Key Takeaways

With 69% of Medicaid beneficiaries enrolled in comprehensive managed care plans, plans play a critical role in responding to the COVID-19 pandemic and in the fiscal implications for states. Given unanticipated costs related to COVID-19 testing and treatment, as well as depressed utilization affecting the financial stability of many Medicaid providers, states are currently evaluating options to adjust current managed care organization (MCO) payment rates and/or risk sharing mechanisms as well as evaluating options and flexibilities under existing managed care rules to direct payments to Medicaid providers (Figure 1). This brief provides an overview of how MCO capitation rates are developed by states and approved by CMS, highlights options available to states to adjust current MCO payment rates and/or risk sharing mechanisms, describes how MCOs pay providers, and outlines state options to direct MCO payments to providers in response to conditions created by the pandemic. Key takeaways are discussed below.

Under federal law, payments to Medicaid managed care organizations (MCOs) must be actuarially sound. Actuarial soundness means that “the capitation rates are projected to provide for all reasonable, appropriate, and attainable costs that are required under the terms of the contract and for the operation of the managed care plan for the time period and the population covered under the terms of the contract.” Unlike fee-for-service (FFS), capitation provides upfront fixed payments to plans for expected utilization of covered services, administrative costs, and profit.

Under existing Medicaid managed care authority, states have several options to address payment issues that have arisen as a direct result of the COVID-19 pandemic. CMS has outlined state options to modify managed care contracts and rates in response to COVID-19 including risk mitigation strategies, adjusting capitation rates, covering COVID-19 costs on a non-risk basis, and carving out costs related to COVID-19 from MCO contracts. These options vary widely in terms of implementation/operational complexity, and all options will require CMS approval.

States can direct that managed care plans make payments to their network providers using methodologies approved by CMS to further state goals and priorities, including COVID-19 response. This strategy can address the scenario in which states are making capitation payments to plans, but providers are not receiving reimbursement from plans due to decreased service utilization while non-urgent services are suspended or patients are hesitant to seek care. For example, states could require plans to adopt a uniform temporary increase in per-service provider payment amounts for services covered under the managed care contract, or states could combine different state directed payments to temporarily increase provider payments.

Key considerations in evaluating payment options include: mitigating MCO and state risk, MCO and provider cash flow, beneficiary access and continuity of care, and administrative burden. States may consider how payment policy options impact both MCO and state financial risk, as capitation rates do not include costs associated with COVID-19 but the pandemic has also led to decreased utilization of services. States may also consider MCO cash flow issues that may arise due to unanticipated costs associated with COVID-19 as well as provider cash flow issues as a result of depressed utilization. It will also be important to consider the impact of payment options on beneficiary access and continuity of care (i.e., will beneficiaries be able to maintain relationships with existing providers). Finally, all options require CMS review and approval but options may vary widely in terms of implementation/operational complexity.

Introduction

Today, capitated managed care is the dominant way in which states deliver services to Medicaid enrollees. States pay Medicaid managed care organizations (MCOs) a set per member per month payment for the Medicaid services specified in their contracts. Current MCO capitation rates were developed and implemented prior to the onset of the COVID-19 pandemic. Consequently, these rates do not include costs for COVID-19 testing and treatment. At the same time, utilization of non-urgent care is decreasing as individuals seek to limit risks/exposure to contracting the coronavirus. As a result, many states are currently evaluating options for making adjustments to existing MCO rates and risk sharing mechanisms in response to unanticipated COVID-19 costs and conditions that have led to decreased utilization. However, there is still a lot of uncertainty about the impact of the pandemic on managed care financing/rates, particularly in current plan rating periods (which typically run on a calendar year or state fiscal year basis), as it’s still too early to know if there is pent up demand that may drive utilization upward as the pandemic abates as well as unknowns about where/when COVID-19 cases and related costs will spike as the pandemic continues.

Additionally, many Medicaid providers may be under fiscal strain, facing substantial losses in revenue.¹  For providers in states that rely heavily on managed care, states are making payments to plans but those funds may not be flowing to providers where utilization has decreased. In contrast, many health insurance companies are reporting record earnings during the pandemic.²  As a result, states are also evaluating options and flexibilities under existing managed care rules to direct/bolster payments to Medicaid providers.

This brief provides an overview of how MCO capitation rates are developed by states and approved by CMS, highlights options available to states to adjust current MCO payment rates and/or risk sharing mechanisms, describes how MCOs pay providers, and outlines state options to direct MCO payments to providers in response to conditions created by the pandemic.

Background

As of July 2019, 40 states, including DC, contract with comprehensive, risk-based managed care plans to provide care to at least some of their Medicaid beneficiaries. Medicaid managed care organizations (MCOs) provide comprehensive acute care (i.e., most physician and hospital services) and in some cases long-term services and supports to Medicaid beneficiaries. MCOs accept a set per member per month payment for these services and are at financial risk for the Medicaid services specified in their contracts.

As of July 2018, almost 54 million Medicaid enrollees received their care through risk-based MCOs – or over two thirds (69%) of all Medicaid beneficiaries. Twenty-five MCO states covered more than 75% of Medicaid beneficiaries in MCOs. In FY 2018, state and federal spending on Medicaid services totaled nearly $593 billion. Payments made to MCOs accounted for about 45% of total Medicaid spending. State-to-state variation in MCO spending reflects many factors, including the proportion of the state Medicaid population enrolled in MCOs, the health profile of the Medicaid population, whether high-risk/high cost beneficiaries (e.g., persons with disabilities, dual eligible beneficiaries) are included in or excluded from MCO enrollment, and whether or not long-term services and supports are included in MCO contracts. Six firms – UnitedHealth Group, Centene, Anthem, Molina, Aetna, and Wellcare – accounted for over 47% of all Medicaid MCO enrollment in July 2018.

How do states set payment rates for MCOs?

Under federal law, payments to Medicaid managed care organizations (MCOs) must be actuarially sound.³  Actuarial soundness means that “the capitation rates are projected to provide for all reasonable, appropriate, and attainable costs that are required under the terms of the contract and for the operation of the managed care plan for the time period and the population covered under the terms of the contract.” The 2016 final rule on Medicaid managed care significantly strengthened the standards that states must meet in developing actuarially sound capitation rates and that CMS will apply in its review and approval of rates. Payments made to Medicaid managed care plans vary depending on the scope of services and populations covered by the plan. Unlike fee-for-service, capitation provides upfront fixed payments to plans for expected utilization of covered services, administrative costs, and profit.

In developing actuarially sound rates, states must follow accepted actuarial methods and specific federal requirements outlined in regulations and other guidance. Plan rates, usually for a 12-month rating period, are set using baseline utilization and cost data based on historical FFS claims, health plan services and utilization data (i.e., encounter data), and/or health plan financial data for the populations enrolled. Baseline spending data is trended forward to determine per member per month payment amounts and must take into account/adjust for factors such as medical cost inflation, expected changes in utilization, and state Medicaid program changes (e.g., changes to eligibility, benefits, cost-sharing, FFS payment rate changes (if state bases managed care rates on FFS rates)). Different rates are set for population subgroups (referred to as “rate cells”) taking into account eligibility category, age, gender, location, among other factors. Rates also include expected non-benefit costs including administration, taxes, licensing and regulatory fees, contribution to reserves, risk margin, and cost of capital.^4 ,5 ,6 

States may use a variety of other mechanisms to adjust plan risk (including catastrophic claims), incentivize plan performance, and ensure payments are not too high or too low, including:

• Risk & Acuity Adjustments – Rates can be risk adjusted to account for the health status (or other demographic factors) of enrollees which may reduce the incentive for plans to avoid sicker members. Acuity adjustments are applied to total payments across all managed care plans to account for significant uncertainty about the health status or risk of a population.
• Risk Sharing Arrangements– Rates may take into consideration the use of plan risk sharing mechanisms including risk corridors, stop-loss, or reinsurance. Under risk corridor arrangements, states and plans agree to share profit or losses (at percentages specified in plan contracts) if aggregate spending falls above or below specified thresholds (two-sided risk corridor). Stop-loss and reinsurance arrangements protect plans from losses beyond a specified threshold.
• Medical Loss Ratio (MLR) – The MLR reflects the proportion of total capitation payments received by an MCO spent on clinical services and quality improvement (where the remainder goes to administrative costs and profit). CMS published a final rule in 2016 that requires states to develop capitation rates for Medicaid to achieve an MLR of at least 85% in the rate year. Contracts must include a requirement for plans to calculate and report an MLR. There is no federal requirement for Medicaid plans to pay remittances to the state if they fail to meet the MLR standard but states have discretion to require remittances – 24 states reported that they “always” require MCOs to pay remittances while six states indicated they “sometimes” require MCOs to pay remittances, as of July 1, 2019.⁷ 
• Incentive & Withhold Arrangements – States may factor payment mechanisms like incentive and withhold arrangements into rate development. Using incentive arrangements, states may make payments over and above capitation rates to plans for meeting specified performance targets. Under withhold arrangements, states may hold back a portion of capitation rates to be paid if/when plans meet specified performance targets (e.g., quality performance measures or quality-based outcomes).

After CMS approval, states may increase or decrease rates by 1.5% (per rate cell) without requiring new approval. CMS must review and approve capitation rates including supporting data and documentation for a 12-month rating period. States must update capitation rates and seek approval from CMS each year before new rates become effective. States must obtain federal approval for adjustments that exceed 1.5%.

What guidance has CMS provided to states to address MCO payment issues in response to COVID-19?

Under existing Medicaid managed care authority, states have several options to address payment issues that have arisen as a direct result of the COVID-19 pandemic. CMS has acknowledged that costs associated with the COVID-19 pandemic could not have been reasonably prospectively included in the development of current MCO rates. CMS has also noted that the COVID-19 public health emergency is causing major shifts in utilization across the healthcare industry, causing uncertainty for health care providers. In response, CMS has outlined state options to modify current managed care contracts and rates in response to COVID-19.⁸  CMS discussed several options available to states and outlined key considerations in evaluating these options including: mitigating MCO and state risk, MCO cash flow, beneficiary continuity of care, and administrative burden (for states, the federal government, and MCOs). The following options described below will all require CMS approval. States will need to submit MCO contract amendments and in most instances revised actuarial certifications. (Also see box below for specific examples discussed on CMS stakeholder call related to options for covering COVID-19 tests.)

States can implement two-sided risk corridors. Most states have experience with risk corridors. Risk corridors can mitigate MCO risk without impacting enrollee continuity of care, as beneficiaries would be able to continue to see existing providers in their plan network. CMS advises states should implement risk corridors for all medical costs, not just COVID costs, noting it is simpler to implement for all medical costs and doing so accounts for risks related to non-COVID costs changes. Risk corridors provide financial protection to MCOs and limits on financial risk to states but would not address immediate MCO cash flow issues, as they would be reconciled and paid out at the end of the contract period. In its May 2020 guidance, CMS notes states could implement two-sided risk corridor based on a target MLR – where a plan and the state would be required to share in gains or losses if the plan did not meet the target MLR within a specified margin. CMS will consider state requests to retroactively amend or implement (for current rating periods) risk mitigation strategies (e.g., risk corridors) only for the purpose of responding to the COVID-19 pandemic.^9 ,10 

States can adjust capitation rates. Risk corridors can be combined with a capitation rate adjustment to address MCO cash flow risks. Many states are implementing temporary increases in Medicaid FFS provider payment rates as part of disaster State Plan Amendments (SPAs). Some states have CMS approved state directed payments which contractually require managed care plans to adopt Medicaid FFS provider rates for specific provider types or services. States can make rate adjustments in response to COVID-19 that result in an increase or decrease to the capitation rate per rate cell of less than 1.5% with a contract amendment (but do not need a revised actuarial certification). For adjustments of more than 1.5% per rate cell, states must submit a revised actuarial rate certification and contract amendment to CMS. States can submit prospective or retrospective rate amendments (for the current rating period) associated with COVID-19 (or adjust for non-COVID-19 costs) but CMS has noted that a significant amount of uncertainty still exists around the spread of the pandemic and costs associated with treatment, which would make it very difficult to develop assumptions or assess the reasonableness of final rates/rate adjustments.

