underutilized, category 4A APMs seem to have several advantages over other APMs, particularly for safety-net providers.

At a time of change and uncertainty in health care delivery, flexibility and revenue stability are critical to provider stability—one of many lessons of the pandemic. Category 4A payments also offer safety-net providers more freedom to use services such as telehealth and to address nonmedical and social factors such as food and housing insecurity, which contribute to poor health and high health care costs. By providing a consistent revenue stream, population-based, prospective payments might also attract providers to work in specialties with persistent shortages, such as primary care and mental health.

CMS recognizes that relatively few of its APM demonstrations have focused on Medicaid or been relevant to safety-net providers or the populations they serve.⁴ The latest CMS innovation strategy calls on MCOs and other payers to consider many types of risk, beyond the financial, in establishing value-based payment arrangements. Instruments such as the APM Risk Evaluation Tool for Financial and Non-Financial Risk¹⁰ can help inform states, MCOs, and other payers as they evaluate various APM options and their appropriateness for safety-net providers. The APM assessment process can also inform where additional supports, learning collaboratives, technical assistance, or directed value-based payments¹¹ might be targeted.

At the same time, the CMS strategy calls for a more streamlined portfolio of models to deliver high-quality, person-centered care. In light of this commitment, there is room at the federal level to expand demonstration and evaluation of category 4A models across services such as primary care, mental health, and substance use disorder treatment. These might serve as stepping stones to safety-net providers successfully adopting APMs with greater risk.

At the state level, states and MCOs should take a rigorous approach to evaluating the appropriateness of the APMs they adopt, particularly for safety-net providers, and make future model selection decisions based on a wider understanding of the risks inherent in these models. Safety-net providers are in short supply and urgently needed to reduce health disparities. Careful model consideration and rigorous evaluation will reveal which APMs work best for attracting and sustaining safety-net providers. Given CMS’ declared intention to include equity in evaluation criteria for Medicaid demonstrations and other payment initiatives,¹² finding payment models that support providers should help advance equity in the communities they serve.

Author Affiliations: George Mason University (AEC), Fairfax, VA; Bailit Health (MBD), Needham, MA; Bowling Business Strategies (RM), Philadelphia, PA.

Source of Funding: None.

Author Disclosures: Dr Cuellar reports consultancies for Casey Family Foundation, grants received from the National Institute for Health Care Management and the National Institute for Mental Health, and honoraria from federal grant review panels. Ms Dyer is employed by Bailit Health, which has contracts with state Medicaid programs and managed care organizations; these funding sources had no role in the conceptualization, preparation, review, or approval of the manuscript. Ms Matulis has received honoraria from the National Council for Mental Wellbeing and the Rhode Island Primary Care Association, and she is employed by Bowling Business Strategies, which has contracts with state government agencies, government consulting firms, provider organizations, and associations; these funding sources had no role in the conceptualization, preparation, review, or approval of the manuscript.

Authorship Information: Concept and design (AEC, MBD, RM); drafting of the manuscript (AEC, MBD, RM); critical revision of the manuscript for important intellectual content (AEC, MBD, RM); administrative, technical, or logistic support (AEC); and supervision (AEC).

Address Correspondence to: Alison E. Cuellar, PhD, George Mason University, 4400 University Dr, Fairfax, VA 22030. Email: aevanscu@gmu.edu.

REFERENCES

1. Value-based care opportunities in Medicaid. CMS. September 15, 2020. Accessed July 2, 2021. https://www.medicaid.gov/Federal-Policy-Guidance/Downloads/smd20004.pdf

2. Medicaid managed care contracts as instruments of payment reform: a compendium of contracting strategies. Catalyst for Payment Reform. 2019. Accessed July 2, 2021. https://www.catalyze.org/wp-content/uploads/woocommerce_uploads/2019/12/Medicaid-MCOs-as-Agents-of-Payment-Reform-1.pdf

3. Brooks-LaSure C, Fowler E, Seshamani M, Tsai D. Innovation at the Centers for Medicare and Medicaid Services: a vision for the next 10 years. Health Affairs. August 12, 2021. Accessed August 14, 2021. doi:10.1377/forefront.20210812.211558

4. Innovation Center strategy refresh. CMS. 2021. Accessed November 12, 2021. https://innovation.cms.gov/strategic-direction-whitepaper

5. Spector JM, Studebaker B, Menges EJ. Provider payment arrangements, provider risk, and their relationship with the cost of health care. Society of Actuaries. October 2015. Accessed July 2, 2021. https://www.soa.org/globalassets/assets/Files/Research/Projects/research-2015-10-provider-payment-report.pdf

6. Nerenz DR, Austin JM, Deutscher D, et al. Adjusting quality measures for social risk factors can promote equity in health care. Health Aff (Millwood). 2021;40(4):637-644. doi:10.1377/hlthaff.2020.01764

7. Alternative payment model: APM framework. Health Care Payment Learning & Action Network. 2017. Accessed July 2, 2021. https://hcp-lan.org/workproducts/apm-refresh-whitepaper-final.pdf

8. APM Measurement: Progress of Alternative Payment Models. 2019 Methodology and Results Report. Health Care Payment Learning & Action Network. 2019. Accessed July 2, 2021. http://hcp-lan.org/workproducts/apm-methodology-2019.pdf

9. APM measurement effort. Health Care Payment Learning & Action Network. 2019. Accessed July 2, 2021. https://hcp-lan.org/workproducts/apm-infographic-2019.pdf

10. Matulis R, Dyer MB, Cuellar AE. Alternative payment models: risk evaluation tool for financial and non-financial risk. George Mason University. 2021. Accessed November 1, 2021. https://hap.gmu.edu/health-policy-issue-briefs

11. Directed payments in Medicaid managed care. Medicaid and CHIP Payment and Access Commission. June 2022. Accessed July 7, 2022. https://www.macpac.gov/publication/directed-payments-in-medicaid-managed-care/

12. Paving the way to equity: a progress report. CMS. January 2021. Accessed December 13, 2021. https://www.cms.gov/files/document/paving-way-equity-cms-omh-progress-report.pdf