States can incorporate supplemental kick payments. Kick payments are one-time fixed, supplemental payments made to plans, allowing them to cover certain services (e.g., maternity care) without assuming financial risk for their use. States could use kick payments to cover COVID-related costs, which would require a contract amendment and rate certification.

States can cover COVID-19 costs on a non-risk basis. This could include all COVID-19 related service costs or all service costs for beneficiaries with a COVID-19 diagnosis. States would then reimburse MCOs for these costs net of capitation payments paid. CMS noted that this option could also be combined with a risk corridor to reduce the risk that remaining costs are significantly lower than originally projected. Covering costs on a non-risk basis would eliminate MCO risk, reduce MCO cash flow problems, and ensure continuity of care for beneficiaries. However, the success of the non-risk model would depend on accurately identifying relevant costs/enrollees.

States can carve-out costs related to COVID-19 and cover them on a FFS basis. States could either carve out all COVID-19 related service costs or all service costs for beneficiaries with a COVID-19 diagnosis. Again, CMS would recommend implementing with a risk corridor. A COVID-19 carve-out would be administratively burdensome and would disrupt beneficiary continuity of care, especially in states that heavily use managed care that don’t have much of a FFS provider network to utilize. This model would also depend on accurately identifying all COVID-19 costs and/or beneficiaries.

States may also consider adjusting their managed care contract quality measurement requirements. States may need to revisit contract provisions that have been affected by pandemic in ways that were not anticipated. CMS has indicated that the COVID-19 pandemic is likely to affect clinical practices and timely reporting of quality data. States may need to reexamine arrangements tied to performance metrics/reporting requirements such as withhold and incentive arrangements, state-directed payments, as well as other contract requirements and penalties. Depending on the nature of the changes, rate certification amendments may or may not be needed; however, states will need to submit contract amendments to reflect any revisions to these provisions.

How do MCOs pay providers?

States generally pay the plans a capitation payment, but then plans determine how to pay the providers in their network.¹²  Plans generally have wide latitude to determine how to pay their contracted providers. Medicaid MCOs may pay the providers in their networks on a FFS basis, capitation basis, or on other terms. Although plans may use alternative provider payment models (e.g., capitation, bundled payments etc.) for some providers, MCOs still widely use FFS reimbursements to pay providers.

Under current MCO rules, states are prohibited from directing how a managed care plan pays its providers except for certain payment methodologies that have been approved and reviewed by CMS. States may require MCOs to adopt minimum or maximum provider payment fee schedules or provide uniform dollar or percentage increases for network providers that provide a particular service under the contract, as approved by CMS. States also can seek CMS approval to require MCOs to implement value-based purchasing models for provider reimbursement (e.g., pay for performance, bundled payments) or participate in multi-payer or Medicaid-specific delivery system reform or performance improvement initiatives. State directed payments must be based on utilization and delivery of services covered under the managed care plan contract and must be reflected in capitation rate development and certification.¹³ 

What guidance has CMS issued related to state options to direct MCO provider payments in response to conditions created by the COVID-19 pandemic?

States can direct that managed care plans make payments to their network providers using methodologies approved by CMS to further state goals and priorities, including COVID-19 response. This strategy can address the scenario in which states are making capitation payments to plans, but providers are not receiving reimbursement from plans due to decreased service utilization while social distancing measures are in place and non-urgent services are suspended. For example, states could require plans to adopt a uniform temporary increase in per-service provider payment amounts for services covered under the managed care contract, or states could combine different state directed payments to temporarily increase provider payments, according to recent CMS guidance. Specific examples from the CMS guidance are outlined in the box below. CMS will allow states to develop and implement these specific state directed payments retrospectively to the start of the current contract rating period.

CMS explains that state directed increased payments for actual utilization of services can preserve the availability of covered services for enrollees during a time when providers may be experiencing dramatic utilization declines or incurring additional costs due to the public health emergency. The guidance also says that states may use directed payments to address increased use of telehealth or other approaches to maintain access to care for all enrollees or specific subgroups with specialized needs during the emergency. States must direct payments to a class of providers, such as dental, behavioral health, home health and personal care, pediatric, federally-qualified health centers, or safety-net hospitals, to support providers that may serve a high proportion of Medicaid enrollees and may be disproportionately affected by the public health emergency. Directed payments must be appropriate and reasonable compared to the total payments the provider would have received in the absence of the public health emergency.¹⁵  For states that have approved directed payment proposals, CMS guidance says that states wishing to make changes to such arrangements in light of COVID-19 can submit an amended directed payment preprint and/or contract and rate certification amendments to CMS.

To the extent that home and community-based services (HCBS) are included in MCO contracts, states may contractually require MCOs to make retainer payments to allow certain HCBS providers to continue to bill for individuals enrolled in Medicaid even if HCBS services (e.g., habilitation and personal care) cannot be provided during a public health emergency. The retainer payments must be authorized as part of the Section 1915(c) HCBS waiver, Section 1115 demonstration waiver, or other Medicaid authority. To effectuate these payments, states must submit a directed payment preprint to CMS for approval.

Looking Ahead: What to Watch

With 69% of beneficiaries enrolled in comprehensive Medicaid managed care plans, plans play a critical role in responding to the COVID-19 pandemic. Given unanticipated costs related to COVID-19 testing and treatment as well as depressed utilization affecting the financial stability of many Medicaid providers, states are currently evaluating options to adjust current MCO payment rates and/or risk sharing mechanisms as well as evaluating options and flexibilities under existing managed care rules to direct MCO payments to Medicaid providers. Key considerations in evaluating these options include:

• Does the policy mitigate risks to MCO and states? Some MCO payment options like rate adjustments, kick-payments, or covering costs on a non-risk basis may mitigate risks to MCOs. Other options like rate adjustments and the use of two-sided risk corridors may mitigate both risks to MCOs and states. However, states may need to consider where they are in their rate cycle, as money may not be returned to states, or additional funding provided to MCOs, until after the end of the rating period (when options like risk corridors or retrospective rate adjustments are used). Going into the next contract period, states will face continued challenges in setting rates as uncertainty about new costs and utilization will remain. Additionally, will CMS revisit some proposals in its Proposed Medicaid Managed Care Rule like the current provision which would prohibit states from retroactively adding or modifying risk sharing mechanisms after the start of the rating period?
• Does the option address MCO or provider cash flow issues? States may consider the fiscal stability/viability of both MCOs and providers and the timing of when additional funding might be paid out. Some policy options, like HCBS retainer payments, are limited to a 90-day cap per individual. States and providers may need to consider alternative options when this period ends.
• Does the policy enhance enrollee access or continuity of care for enrollees? States may consider the impact of payment policy options on beneficiary continuity of care, as some options – like carving out services to FFS systems – may disrupt established beneficiary-provider relationships. States considering directed payments could consider how policy options may enhance access to care for enrollees (including access to telehealth services).
• How feasible are the options given uncertainty about COVID-19 costs and the spread of the pandemic? And, how administratively burdensome are the options for states, the federal government, and plans? Both retroactive and prospective rate adjustments may be extraordinarily difficult to develop and implement because of significant uncertainty related to COVID-19 costs and utilization. States may consider what is required for approval – contract amendments, actuarial rate certification amendments, and/or state directed payment pre-prints and whether options under consideration may meet criteria for expedited review at CMS as well as the implementation/operationalization complexity and how quickly they may be able to get new monies out to plans and providers.

Looking to the future, the duration of the temporary increase in the federal Medicaid match rate as well as overall state fiscal conditions will also be factors states must consider when evaluating policy options in response to the COVID-19 pandemic. Reduced state revenues and projected state budget shortfalls will likely put pressure on state Medicaid programs and states’ ability to pay for any policies that increase state costs, particularly without additional federal support or certainty about the duration of the current enhanced federal funding. States will want to carefully review options to mitigate risks related to overpayments to MCOs in a time of heightened uncertainty and state fiscal constraints. Due to significant challenges around revising rates for current contract periods (either retrospectively or prospectively), CMS is likely to encourage states to include two-sided risk corridors as part of state strategies to guard against overpayment.

1. The Coronavirus Aid, Relief, and Economic Security (CARES) Act and the Paycheck Protection Program and Health Care Enhancement Act provide $175 billion in provider relief funds to reimburse eligible health care providers for health care related expenses or lost revenues that are attributable to coronavirus. HHS has allocated $15 billion to Medicaid providers however, there have been some delays and challenges in applying for these funds and the allocation may not be sufficient to remedy the fiscal issues faced by some providers. ↩︎
2. Abelson, Reed. “Major U.S. Health Insurers Report Big Profits, Benefiting from the Pandemic.” The New York Times. Aug. 5, 2020. https://www.nytimes.com/2020/08/05/health/covid-insurance-profits.html (accessed Aug. 12, 2020). ↩︎
3. These requirements apply to comprehensive risk-based plans as well as limited-benefit plans (e.g., those providing only dental or behavioral health services). ↩︎
4. MACPAC, Setting Per Capita Caps: Significant differences between current methods and those anticipated under financing reforms, March 2017, https://www.macpac.gov/wp-content/uploads/2017/03/Setting-Per-Capita-Caps.pdf. ↩︎
5. MACPAC, Payment Policy in Medicaid Managed Care, June 2011, https://www.macpac.gov/wp-content/uploads/2011/06/Payment-Policy-in-Medicaid-Managed-Care.pdf. ↩︎
6. Centers for Medicare & Medicaid Services (CMS), “2020-2021 Medicaid Managed Care Rate Development Guide For Rating Periods Starting between July 1, 2020 and June 30, 20211,” July 2020, https://www.medicaid.gov/medicaid/managed-care/downloads/2020-2021-medicaid-rate-guide.pdf. ↩︎
7. Kathleen Gifford, et al., A View from the States: Key Medicaid Policy Changes: Results from a 50-State Medicaid Budget Survey for State Fiscal Years 2019 and 2020 (Washington, DC, Kaiser Family Foundation (KFF), Oct. 2019), https://modern.kff.org/medicaid/report/a-view-from-the-states-key-medicaid-policy-changes-results-from-a-50-state-medicaid-budget-survey-for-state-fiscal-years-2019-and-2020/. ↩︎
8. CMS Coronavirus COVID-19 Stakeholder Calls. Friday, April 10, 2020 CMS Medicaid and CHIP All State Call, https://www.cms.gov/Outreach-and-Education/Outreach/OpenDoorForums/PodcastAndTranscripts. ↩︎
9. Although the proposed rule at CMS would prohibit states from implementing retroactive risk mitigation strategies, given the unique and unanticipated circumstances presented by the COVID-19 pandemic CMS will consider state requests to retroactively amend or implement risk mitigation strategies (e.g., risk corridors) only for the purpose of responding to the COVID-19 pandemic ↩︎
10. CMS, COVID-19 Frequently Asked Questions (FAQs) for State Medicaid and Children’s Health Insurance Program (CHIP) Agencies (last updated June 30, 2020), https://www.medicaid.gov/state-resource-center/downloads/covid-19-faqs.pdf (pgs. 83-84 V.C.6). ↩︎
11. CMS Coronavirus COVID-19 Stakeholder Calls. Friday, March 27, 2020, CMS Medicaid & CHIP All State Call, https://www.cms.gov/Outreach-and-Education/Outreach/OpenDoorForums/PodcastAndTranscripts. ↩︎
12. Some plans may include certain services in their contract with the state (e.g., pharmacy, NEMT, dental) but may subcontract these services to other entities. ↩︎
13. The proposed MCO rule pending at CMS would make some changes to minimum fee schedule arrangements for directed payments. ↩︎
14. CMCS Informational Bulletin, May 14, 2020, “Medicaid Managed Care Options in Responding to COVID-19,” https://www.medicaid.gov/sites/default/files/Federal-Policy-Guidance/Downloads/cib051420.pdf. ↩︎
15. CMS will require the implementation of a two-sided risk corridor when states implement state-directed payments intended to mitigate impact of the public health emergency. ↩︎

The COVID-19 pandemic has led to dramatic decreases in health care spending, as patients and providers have delayed a wide range of health care services. The decrease in service use and spending resulted in a decline in revenue for many providers at the same time that some are facing increased costs due to the pandemic. Given the uncertain timing of a “return to normal” and potentially lingering effects of the current economic crisis, some providers may continue to experience sustained declines in revenue even with the federal assistance that has been made available.¹ 

Depending on the severity and duration of revenue loss, some hospitals and physician practices may find it difficult to operate independently, which could increase the rate of consolidation among health care providers. Lower margins among some providers may create new opportunities for large chains to acquire smaller providers. The Coronavirus Aid, Relief, and Economic Security (CARES) Act and the Paycheck Protection Program and Health Care Enhancement Act allocated $175 billion for grants to providers that were partly intended to help make up for revenue lost due to coronavirus, but analysis shows that the first $50 billion in grants were not targeted to providers most vulnerable to revenue losses.²  Another $13 billion was subsequently targeted to safety net hospitals and $11 billion has been targeted to rural providers.³  However, it is not clear whether this infusion of funds plus other government loans—including those from the Paycheck Protection Program—will be sufficient to stabilize providers who are least equipped to weather this revenue decline. Even if sufficient government assistance is provided, the disruption of the COVID-19 pandemic may make operating independently seem less attractive and riskier to some smaller providers. Therefore, financial assistance to providers may not be sufficient to prevent an increase in the pace of consolidation.

This brief provides an overview of existing research that examines the impact of provider consolidation on health care costs and quality. There are two major types of consolidation among health care providers, both of which are discussed in this brief. The first is horizontal consolidation, which occurs when two providers performing similar functions join, such as when two hospitals merge or groups of physician practices merge to form larger group practices. The second type is vertical integration, which refers to one type of entity purchasing another in the supply chain such as hospitals acquiring physician practices.⁴ 

Provider consolidation leads to higher prices

A wide body of research has shown that provider consolidation leads to higher health care prices for private insurance; this is true for both horizontal and vertical consolidation. In Medicare, payment policies protect Medicare from increased prices due to horizontal consolidation but have led to higher Medicare costs in the case of vertical consolidation. However, recent administrative and legislative changes are bringing Medicare reimbursement at hospital outpatient departments in line with reimbursement at independent physicians’ offices.

Horizontal consolidation among hospitals

In 2020, the Medicare Payment Advisory Commission (MedPAC) reviewed the published research on hospital consolidation and concluded that the “preponderance of evidence suggests that hospital consolidation leads to higher prices.”⁵  For example, one analysis looking at 25 metropolitan areas with the highest rates of hospital consolidation from 2010 through 2013 found that the price private insurance paid for the average hospital stay increased in most areas between 11% and 54% in the subsequent years.⁶  A separate analysis of data from employer-sponsored coverage found that hospitals that do not have any competitors within a 15-mile radius have prices that are 12% higher than hospitals in markets with four or more competitors.⁷  Another analysis of all hospital mergers over a five year period found that mergers of two hospitals within five miles of one another resulted in an average price increase of 6.2% and that price increases continued in the two years after a merger.⁸  A similar study found that mergers of two hospitals in the same state led to price increases of 7% to 9% for the acquiring hospitals.⁹  Studies have found that these patterns hold even when looking specifically at non-profit hospitals.¹⁰  While health plans may try to keep hospital prices low, health plans’ ability to successfully hold down prices is limited in many parts of the country because they have less market power than hospitals.¹¹ 

Even when a hospital merges with a hospital in a different geographic area, some studies suggest that the merger can impact competition and prices. One analysis found that prices at hospitals acquired by out‐of‐market hospital systems increase by about 17% more than unacquired, stand‐alone hospitals.¹²  This study also found that this type of merger has a spillover impact on market dynamics in the area where the acquired hospital is located— with prices of nearby competitors to acquired hospitals increasing by around 8%.¹³  One reason that prices rise when there are hospital mergers across markets is that they increase hospital bargaining positions with insurers, which seek to have strong provider networks across multiple areas in order to attract employers with employees in multiple locations.¹⁴  Additionally, large hospital systems can influence the dynamics of negotiations with insurers and shift volume to higher cost facilities. For example, hospital systems may require that insurers include all hospitals in their system in a provider network if the insurer wants any hospitals included. This can lead to higher cost hospitals being in a provider network even when there are lower cost hospitals nearby. In one recent anti-trust case, the Sutter Health system was accused of violating California’s antitrust laws by using its market power to illegally drive up prices.¹⁵  In a 2019 settlement, Sutter Health agreed to stop requiring that all of its hospitals be included in an insurer’s network and also agreed to pay damages and make other changes.¹⁶ 

Horizontal consolidation among physicians

Patterns of consolidation leading to higher prices also have been observed when physician practices merge. A national study found that physicians in the most concentrated markets charged fees that were 14% to 30% higher than the fees charged in the least concentrated markets.¹⁷  Another national study comparing physician prices in counties with highly concentrated physician markets to counties with the least concentrated physician markets found higher prices for physicians practicing in the most concentrated counties across specialty types.¹⁸  A study that examined the effects of a merger of six orthopedic groups in Pennsylvania found that the merger was associated with price increases ranging from 15% to 25% across payers.¹⁹ 

Vertical consolidation

Vertical consolidation also leads to higher prices, which can then lead to higher premiums. One study analyzing highly concentrated hospital markets in California found that an increase in the share of physicians in practices owned by a hospital was associated with a 12% increase in premiums for private plans sold in the state’s Marketplace.²⁰  Another study that used private insurer data found that an increase in physician-hospital integration was associated with an average price increase of 14% for the same service.²¹  Those findings are consistent with another study that used private insurance data that found a large increase in physician-hospital vertical integration was associated with an increase in outpatient prices.²²  A study looking at Medicare beneficiaries’ patterns of health care utilization found that “patients are more likely to choose a high-cost, low-quality hospital when their physician is owned by that hospital.”²³ 

Insurance markets and consolidation

When insurance markets become more consolidated, there are two distinct impacts. As insurance companies consolidate and have more market power, evidence suggests that they are able to obtain lower prices from providers. For example, one study looking at the impact of health plan concentration on hospital prices found that hospital prices in the most concentrated health plan markets were approximately 12% lower than in more competitive health plan markets.²⁴  Another study found a similar pattern for both hospitals and some types of specialists.²⁵  However, these lower prices do not necessarily lead to lower premiums. A national study found that lower provider prices only translate into lower premiums if the insurance market is sufficiently competitive; where health insurance markets are more concentrated, premiums tend to be higher.²⁶  The impact on premiums may be somewhat mitigated for fully insured plans by the minimum loss ratio requirement in the Affordable Care Act, which limits the amount of the premium that insurers can keep.

Consolidation and Medicare prices

Private insurance rates are the result of negotiations between providers and payers, which means providers with market power due to consolidation have greater leverage to raise prices in these negotiations. In contrast, Medicare prices are set by formulas and government policies. Horizontal consolidation does not impact Medicare prices for physicians or hospitals that are generally paid based on the prospective payment systems.²⁷  However, once a physician’s office has been purchased by a hospital, that hospital had historically been able to obtain higher Medicare rates by billing as a hospital outpatient department for services at that physician’s location.

Both Congress and the Department of Health & Human Services (HHS) have made policy changes over the past several years to lower costs for off-campus hospital outpatient clinics to bring them in line with physicians’ offices, despite industry opposition. The Bipartisan Budget Act of 2015 (BBA) required that Medicare reimburse for services delivered at new, off-campus hospital outpatient departments using rates based on the physician fee schedule instead of the higher rates for outpatient hospital departments. However, this change grandfathers off-campus outpatient departments that billed for services, rendered services, or were being constructed before November 2, 2015. Beginning in 2019, the Centers for Medicare & Medicaid Services (CMS) lowered payment rates in grandfathered off-campus departments for a clinic visit—the single highest volume service provided by hospital outpatient departments—to 70% of the full hospital outpatient rate in 2019 and 40% of the full hospital outpatient rate in 2020. Several hospital associations challenged CMS’ authority for this policy. On September 17, 2019, the DC District Court vacated CMS’s regulation for being inconsistent with the statute. On July 17, 2020, the DC Circuit Court reversed the District Court’s decision, allowing the regulation to stay in place.

HHS’s regulatory change to lower payments at hospital outpatient departments is consistent with MedPAC’s recommendation to adjust Medicare payments so that those locations are reimbursed at the same rates as physician’s offices.²⁸  While HHS’s change does not directly impact Medicare Advantage plans, there is some evidence that Medicare Advantage plans typically pay rates that are similar to payments under traditional Medicare.

Mergers have led to more consolidation, even before the financial pressures brought on by COVID-19

Between 2010 and 2017, there were 778 hospital mergers.²⁹  Over time, the number of independent hospitals has declined as a result of these mergers, while the number of hospitals that are part of larger systems has risen (Figure 1). By 2017, two thirds (66%) of all hospitals were part of a larger system, as compared to 53% in 2005.³⁰ 

In 2010, most hospital markets were already dominated by a limited number of health systems: on average, the three largest health systems in a given area accounted for more than three-quarters of admissions.³¹  In the subsequent years, health care markets have continued to become more consolidated, as measured using the Herfindahl–Hirschman Index. This index is a commonly used measure of market concentration that is calculated for a given market based on the number of competing providers and each of these providers’ relative market share. From 2010 to 2016, the mean Herfindahl-Hirschman Index for metropolitan statistical areas in the United States for hospitals and specialist physician organizations each increased by about 5% on average.³²  Over the same period, the Herfindahl-Hirschman Index for primary care practices increased by 29% on average in metropolitan statistical areas nationwide.³³  Using this index, 90% of metropolitan statistical areas were highly concentrated for hospitals, 65% were highly concentrated for specialists and 39% were highly concentrated for primary care physicians by 2016.³⁴ 

Much of the increase in consolidation among physician practices is due to acquisition by hospitals. The proportion of primary care physicians practicing in organizations owned by a hospital or health system grew from 28% in 2010 to 44% in 2016.³⁵  By 2018, data from the American Medical Association shows that 35% of all practicing physicians worked either directly for a hospital or in a practice at least partly owned by a hospital in 2018.³⁶ 

Among health insurance markets, 57% of metropolitan statistical areas were highly concentrated in 2016 for private insurance, and the average Herfindahl-Hirschman Index for insurers was relatively steady between 2010 and 2016.³⁷  Meanwhile, the market for the private Medicare Advantage plans available to Medicare beneficiaries has become increasingly concentrated. Medicare Advantage plans are mainly health maintenance organizations (HMOs) and preferred provider organizations (PPOs) and receive payments from Medicare to cover Medicare enrollees. The total market share of the top four Medicare Advantage insurers increased from 48% in 2011 to 61% in 2015.³⁸  As of 2020, the top four Medicare Advantage insurers controlled 70% of the market.³⁹ 

The role of private equity

Private equity has started to play a role in this consolidation in recent years. These firms typically invest in businesses by taking a majority stake with the goal of increasing the value of the business and potentially selling it at a profit. One study found that private equity firms acquired 355 physician practices (1,426 sites and 5,714 physicians) from 2013 to 2016.⁴⁰  The pace of these acquisitions increased over the study period, with 59 practices acquired in 2013 and 136 practices acquired in 2016.⁴¹  While these acquisitions represent a small share of the 18,000 unique group medical practices in the United States, the trend is worth monitoring given the unique business model of these firms.⁴²  Private equity firms often sell their investments within three to seven years, so they may have a short time horizon for evaluating investments in improving medical providers.⁴³  Acquisition by a private equity firm can lead to more consolidation later, as these firms often then acquire additional nearby practices as part of their business model.⁴⁴  A study on the impact of private equity acquisitions of hospitals found that hospitals acquired by private equity firms reported larger increases in annual net income and hospital charges than similarly situated hospitals not acquired by private equity firms.⁴⁵ 

Anti-trust enforcement challenges and opportunities

In health care, along with other sectors of the economy, enforcement of federal and state anti-trust laws is supposed to ensure competitive markets that benefit consumers. At the federal level, the Federal Trade Commission (FTC) is charged with reviewing mergers. In the past, the FTC has blocked some hospital and physician mergers,⁴⁶  but the overall health care market has continued to become increasingly consolidated. FTC officials have cited several constraints on their ability to enforce anti-trust laws in the health care sector that may be contributing to the increases in consolidation in recent years.⁴⁷  Specifically, the FTC and Department of Justice’s anti-trust division have seen their budgets remain flat from 2010 to 2016, even as the pace of health care mergers has increased.⁴⁸  Vertical integration is particularly challenging for the FTC to monitor because it is often the result of hospitals acquiring many smaller practices and each of those transactions may fall under the threshold of having to notify FTC.^{49 , 50} 

Once a merger has taken place, states and the federal government can still enforce anti-trust laws. This can include pursing actions to stop anti-competitive practices such as a health care provider with significant market power preventing insurers from giving incentives to enrollees to go to less expensive providers.⁵¹  However, there is an important limitation on the FTC’s enforcement ability. The FTC Commissioner, Rebecca Kelly Slaughter, has raised concerns that the FTC is not able to enforce anti-trust rules on non-profit hospitals, although it can review mergers that involve a non-profit hospital.⁵²  Nationally, 57% of all hospitals are non-profit.⁵³  In 2019, 66% of the total hospital and health system mergers and acquisitions involved a non-profit entity purchasing another non-profit entity.⁵⁴ 

States can serve as another potential check on anti-competitive mergers and can sue under federal anti-trust law and enforce their own laws. A recent review of state anti-trust enforcement in health care identified several practices that support robust enforcement.⁵⁵  These include adequate notice requirements for potential mergers and waiting periods for state reviews; established criteria for merger review and the ability to conduct a full analysis of economic and health care implications; and the ability to implement post-merger monitoring.⁵⁶ 

There is no clear evidence that consolidation improves quality of care

While provider consolidation holds the promise of greater efficiencies and better care coordination, evidence of the benefits of improved quality after a merger are mixed at best, and some studies suggest that market consolidation—particularly for horizontal consolidation—can actually lead to lower quality care. It is difficult and takes time and resources to achieve true integration of care among newly merged health systems, while price increases often occur immediately after consolidation.⁵⁷ 

Regarding vertical integration, many studies showing that quality does not improve (or gets worse) after vertical integration, while some analysis has shown modest improvements.⁵⁸  One study of 15 integrated delivery networks finds no evidence that hospitals in these systems provide better clinical quality or safety scores than their competitors.⁵⁹  Another study found that larger hospital-based provider groups had higher per beneficiary Medicare spending and higher readmission rates than smaller groups.⁶⁰  However, one study looking at hospital quality measures from 2008 to 2015 found that vertical integration had a limited positive effect on a small subset of quality measures.⁶¹ 

Studies of markets where there has been horizontal consolidation largely have found that mergers do not improve quality, and that quality may actually be worse in highly concentrated markets than in markets with more competition. One study found that risk-adjusted one-year mortality for heart attacks in Medicare patients was 4.4% higher in more highly concentrated hospital markets compared to less concentrated markets.⁶²  Another study that analyzed Medicare claims for patients treated for hypertension, a cardiac condition or an acute myocardial infarction found that patients in areas higher cardiology market concentration had worse health outcomes and higher health care expenditures.⁶³ 

A study published in 2020 that followed hospitals for three years after a merger and compared those hospitals to a “control” group of hospitals that did not have a change in ownership found that the acquired hospitals’ outcome measures did not improve, when looking at scores for 30-day readmission and mortality rates among patients discharged from a hospital.⁶⁴  That analysis also found that patient experience worsened slightly after a merger, as measured by patients’ responses to questions about whether they would recommend the hospital and whether doctors and nurses always communicated well. The one improvement noted in that study was in process measures, which improved in the years before the acquisition and so could not be conclusively attributed to a change in ownership. The American Hospital Association funded a study using a similar design to look at the impact of hospital mergers on inpatient quality that found “small improvements in quality for some quality measures.”⁶⁵ 

Beyond health care quality, there have also been questions about the impact of hospital consolidation on the provision of charity care and other community benefits, and whether the impact differs depending on whether a hospital is for-profit or non-profit. A new study looking at this issue finds no evidence that non-profit hospitals provided more community benefits as their market share increased.⁶⁶ 

The Coronavirus and Resulting Economic Crisis May Make More Providers Likely to Consolidate

The COVID-19 pandemic and the resulting economic crisis are leading to unprecedented financial pressure on health care providers. It is still unclear if this pressure will lead to more mergers or cause providers to close, but it is possible that the pace of consolidation will increase due to the economic impact of the pandemic. Health care spending dropped dramatically after the start of the coronavirus pandemic (Figure 2) due to delayed or forgone medical care.⁶⁷  While spending started to pick up in May, it is not clear if that trend will continue given the record number of new COVID-19 cases in the subsequent months. This decrease in spending on health care services is leading to declines in provider revenue that could spur mergers, depending on the severity and duration of the revenue loss.

Figure 2

Compounding the impact of COVID-19, the current economic crisis could put additional financial pressure on some providers, particularly if the number of uninsured people rises. KFF has estimated that by early May 2020, nearly 27 million people were at risk of losing employer-sponsored coverage due to a job loss.⁶⁸  About half of those individuals were estimated to be eligible for Medicaid and about 30% were estimated to be eligible for subsidized marketplace coverage.⁶⁹  This shift from employer coverage to Medicaid alone will lead to lower revenues for providers, because employer-sponsored insurance tends to reimburse at much higher rates than Medicaid.⁷⁰ 

The federal government has made different types of funding available to providers to help them weather the coronavirus pandemic. It is not yet clear if this infusion of funds is sufficient to help the providers most vulnerable to declines in revenue due to coronavirus and the ensuing economic fallout. If providers are not able to pay their bills, they may be financially motivated to merge with a larger system. Below we outline the three main sources of stimulus funding sources for providers.

• $175B in provider relief grants: The Coronavirus Aid, Relief, and Economic Security (CARES) Act and the Paycheck Protection Program and Health Care Enhancement Act together allocated $175 billion for grants to providers to help cover expenses related to coronavirus and lost revenue due to the pandemic. About $118 billion in grants have already been allocated to providers, and a portion of the remaining money will be used to reimburse providers who treat uninsured COVID-19 patients.⁷¹  However, $50 billion of this fund was allocated to Medicare providers using a formula that gave more money to providers that tend to have higher margins.⁷²  While these grants provide some assistance for providers, it is a small share of money compared to total hospital spending, which was $1.2 trillion in 2018.⁷³ 
• $100B in advanced payments to Medicare providers: The Centers for Medicare & Medicaid Services distributed about a $100 billion in advanced Medicare payments to providers.⁷⁴  Most Medicare providers qualified for advanced payments representing three months of reimbursement from traditional Medicare in the period before coronavirus. About 80% of the advanced payments went to hospitals, with the remaining money going to physicians and other providers.⁷⁵  Repayment for those advanced payments was scheduled to begin in August 2020, 120 days after payment was issued.⁷⁶  However, providers are currently pushing for those loans to be forgiven, and Congress is also considering more favorable terms for repayment of the loans.⁷⁷ 
• Treasury department and Small Business Administration loans: Along with other businesses and employers, health care providers are also potentially eligible for some of the loans included in the CARES Act and the Paycheck Protection Program and Health Care Enhancement Act that the Treasury department, the Federal Reserve, and Small Business Administration are distributing. These loans include the Paycheck Protection Program (PPP) for small businesses, which forgives loans if employers do not lay off workers and meet other criteria. According to a Treasury Department analysis, health care providers received 13% of the $520 billion in PPP loans that have been distributed to small businesses.⁷⁸  While the CARES Act also appropriated $454 billion⁷⁹  for loans to qualifying larger businesses—including hospitals and other large health care entities—there have been delays in distributing those loans.⁸⁰  However, the Federal Reserve announced it is making changes to some of these loans so that it would be easier for non-profit institutions such as hospitals to qualify.⁸¹ 

Providers that accept any of these federal funds are not barred from future mergers, and most of this aid was not targeted to health care providers that may be most vulnerable to financial shocks from the coronavirus pandemic. Given that health care markets were consolidating even before the COVID-19 pandemic, aid to providers is unlikely to prevent health care markets from continuing to become more concentrated. However, it is possible that for some providers, this assistance may help stabilize their finances and increases the likelihood they can operate independently if that is their goal.

Discussion

There is now a large body of research showing that health care provider consolidation tends to raise prices without clear indications of quality improvements. Even before the pandemic, the U.S. health care system was becoming increasingly consolidated. The financial strains of the pandemic could increase the pace of consolidation among hospitals and physicians, which threatens to increase health care costs and premiums, without compelling evidence of commensurate quality improvements. Remedial action from policymakers could come in the form of increasing anti-trust enforcement—including taking steps to address any potential anti-competitive behavior in markets that are already consolidated—or targeted assistance to struggling providers that are trying to remain independent.

1. Cynthia Cox, Rabah Kamal, and Daniel McDermott, “How have healthcare utilization and spending changed so far during the coronavirus pandemic?” Peterson-KFF Health System Tracker. May 29, 2020. https://www.healthsystemtracker.org/chart-collection/how-have-healthcare-utilization-and-spending-changed-so-far-during-the-coronavirus-pandemic/.   ↩︎
2. Karyn Schwartz and Damico, Anthony. “Distribution of CARES Act Funding Among Hospitals.” KFF (blog), May 13, 2020. https://modern.kff.org/coronavirus-covid-19/issue-brief/distribution-of-cares-act-funding-among-hospitals/. ↩︎
3. “HHS Announces Over $4 Billion in Additional Relief Payments to Healthcare Providers Impacted by the Coronavirus Pandemic” Press Release, July 10, 2020. https://www.hhs.gov/about/news/2020/07/10/hhs-announces-over-4-billion-in-additional-relief-payments-to-providers-impacted-by-coronavirus-pandemic.html. ↩︎
4. Health plans merging with health care providers is another form of vertical integration, but that type of merger does not impact prices in the same way and so is not the focus of this brief. ↩︎
5. MedPAC, “March 2020 Report to the Congress: Medicare Payment Policy,” March 13, 2020. ↩︎
6. Reed Abelson. “When Hospitals Merge to Save Money, Patients Often Pay More” New York Times, November 18, 2018, Sec. B. https://www.nytimes.com/2018/11/14/health/hospital-mergers-health-care-spending.html ↩︎
7. Zack Cooper, Stuart V Craig, Martin Gaynor, and John Van Reenen. “The Price Ain’t Right? Hospital Prices and Health Spending on the Privately Insured.” Working Paper. National Bureau of Economic Research, December 2015. https://doi.org/10.3386/w21815 ↩︎
8. Martin Gaynor. “Examining the Impact of Health Care Consolidation,” Committee on Energy and Commerce Oversight and Investigations Subcommittee (2018). https://docs.house.gov/meetings/IF/IF02/20180214/106855/HHRG-115-IF02-Wstate-GaynorM-20180214.pdf ↩︎
9. Leemore Dafny, Kate Ho, and Robin Lee. “The Price Effects of Cross-Market Hospital Mergers.” Cambridge, MA: National Bureau of Economic Research, March 2016. https://doi.org/10.3386/w22106 ↩︎
10. John Simpson, and Richard Shin. “Do Nonprofit Hospitals Exercise Market Power?” Federal Trade Commission, November 1, 1996. https://www.ftc.gov/reports/do-nonprofit-hospitals-exercise-market-power; Michael G. Vita and Seth Sacher. “The Competitive Effects of Not-for-Profit Hospital Mergers: A Case Study.” The Journal of Industrial Economics 49, no. 1 (2001): 63–84. https://doi.org/10.1111/1467-6451.00138; Steven Tenn. “The Price Effects of Hospital Mergers: A Case Study of the Sutter–Summit Transaction.” International Journal of the Economics of Business 18, no. 1 (February 1, 2011): 65–82. https://doi.org/10.1080/13571516.2011.542956 ↩︎
11. Glenn Melnick, Yu-Chu Shen and Vivian Wu. “The Increased Concentration Of Health Plan Markets Can Benefit Consumers Through Lower Hospital Prices.” Health Affairs 30, no. 9 https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2010.0406 ↩︎
12. Lewis, Matthew S., and Kevin E. Pflum. “Hospital Systems and Bargaining Power: Evidence from out-of-Market Acquisitions.” The RAND Journal of Economics 48, no. 3 (2017): 579–610. https://doi.org/10.1111/1756-2171.12186 ↩︎
13. Lewis and Pflum, 2017. ↩︎
14. Gaynor, 2018 ↩︎
15. Gold, Jenny. “Surprise Settlement In Sutter Health Antitrust Case.” Kaiser Health News (blog), October 16, 2019. https://kffhealthnews.org/news/surprise-settlement-in-sutter-health-antitrust-case/ ↩︎
16. Gold, Jenny. “California AG Details ‘Historic’ Settlement Agreement In Sutter Health Antitrust Case.” California Healthline (blog), December 20, 2019. https://californiahealthline.org/news/california-ag-details-historic-settlement-agreement-in-sutter-health-antitrust-case/ ↩︎
17. Abe Dunn and Adam Shapiro. “Do Physicians Possess Market Power?” Journal of Law and Economics 57, no. 1 (January 1, 2014). https://chicagounbound.uchicago.edu/jle/vol57/iss1/6 ↩︎
18. Laurence C. Baker, M. Kate Bundorf, Anne B. Royalty, and Zachary Levin. “Physician Practice Competition and Prices Paid by Private Insurers for Office Visits.” JAMA 312, no. 16 (October 22, 2014): 1653–62. https://doi.org/10.1001/jama.2014.10921 ↩︎
19. Thomas Koch and Shawn W. Ulrick. “Price Effects of a Merger: Evidence from a Physicians’ Market.” SSRN Scholarly Paper. Rochester, NY: Social Science Research Network, August 1, 2017. https://doi.org/10.2139/ssrn.3026344 ↩︎
20. Richard Scheffler, Arnold, Daniel and Whaley, Christopher. “Consolidation Trends In California’s Health Care System: Impacts On ACA Premiums And Outpatient Visit Prices.” Health Affairs 37, no. 9 (September 1, 2018): 1409–16. https://doi.org/10.1377/hlthaff.2018.0472 ↩︎
21. Cory Capps, David Dranove, and Christopher Ody. “The Effect of Hospital Acquisitions of Physician Practices on Prices and Spending.” Journal of Health Economics 59 (May 1, 2018): 139–52. https://doi.org/10.1016/j.jhealeco.2018.04.001 ↩︎
22. Hannah T. Neprash et al. “Association of Financial Integration Between Physicians and Hospitals With Commercial Health Care Prices.” JAMA Internal Medicine 2015;175(12):1932-1939.   ↩︎
23. Laurence C. Baker, M. Kate Bundorf, and Daniel P. Kessler. “The Effect of Hospital/Physician Integration on Hospital Choice.” Journal of Health Economics 50 (December 1, 2016): 1–8. https://doi.org/10.1016/j.jhealeco.2016.08.006 ↩︎
24. Melnick et al. 2011.   ↩︎
25. Richard M. Scheffler and Daniel R. Arnold. “Insurer Market Power Lowers Prices in Numerous Concentrated Provider Markets.” Health Affairs. 2017 36:9, 1539-1546.   ↩︎
26. Erin E. Trish and Bradley J. Herring. “How Do Health Insurer Market Concentration and Bargaining Power with Hospitals Affect Health Insurance Premiums?.” Journal of Health Economics. Vol. 42 (July 2015). ↩︎
27. Unlike hospitals paid on the prospective payment system, Critical Access Hospitals are paid based on a percent of their costs.   ↩︎
28. MedPAC, “March 2020 Report to the Congress: Medicare Payment Policy,” March 13, 2020.   ↩︎
29. Gaynor, 2018 ↩︎
30. “TrendWatch Chartbook 2018 | AHA.” Chartbook. American Hospital Association, 2018. https://www.aha.org/guidesreports/2018-05-22-trendwatch-chartbook-2018 ↩︎
31. David M. Cutler and Fiona Scott Morton. “Hospitals, Market Share, and Consolidation.” JAMA vol. 310 no. 18 (November 13, 2013). ↩︎
32. Brent D. Fulton. “Health Care Market Concentration Trends In The United States: Evidence And Policy Responses.” Health Affairs 36, no. 9 (September 1, 2017): 1530–38. https://doi.org/10.1377/hlthaff.2017.0556 ↩︎
33. Fulton, 2017. ↩︎
34. Fulton, 2017.   ↩︎
35. Fulton, 2017. ↩︎
36. Carol K. Kane. “Updated Data on Physician Practice Arrangements: For the First Time, Fewer Physicians are Owners Than Employees.” American Medical Association. April 2019. Available at: https://www.ama-assn.org/system/files/2019-07/prp-fewer-owners-benchmark-survey-2018.pdf ↩︎
37. Fulton, 2017.   ↩︎
38. Dafny, 2015.   ↩︎
39. Meredith Freed, Anthony Damico and Tricia Neuman. “A Dozen Facts About Medicare Advantage in 2020.” Kaiser Family Foundation. Apr 22, 2020.   ↩︎
40. Jane M. Zhu, Lynn M. Hua, and Daniel Polsky. “Private Equity Acquisitions of Physician Medical Groups Across Specialties, 2013-2016.” JAMA 323, no. 7 (February 18, 2020): 663–65. https://doi.org/10.1001/jama.2019.21844 ↩︎
41. Zhu et all , 2020. ↩︎
42. Zhu et al., 2020. ↩︎
43. Jack S. Resneck. “Dermatology Practice Consolidation Fueled by Private Equity Investment: Potential Consequences for the Specialty and Patients.” JAMA Dermatology 154, no. 1 (01 2018): 13–14. https://doi.org/10.1001/jamadermatol.2017.5558 ↩︎
44. Suhas Gondi and Zirui Song. “Potential Implications of Private Equity Investments in Health Care Delivery.” JAMA 321, no. 11 (March 19, 2019): 1047–48. https://doi.org/10.1001/jama.2019.1077 ↩︎
45. Joseph D. Bruch, Suhas Gondi and Zirui Song. “Changes in Hospital Income, Use, and Quality Associated with Private Equity Acquisition.” JAMA Internal Medicine (August 24, 2020).   ↩︎
46. Christine S. Wilson. “The FTC’s Ongoing Efforts to Promote Competition and Choice in Our Health Care System.” Keynote Remarks at the Council for Affordable Health Coverage: The Price of Good Health – 2020 and Beyond, January 16, 2020. ↩︎
47. Rebecca Slaughter. “Antitrust and Health Care Providers Policies to Promote Competition and Protect Patients Center for American Progress.” Washington, DC: Federal Trade Commission, May 14, 2019. ↩︎
48. Slaughter, 2019.   ↩︎
49. MedPAC, “March 2020 Report to the Congress: Medicare Payment Policy,” March 13, 2020.   ↩︎
50. Slaughter, 2019.   ↩︎
51. Gaynor, 2018.   ↩︎
52. Slaughter, 2019.   ↩︎
53. Kaiser Family Foundation. “Hospitals by Ownership Type.” 2018. ↩︎
54. Kaufman, Hall & Associates, LLC. “2019 M&A in Review: In Pursuit of the New Bases of Competition,” 2020. https://www.kaufmanhall.com/sites/default/files/documents/2020-01/2019_mergers_and_acquisitions_report_kaufmanhall.pdf ↩︎
55. Jaime S. King “Preventing Anticompetitive Healthcare Consolidation: Lessons from Five States.” The Source on Healthcare Price and Competition. Petris Center. June 2020. https://2zele1bn0sl2i91io41niae1-wpengine.netdna-ssl.com/wp-content/uploads/2020/06/PreventingAnticompetitiveHealthcareConsolidation.pdf   ↩︎
56. Ibid.   ↩︎
57. Gaynor, 2018 ↩︎
58. MedPAC, “March 2020 Report to the Congress: Medicare Payment Policy,” March 13, 2020. ↩︎
59. Jeff Goldsmith, Lawton Burn, Aditi Sen, and Trevor Goldsmith. “Integrated Delivery Networks: In Search of Benefits and Market Effects.” National Academy of Social Insurance, February 2015. https://www.nasi.org/research/2015/integrated-delivery-networks-search-benefits-market-effects. ↩︎
60. J. Michael McWilliams, Michael E. Chernew, Alan M. Zaslavsky, Pasha Hamed, and Bruce E. Landon. “Delivery System Integration and Health Care Spending and Quality for Medicare Beneficiaries.” JAMA Internal Medicine 173, no. 15 (August 12, 2013): 1447–56. https://doi.org/10.1001/jamainternmed.2013.6886. ↩︎
61. Marah Noel Short and Vivian Ho. “Weighing the Effects of Vertical Integration Versus Market Concentration on Hospital Quality.” Medical Care Research and Review: MCRR, February 9, 2019, 1077558719828938. https://doi.org/10.1177/1077558719828938. ↩︎
62. Daniel P Kessler and Mark B McClellan. “Is Hospital Competition Socially Wasteful?” NBER Working Paper, July 1999. ↩︎
63. Thomas Koch, Brett Wendling, and Nathan E. Wilson. “Physician Market Structure, Patient Outcomes, and Spending: An Examination of Medicare Beneficiaries.” Health Services Research 53, no. 5 (2018): 3549–68. https://doi.org/10.1111/1475-6773.12825. ↩︎
64. Nancy D. Beaulieu, Leemore S. Dafny, Bruce E. Landon, Jesse B. Dalton, Ifedayo Kuye, and J. Michael McWilliams. “Changes in Quality of Care after Hospital Mergers and Acquisitions.” The New England Journal of Medicine 382, no. 1 (02 2020): 51–59. https://doi.org/10.1056/NEJMsa1901383. ↩︎
65. M. Noether and S. May, “Hospital Merger Benefits, a Review and Extension” American Hospital Association, December 2018. https://www.aha.org/system/files/2019-01/hospital-merger-benefits-a-review-and-extension-jan-2-19.pdf ↩︎
66. Cory Capps, Dennis W. Carlton, and Guy David, “Antitrust Treatment of Nonprofits: Should Hospitals Receive Special Care?.” Econ Inquiry, 58: 1183-1199. (2020) doi:10.1111/ecin.12881 ↩︎
67. Liz Hamel, Audrey Kearney, Ashley Kirzinger, Lunna Lopes, Cailey Munana, and Mollyann Brodie. “KFF Health Tracking Poll – May 2020 – Health and Economic Impacts.” KFF (blog), May 27, 2020. https://modern.kff.org/report-section/kff-health-tracking-poll-may-2020-health-and-economic-impacts/. ↩︎
68. Rachel Garfield, Gary Claxton, Anthony Damico, and Larry Levitt, “Eligibility for ACA Health Coverage Following Job Loss” KFF, May 13, 2020. https://modern.kff.org/coronavirus-covid-19/issue-brief/eligibility-for-aca-health-coverage-following-job-loss/ ↩︎
69. Garfield et al, 2020. ↩︎
70. Gary Claxton, Matthew Rae, Larry Levitt, and Cynthia Cox. “How have healthcare prices grown in the U.S. over time?” Peterson-KFF Health System Tracker. May 8, 2018.   ↩︎
71. Karyn Schwartz and Jennifer Tolbert. “Limitations of the Program for Uninsured COVID-19 Patients Raise Concerns.” Kaiser Family Foundation. August 24, 2020. ↩︎
72. Schwartz and Damico, 2020 ↩︎
73. Peterson-KFF Health System Tracker, “National Health Spending Explorer.” Peterson-KFF Health System Tracker. Available at: https://www.healthsystemtracker.org/health-spending-explore.   ↩︎
74. Juliette Cubanski, Karyn Schwartz, Jeannie Fuglesten Biniek, and Tricia Neuman. “Medicare Accelerated and Advance Payments for COVID-19 Revenue Loss: Time to Repay?” Kaiser Family Foundation. August 07, 2020.   ↩︎
75. CMS. “Medicare Accelerated and Advance Payments State-by-State and by Provider Type.” Available at: https://www.cms.gov/files/document/covid-accelerated-and-advance-payments-state.pdf.   ↩︎
76. HHS, “Fact Sheet: Expansion of the Accelerated and Advance Payment Program for Providers and Suppliers During COVID-19 Emergency.” https://www.cms.gov/files/document/accelerated-and-advanced-payments-fact-sheet.pdf. ↩︎
77. Bertha Coombs. “Hospitals push Congress to forgive $100 billion in coronavirus aid Medicare loans.” CBNC. July 28, 2020. https://www.cnbc.com/2020/07/28/hospitals-push-congress-to-forgive-100-billion-in-medicare-loans.html ↩︎
78. Small Business Administration. “Paycheck Protection Program (PPP) Report: Approvals through 07/10/2020.” Available at: https://home.treasury.gov/system/files/136/SBA-Paycheck-Protection-Program-Loan-Report-Round2.pdf.   ↩︎
79. This total excludes the $46 billion appropriated for loans for the aviation industry and businesses “critical to maintaining national security.”   ↩︎
80. Glenn Hubbard and Hal Scott. “‘Main Street’ Program Is Too Stingy to Banks and Borrowers.” Wall Street Journal. July 20, 2020.   ↩︎
81. Federal Reserve Board, “Federal Reserve Board modifies Main Street Lending Program to provide greater access to credit for nonprofit organizations such as educational institutions, hospitals, and social service organizations,” July 17, 2020. Available at: https://www.federalreserve.gov/newsevents/pressreleases/monetary20200717a.htm.   ↩︎

The Affordable Care Act’s (ACA) future continues to be uncertain as the law’s constitutionality will once again be considered by the U.S. Supreme Court in California v. Texas¹  (known as Texas v. U.S. in the lower courts). Oral argument is scheduled for Tuesday, November 10, 2020. This ongoing litigation challenges the ACA’s minimum essential coverage provision (known as the individual mandate) and raises questions about the entire law’s survival. The individual mandate provides that most people must maintain a minimum level of health insurance coverage; those who do not do so must pay a financial penalty (known as the shared responsibility payment) to the IRS. The individual mandate was upheld as a constitutional exercise of Congress’ taxing power by a five member majority of the Supreme Court in NFIB v. Sebelius in 2012.

In the 2017 Tax Cuts and Jobs Act (TCJA), Congress set the shared responsibility payment at zero dollars as of January 1, 2019, leading to the current litigation. In December 2019, the U.S. Court of Appeals for the 5^th Circuit affirmed the trial court’s decision that the individual mandate is no longer constitutional because the associated financial penalty no longer “produces at least some revenue” for the federal government.²  But, instead of deciding whether the rest of the ACA must be struck down, the 5^th Circuit sent the case back to the trial court for additional analysis. However, the Supreme Court has now agreed to review the case.

The ACA remains in effect while the litigation is pending. However, if all or most of the law ultimately is struck down, it will have complex and far-reaching consequences for the nation’s health care system, affecting nearly everyone in some way. A host of ACA provisions could be eliminated, including protections for people with pre-existing conditions, subsidies to make individual health insurance more affordable, expanded eligibility for Medicaid, coverage of young adults up to age 26 under their parents’ insurance policies, coverage of preventive care with no patient cost-sharing, closing of the doughnut hole under Medicare’s drug benefit, and a series of tax increases to fund these initiatives.

This issue brief answers key questions about the litigation as we await a decision from the Supreme Court about the ACA’s survival.

1.  Who Is Challenging the ACA?

A group of 20 states, led by Texas, sued the federal government in February 2018, seeking to have the entire ACA struck down (the “state plaintiffs”).³  These states are represented by 18 Republican attorneys general and 2 Republican governors. After Democratic victories in the 2018 mid-term elections, two of these states, Wisconsin and Maine, withdrew from the case in early 2019, leaving 18 states challenging the ACA on appeal (Figure 1).⁴ 

Two individuals joined the lawsuit in the trial court in April 2018, as plaintiffs challenging the ACA.⁵  These plaintiffs are self-employed residents of Texas who claim that the individual mandate requires them to purchase health insurance that they otherwise would not buy, although there is no penalty if they fail to buy coverage.

2.  What Is the Federal Government’s Position in the Case, and How Has It Changed Over Time?

Throughout the litigation, the federal government has not defended the constitutionality of the ACA’s individual mandate. Instead, the federal government agrees with the state and individual plaintiffs that the individual mandate is no longer constitutional under Congress’s taxing power as a result of the TCJA provision that set the financial penalty at zero.⁶  It is unusual for the federal government to take a position that does not seek to uphold a federal law.

Unlike the plaintiffs, the federal government argued at the trial court that only the ACA’s protections for people with pre-existing conditions, including guaranteed issue and community rating, should be struck down along with the individual mandate. The federal government took the position that these provisions cannot function effectively without the individual mandate but the rest of the ACA should be allowed to survive.

Notably, the federal government changed its position while the case was on appeal at the 5^th Circuit (Figure 2). First, the federal government took what the 5^th Circuit called a “significant change in litigation position”⁷  by deciding to support the trial court’s decision that the individual mandate is inseverable from the entire ACA.⁸  This change came after the federal government had appealed, asking the 5^th Circuit to review the trial court’s decision. Next, the federal government raised new arguments about the scope of relief that the court should grant, asserting that the federal government should be enjoined from enforcing only the ACA provisions that injure the plaintiffs. For example, the federal government identified “several criminal statutes used to prosecute individuals who defraud our healthcare system” that are part of the ACA that it believes should survive.⁹  The federal government also argued for the first time in the 5^th Circuit that any injunction prohibiting enforcement of the ACA should apply only in the plaintiff states.¹⁰ 

The federal government is asking the Supreme Court to prohibit it from enforcing only the ACA provisions that are found to harm the individual plaintiffs. Even though the federal government is arguing that the entire ACA should be found invalid (because the individual mandate is no longer constitutional and cannot be severed from the rest of the law), the federal government does not want the Court to necessarily prevent it from still enforcing parts of the law. Instead, the federal government is seeking a more limited remedy:  it contends that “relief should reach only the enforcement of the ACA provisions that injure the individual plaintiffs.”¹¹  The federal government has not clearly identified which specific ACA provisions fall into this category and is asking the Supreme Court to send the case back to the lower courts to determine this issue.¹² 

3.  Who is Defending the ACA?

Another 17 states, led by California, were permitted by the trial court to intervene in the case and defend the ACA (the “state intervener-defendants”). Subsequently, the 5^th Circuit allowed four more states to intervene in the case on appeal, bringing the total number of states defending the ACA in the case to 21 .¹³  In addition, six states filed an amicus brief in the Supreme Court in support of the ACA (Figure 1).

The 5^th Circuit also allowed the U.S. House of Representatives to intervene in the case to defend the ACA on appeal.¹⁴  However, the 5^th Circuit did not decide whether the House has standing to pursue the appeal.¹⁵  The standing of the state intervener-defendants and/or the House is particularly important in this case, since the federal government is not defending the ACA (Figure 3). At the Supreme Court, the parties are not contesting, and the Court has not asked for briefing on, California’s ability to pursue an appeal (California and the House both filed cert petitions raising the same issues, and the Court accepted California’s petition).

4.  What Did the 5^th Circuit Decide?

The 5^th Circuit issued a 2:1 decision finding the individual mandate unconstitutional and sending the case back to the trial court for additional analysis about whether the rest of the ACA can survive. There are three main issues in the case: (A) whether the parties have standing to invoke the court’s jurisdiction; (B) whether the ACA’s individual mandate, as amended by the TCJA, is constitutional; and (C) if the mandate is unconstitutional, whether it can be severed from the rest of the ACA, or on the other hand, whether other provisions of the ACA also must be invalidated. Figure 4 illustrates the legal questions and potential outcomes in the case.

(A) The parties have standing to litigate the case.

The 5^th Circuit decided that the case presented a live controversy for it to resolve, despite the unusual alignment of the parties’ positions. Although the federal government is “in almost complete agreement on the merits of the case” with the plaintiffs, it also has indicated that it will continue to enforce the ACA unless or until a court issues a final order striking the law down.¹⁶  The state intervener-defendants have standing to pursue an appeal because they would be injured by the loss of federal ACA funding, such as funding for the Medicaid expansion and the Medicaid Community First Choice attendant care program, if the trial court’s decision is upheld.¹⁷ 

The 5^th Circuit decided that the both the individual and state plaintiffs have standing to challenge the ACA in court. Standing ensures that federal courts are deciding actual cases or controversies as required by the U.S. Constitution. Standing is essential for the court to have jurisdiction to decide a case and therefore cannot be waived. To establish standing, a party must suffer an injury that is concrete and actual or imminent; fairly traceable to the challenged conduct; and likely to be redressed by a favorable court ruling. The 5^th Circuit agreed with the trial court that the individual plaintiffs have standing because they have spent money that they otherwise would not have spent, absent the individual mandate, to purchase health insurance.¹⁸  The 5^th Circuit also decided that the state plaintiffs have standing because they are incurring costs from the individual mandate from having to verify which state employees have minimum essential coverage.¹⁹ 

The dissent reached the opposite conclusion, finding that neither the individual nor the state plaintiffs has standing to bring the case. According to the dissent, any injury experienced by the individual plaintiffs “is entirely self-inflicted” because “absolutely nothing” will happen to them if they do not purchase insurance to meet the individual mandate now that the penalty is set at zero.²⁰  The dissent also concluded that the state plaintiffs lack standing because they failed to provide evidence showing that “at least some state employees have enrolled in employer-sponsored health insurance” or that “anyone has enrolled in their Medicaid programs solely because of the unenforceable coverage requirement.”²¹ 

(B) The individual mandate is unconstitutional after the TCJA set the financial penalty at zero.

The 5^th Circuit decided that the individual mandate as amended by the TCJA is unconstitutional. The court agreed with the state and individual plaintiffs and the federal government’s assertion that the requirement to produce some revenue is “essential” to the Supreme Court’s earlier finding in NFIB that the individual mandate could be saved as a valid exercise of Congress’s power to tax.²²  Without that feature, the mandate is a command to purchase health insurance, which as the Supreme Court held in NFIB, is an unconstitutional exercise of Congress’ power to regulate interstate commerce.

The dissent concluded that the individual mandate remains constitutional because the TCJA amendment is “a law that does nothing.”²³  The dissent reasoned that the TCJA did not change the text of the coverage requirement and therefore did not change the individual mandate into a mandatory command to purchase insurance. Rather, Congress “changed the parameters” of the choice about whether to purchase insurance from paying a tax penalty to “no consequences at all.”²⁴ 

(C) The trial court’s analysis about whether the individual mandate is severable from the rest of the ACA was incomplete.

The 5^th Circuit sent the case back to the trial court for additional analysis about which ACA provisions should survive without the individual mandate. The trial court incorrectly focused on the intent of Congress in 2010 when passing the ACA and instead should have considered Congress’ intent when enacting the TCJA and setting the shared responsibility payment at zero in 2017.²⁵  In so doing, the trial court should “employ a finer-toothed comb. . . and conduct a more searching inquiry into which provisions of the ACA Congress intended to be inseverable from the individual mandate. . . us[ing] its best judgment to determine how best to break the ACA down into constituent groups, segments, or provisions to be analyzed.”²⁶ 

The 5^th Circuit also directed the trial court to consider the federal government’s new argument that any order prohibiting enforcement of the ACA should extend only to provisions that injure the plaintiffs and apply only in the plaintiff states. The trial court may consider whether the federal government timely raised this argument and whether Supreme Court precedent supports limiting the remedy in this way.²⁷ 

The dissent criticized the majority’s failure to send the case back to the trial court instead of resolving the severability issue. Severability is a question of law, which the 5^th Circuit could have resolved without sending the case back to the trial court. The dissent agreed with the majority that the severability analysis should look to the intent of Congress when passing the TCJA in 2017. However, the dissent concluded that the fact that Congress changed the tax penalty amount to zero while leaving the rest of the ACA in place indicates that Congress intended for all of the other provisions to remain in effect.²⁸ 

5.  What is Happening at the Supreme Court?

The Supreme Court has agreed to review four legal questions in the case. First, the Court will consider whether Texas and the individual plaintiffs have standing to bring the lawsuit to challenge the individual mandate. If so, the Court will determine whether the TCJA rendered the individual mandate unconstitutional. If the mandate is unconstitutional, the Court will decide whether the rest of the ACA can survive. Finally, if the entire ACA is held invalid, the Court will resolve whether the entire law should be unenforceable nationwide or whether it should be unenforceable only to the extent that provisions injure the individual plaintiffs.

The case will be argued at the Supreme Court on November 10, 2020. The Court has allotted one hour and twenty minutes for oral argument, with 40 minutes for each side. California will argue for 30 minutes of the time allotted to the parties defending the ACA, with the remaining 10 minutes argued by the House. The time allotted to the parties challenging the ACA will be evenly divided between the federal government and Texas, with 20 minutes for each. The Court denied Ohio and Montana’s motion to participate in oral argument as amici curiae in support of neither side. The decision could come as late as the end of term in June 2021.

Looking Ahead

If the Supreme Court finds that the individual mandate is unconstitutional and invalidates only that provision, the practical result will be essentially the same as the ACA exists today, without an enforceable mandate. If the Supreme Court adopts the position that the federal government took during the trial court proceedings and invalidates the individual mandate as well as the protections for people with pre-existing conditions, then federal funding for premium subsidies and the Medicaid expansion would stand, and it would be up to states whether to reinstate the insurance protections. The Supreme Court also could decide that Texas and the individual plaintiffs do not have standing to bring the lawsuit, which would allow the ACA as it exists today to remain in effect.

The most far-reaching consequences, affecting nearly every American in some way, will occur if the Supreme Court ultimately decides that all or most of the ACA must be overturned, as the federal government now argues. The number of non-elderly individuals who are uninsured decreased by 18.6 million from 2010 to 2018, as the ACA went into effect. The ACA made significant changes to the individual insurance market, including requiring protections for people with pre-existing conditions, creating insurance marketplaces, and authorizing premium subsidies for people with low and modest incomes. The ACA also made other sweeping changes throughout the health care system including expanding Medicaid eligibility for low-income adults; requiring private insurance, Medicare, and Medicaid expansion coverage of preventive services with no patient cost sharing; phasing out the Medicare prescription drug doughnut hole coverage gap; reducing the growth of Medicare payments to health care providers and insurers; establishing new national initiatives to promote public health, care quality, and delivery system reforms; and authorizing a variety of tax increases to finance these changes. All of these provisions could be overturned if all or most of the ACA is struck down by the courts, and it would be enormously complex to disentangle these provisions from the overall health care system.

For now, the ACA remains in effect. The trial court’s original decision that the entire ACA should be invalidated was never implemented and was set aside by the 5^th Circuit. Additionally, the Trump Administration has indicated that it intends to continue enforcing the ACA while the appeal is pending. Although the Supreme Court’s decision in the case could come as late as June 2021, the Court’s decision to review the case now, without waiting for the lower courts to complete their review, will minimize the amount of time that the ACA’s future remains uncertain.²⁹  If the Supreme Court had not agreed to review the case now, the litigation likely would have continued for several more years, while the trial court issued a new decision on severability and that decision was then reviewed by the 5^th Circuit, before returning to the Supreme Court. Still, 10 years after its enactment, the only certainty for the ACA in the foreseeable future is that there is continuing uncertainty about its ultimate survival.

1. No. 19-840, https://www.supremecourt.gov/search.aspx?filename=/docket/docketfiles/html/public/19-840.html. The case has been consolidated with Texas v. California, No. 19-1019, https://www.supremecourt.gov/search.aspx?filename=/docket/docketfiles/html/public/19-1019.html. ↩︎
2. Texas v. U.S., No. 19-10011, slip opin. (5th Cir. Dec. 20, 2019) (revised with technical corrections), https://affordablecareactlitigation.files.wordpress.com/2019/12/fifth-circuit-opinion-technical-revisions-12-20.pdf. ↩︎
3. Texas v. U.S., No. 4:18-cv-00167-O, Compl. (N.D. Tex. Feb. 26, 2018), https://affordablecareactlitigation.files.wordpress.com/2018/09/177111358274.pdf. ↩︎
4. In addition, Montana and Ohio filed an amicus brief in the 5th Circuit and the Supreme Court, arguing that the individual mandate is now unconstitutional but that it should be severed, allowing the rest of the law to survive. ↩︎
5. Texas v. U.S., No. 4:18-cv-00167-O, Amended Compl. (N.D. Tex. April 23,, 2018), https://affordablecareactlitigation.files.wordpress.com/2018/09/texas-v-us-aca-amended-complaint.pdf. ↩︎
6. The federal government and the state and individual plaintiffs also endorse the Supreme Court’s determination in NFIB that the individual mandate is not a constitutional exercise of Congress’ power to regulate interstate commerce. The trial court adopted and the 5th Circuit affirmed both of these conclusions in their decisions. ↩︎
7. Texas v. U.S., supra. n. 2, slip opin. at 54. ↩︎
8. Letter from U.S. Dep’t of Justice to U.S. Court of Appeals for the Fifth Circuit Clerk (March 25, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/03/doj-anti-aca-letter-3-25.pdf. ↩︎
9. It stated that the “relief awarded to the plaintiffs should extend only to the ACA’s provisions that actually injure them. Texas v. U.S., no 19-1001, Brief for the Federal Defendants at 19 (5th Cir. May 1, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/05/5c-us-brief.pdf. ↩︎
10. Texas v. U.S., supra. n. 2, slip opin. at 10-11, 61. ↩︎
11. Id. at 12. ↩︎
12. Texas v. U.S., Nos. 19-840 and 19-1019, Br. for Fed. Respond. at 49 (June 25, 2020), https://www.supremecourt.gov/DocketPDF/19/19-840/146406/20200625205555069_19-840bsUnitedStates.pdf. ↩︎
13. Texas v. U.S., No. 19-1001, Order (5th Cir. Feb. 14, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/02/5c-order-denying-motion-to-expedite-granting-oregon-motion-to-intervene.pdf. ↩︎
14. Texas v. U.S., No. 19-1001, Order (5th Cir. Feb. 14, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/02/5c-order-granting-us-house-motion-to-intervene.pdf. The House also sought to intervene in the trial court proceedings in January 2019, but the trial court stayed briefing on that motion, citing the pending 5th Circuit appeal. ↩︎
15. Texas v. U.S., supra. n. 2, slip opin. at 16. ↩︎
16. Id. at 13. ↩︎
17. Id. at 15. ↩︎
18. Id. at 21-23. ↩︎
19. Id. at 28-33. ↩︎
20. Id. at 69. ↩︎
21. Id. at 75, 77. ↩︎
22. Texas v. U.S., No. 19-1001, Br. for State Appellees at 34 (5th Cir. May 1, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/05/5c-appellees-brief.pdf. The federal government pointed out that prior to the TCJA, the shared responsibility payment for tax year 2019 and beyond would have been the greater of 2.5% of household income or $695. Texas v. U.S., no. 19-1001, Br. for the Fed. Defendants at 13 (5th Cir. May 1, 2019), https://affordablecareactlitigation.files.wordpress.com/2019/05/5c-us-brief.pdf. ↩︎
23. Texas v. U.S., supra. n. 2, slip opin. at 79. ↩︎
24. Id. at 84, 83 (emphasis in original). ↩︎
25. Id. at 56. ↩︎
26. Id. at 59. ↩︎
27. Id. at 62. ↩︎
28. Id. at 84-85. ↩︎
29. The Court denied California’s request to review the cert petition on an expedited basis. California and the House had asked the Court to proceed on an expedited basis so that the case could be heard and decided in the current term, by June 2020. ↩︎

Since the COVID-19 pandemic first surfaced in the United States, the number of cases and deaths in long-term care (LTC) facilities has been rising. As of August 20, 2020, over 70,000 COVID-19 related resident and staff deaths have been reported in nursing homes and other long-term care facilities, which is a conservative estimate because not all states publish these data. The increase in deaths among long-term care facility residents and staff has become an urgent concern for federal and state policymakers, the long-term care industry, family members of residents, residents themselves, and the general public.

While COVID-19 outbreaks and deaths in nursing homes have received a fair amount of attention, assisted living facilities (ALFs), which are home to over 800,000 mostly frail, elderly residents, have been largely overlooked. Unlike nursing homes, assisted living facilities are not federally regulated, leaving states to decide whether or not to publicly report data or to impose restrictions to protect residents. This analysis examines the impact of COVID-19 on assisted living facilities as well as changes over time, using state-level data on COVID-19 cases and deaths reported in early June 2020, and again in early August. These counts are a subset of the state-level COVID-19 cases and deaths in all long-term care facilities, including nursing homes, as reported in other KFF analyses. (See Methods for details).

Less Than Half of All States Report COVID-19 Cases in Assisted Living Facilities and Even Fewer Report Deaths

As of August 2020, 19 states identify COVID-19 cases or deaths specific to assisted living facilities, an increase of four states since June 2020. Of these 19 states, 13 [CO, CT, FL, KY, MA, NV, ND, OH, PA, RI, TN, TX, UT] report COVID-19 data for assisted living facilities in its own, distinct category, and 6 [CA, GA, LA, NC, NY, SC] report COVID-19 data for assisted living facilities along with congregate settings other than nursing homes (Tables 1 and 2). This leaves 31 states and DC that do not identify COVID-19 cases and deaths occurring in assisted living facilities specifically, as of August 2020.

• 18 of the 19 states report COVID-19 cases in assisted living facilities: As of August 2020, 18 states report COVID-19 cases in assisted living facilities, either in its own distinct category (13 states) or combined with other congregate, non-nursing facilities (5 states), and report either cumulatively (13 states) or active cases only (5 states). Of these 18 states, 14 report cases among residents and staff separately, 2 report cases among residents and staff combined, and 2 report cases among residents only.
• 14 of the 19 states report COVID-19 deaths in assisted living facilities: As of August 2020, 14 states report COVID-19 deaths in assisted living facilities, either in its own distinct category (8 states) or combined with other congregate settings other than nursing homes (6 states), and report either cumulatively (12 states) or only among facilities with ongoing outbreaks (2 states). Of these 14 states, 7 report deaths among residents and staff separately, and 7 states report deaths among residents only

COVID-19 Cases and Deaths in Assisted Living Facilities Have Increased, With a Notable Increase in Cases Among Staff

CASES: As of early August 2020, a total of 22,080 COVID-19 cases have been reported among residents and staff in assisted living facilities, based on the 18 states reporting COVID-19 cases data. This total reflects both the number of cases among 14 states that were reporting this information in June and the addition of 4 states that started reporting since then. Among the 14 states that reported COVID-19 cases in both June and August, the number of cases among residents and staff has increased by 66% and the number of cases among residents only has increased by 63%. This is an undercount because it is based on data reported by a minority of states.

As of early August 2020, a total of 7,626 cases were reported among assisted living staff in the 14 states reporting staff cases, including 6 states that started reporting since June. Among the 8 states that reported in both June and August, the number of staff cases has increased by 156% from 2,085 to 5,333 cases in early August.

DEATHS: As of early August 2020, a total of 2,651 deaths among residents and staff have been reported in the 14 states that identify COVID-19 deaths specific to assisted living facilities, including the 10 states that reported deaths in both June and August, and 4 states that started reporting since June. Among the 10 states that reported deaths in both June and August, the total number of deaths increased by 59% from 1,483 to 2,356 deaths in early August. The majority of reported COVID-19 deaths are among assisted living facility residents (2,257); a relatively small number represent deaths among staff (99).

As of early August 2020, a total of 99 deaths were reported among assisted living staff in the 7 states reporting staff deaths, including 2 states that started reporting since June. Among the 5 states that reported in both June and August, the number of deaths has increased by 219% from 31 to 99 deaths in early August.

In the 10 states reporting cumulative COVID-19 CASE numbers for assisted living facilities in both June and August, the percentage increase in the aggregate number of COVID-19 cases in the population overall was significantly greater than the percentage increase in aggregate resident and staff cases in ALFs (223% versus 61%). But the opposite is true in the states reporting cumulative deaths in assisted living facilities. In the nine states reporting cumulative DEATH data for assisted living facilities separately from nursing homes, the aggregate percentage increase in COVID-19 deaths occurring in the overall population in these states was roughly half of the increase in resident and staff deaths occurring in ALFs between June and August (36% vs. 60%).

Discussion

Despite intense scrutiny of the number of COVID-19 cases and deaths in nursing facility settings, less than half of all states are reporting data for COVID-19 in assisted living facilities specifically. As a result, it is difficult to know the extent to which residents and staff in assisted living facilities have been affected by COVID-19 or the extent to which interventions are urgently needed. Our analysis finds a significant increase in COVID-19 cases and deaths among residents and staff in assisted living facilities in the two-month period between June and August. The rise in cases among staff is especially noteworthy. Notably, four out of five states [CA, FL, NV, SC] with the largest increase in cases among staff are also considered “hotspot” states with widespread community transmission. The rise in COVID-19 cases among staff is most likely to disproportionately affect female, Black, and low-wage workers, based on a recent analysis.

Since COVID-19 data for assisted living facilities are reported separately from nursing facilities by a minority of states, the counts of cases and deaths presented in this analysis are undoubtedly conservative. Compounding this data limitation, states that do report for assisted living facilities separately from nursing homes vary significantly in what they report: some states report cases, but not deaths, and some do not report cases or deaths among staff. In fact, only seven states separately report deaths among staff working in assisted living facilities. Additionally, while some states (e.g., NY, LA) have been reporting cumulative case and death data dating back to early March, others do not specify the start date of their retrospective data reporting, leading to potential undercounts of cases and deaths that have occurred since the beginning of the pandemic. The reporting of active cases only by some states (such as Florida) is likely to result in an undercount of the true magnitude of cases and deaths since the numbers do not take into account cases and deaths that may have occurred but are no longer active.

Overall, the incomplete system of state-level reporting of COVID-19 data in assisted living facilities results in an incomplete picture of disease incidence and mortality among staff and residents in these facilities. Based on data from the states that do report, outbreaks in assisted living facilities, and protections for residents and staff, warrant more careful attention.

Table 1:

–

To collect data on COVID-19 cases and deaths in ALFs, we reviewed public reporting of COVID-19 surveillance data displayed on applicable state-run websites in all states, and collected numbers of cases and deaths for ALFs in states where this data is reported separately from cases and deaths in LTC facilities generally, in order to exclude nursing facilities from our analysis. Not all states report COVID-19 data for LTC facilities, and some only report this data for nursing homes. For this analysis, we collected state data on COVID-19 cases and deaths between June 5 and June 8, 2020, and between August 3 and August 6, 2020.

We included COVID-19 case and death data for each state (n=13) that reported for ALFs specifically, as well as other states (n=6) that report ALF data separately from nursing facilities, but do so within a larger category that includes ALFs along with other non-nursing home facility types, such as residential care facilities, personal care homes, and adult care facilities. We included these congregate facilities because assisted living facilities represent a large share of their total residents in many states. We intentionally excluded states reporting in broader categories that include nursing homes because these numbers are reported to CMS separately and our goal is to understand the impact of COVID-19 in assisted living facilities to the extent this is possible with current data limitations.

In order to compare increases in COVID-19 cases and deaths overall to increases in cases and deaths occurring in ALFs over the study period, we calculated the percent increases in cases and deaths occurring between June 8 and August 6, 2020 in the 11 states reporting cumulative ALF data in both the beginning and end of the study period, using the state data section of the KFF COVID-19 Coronavirus Tracker.

Notably, states vary widely in reporting data for cases and deaths. Among the 18 states that report ALF cases, 2 states report aggregate cases for residents and staff, 14 states report cases for residents and staff separately, and 2 states report cases for residents only. Among the 14 states that report ALF deaths, 7 states report deaths for residents and staff separately and 7 states report deaths for residents only. Since June, Connecticut and Texas, which were previously reporting cases and deaths for residents only, are now reporting staff cases, but not staff deaths (Texas) and cases and deaths for residents and staff separately (Connecticut). Most states in our analysis (n=14) report cumulative case and death data, however five states (Florida, North Carolina, North Dakota, Tennessee, Utah) report only active cases, which may lead to a significant undercount of cases and deaths that are no longer active. Within the two-month period, North Carolina switched from reporting cumulative cases to active cases.

States differ slightly in how many cases constitute a facility “outbreak” which triggers reporting – some states report data for facilities with just one active case, others begin reporting when two or more cases are reported by facilities. States also differ in whether they report suspected COVID-19 cases and deaths, in addition to confirmed cases – some only report cases confirmed via diagnostic test. For the purpose of this analysis, we have included both suspected and confirmed COVID-19 cases and deaths reported. Additionally, four states (Massachusetts, Rhode Island, Utah, and California) report ranges of cases and/or deaths, versus specific counts. For these states, we used the median of the reported range, or 5 in the case of “5 or above” reported, and 31 for “greater than 30”.

Long-term care (LTC) facilities have experienced a disproportionate share of deaths during the COVID-19 pandemic. The most recently available data show long-term care facilities making up 8% of all coronavirus cases, but more than 40% of all COVID-19 deaths. This analysis evaluates trends of long-term care cases and deaths between April and August 2020 to evaluate where and when the pandemic has hit long-term care facilities the hardest, how the share of cases and deaths attributed to long-term care facilities has changed over time, and whether states continue to report new cases and deaths in these facilities at the same rate now as they did back in April.

This analysis presents trends in long-term care data for states that have reported long-term care cases (35 states) and deaths (36 states) since early April 2020 (see methods and limitations for more details). Federally available, facility-specific data was not used for this analysis because facilities were not required to begin reporting until May 8, which would miss peaks in cases and deaths in April and May. Key findings from our analysis include:

• Nationally, reported cumulative cases and deaths in long-term care facilities have increased over time, increasing from 10,000 deaths and 50,000 cases in mid-April to over 70,000 deaths and nearly 400,000 cases in mid-August. However, trends in cases and deaths vary notably across states. (Interactives 1 and 2). Some states, such as New York and Massachusetts, experienced a surge of long-term care cases and deaths that peaked and plateaued much earlier than other states where long-term cases and deaths were first reported in later months and have continued to climb, such as Texas and North Carolina (Interactives 1 and 2).
• Nationally, the share of deaths attributed to LTC facilities has stayed relatively constant since May, but the share of cases attributed to LTC facilities has decreased over time (Figures 1 and 2). The decreasing share of cases attributed to long-term care facilities is likely due broader availability of testing that contributed to the identification of more cases in the general population, particularly in younger age groups, in June and July.
• Nationally, the number of new LTC cases and deaths peaked in April, decreased in May and June, before rising in July, followed by a rise in new LTC deaths in August (Figure 3 and Table 2). These patterns of new LTC cases and deaths generally follow the patterns of cases and deaths that we have seen nationally, indicating a strong connection between community spread with cases and deaths in long-term care facilities. Recent analysis on cases in hotspot states supports this notion as well. A key distinction here is the number of new LTC cases per 100,000 residents in July and August was much lower than new LTC cases in April and May, which differs from the national pattern where new cases in the summer months were higher than new cases in April and May. (Figure 3).

1.  How do trends in total cases and deaths in long-term care facilities vary by state?

Total cases and deaths in long-term care facilities have increased over time, with notable differences across states (Interactives 1 and 2). Examining visual trend lines in cumulative cases and deaths since April indicates that there are some states, such as New York, Massachusetts, New Jersey, and Connecticut, that have “flattened the curve”, while other states, such as California, Texas, Georgia, and Ohio, have continued to see a rise in long-term care cases and deaths (Interactives 1 and 2).

As of August 20^th, states with the highest burden of COVID-19 long-term care cases and deaths per 100,000 state residents were among those that experienced the earliest peaks in both long-term care and overall cases and deaths (Table 1). New Jersey, Massachusetts, Louisiana, Rhode Island, and Connecticut had the highest burden of cases and deaths in LTC facilities per 100,000 state residents (>250 cases per 100,000 residents and >40 deaths per 100,000 residents) among all states that report data. With the exception of Louisiana, these states were also among the states that experienced relatively large numbers of long-term care cases and deaths in the earlier months of the pandemic (Interactives 1 and 2, Table 2).

In contrast, while long-term care cases and deaths in states like California and Texas continued to rise into the summer months, after other states had flattened their curve (Interactives 1 and 2, Table 2), these states experienced a smaller number of long-term care cases and deaths per 100,000 residents (11-12 deaths and 118 cases per 100,000 state residents) than states that peaked in March/April (>40 deaths and >250 cases per 100,000 state residents) (Table 1). With the number of cases and deaths in these states continuing to rise, it will be important to re-assess the extent to which this finding holds over time.

Interactive 1: 

–

Interactive 2:

–

2.  How has the national share of deaths and cases attributed to long-term care facilities changed over time?

Nationally, the share of deaths attributed to long-term care facilities has stayed relatively constant over time (Figure 1). The share of reported deaths attributed to long-term care facilities increased from 37% the week of April 19^th to 46% the week of May 3^rd and has stayed relatively constant since then. The increase in share of deaths attributed to LTC facilities from April to May may be attributed to better reporting (Figure 1). Notably, 7 additional states started reporting data on deaths between the week of April 19^th and the week of May 3^rd.

In contrast, the share of cases attributed to long-term care facilities has decreased over time. When long-term care data began to be reported by more states in mid-April, 1 in every 7 coronavirus cases was a long-term care facility resident or staff member. By the beginning of May, that share had increased to nearly 1 of every 5 cases. However, since then, the share of cases attributed to residents and staff in long-term care facilities has decreased. The most recent data from mid-August show that 1 in every 10 cases is a long-term care facility case (Figure 2). This decrease can likely be attributed to an increase in testing of the general population, most likely due to the increased community transmission that occurred in June and July.These new cases primarily skewed towards a younger population, so share of cases attributed to long-term care facilities has decreased.

3.  What has happened with new cases and deaths in long-term care facilities nationally and at the state-level?

Nationally, new reported long-term care cases per week were the highest in April and decreased through May and June, before increasing again in July and August. (Figure 3 and Table 2). The number of new long-term care cases nationwide decreased from April to May (16.6 cases to 10.0 cases per 100,000 US residents) and again in June (4.1 cases per 100,000 US residents). However, beginning in July, new cases began to rise again (5.2 cases per 100,000 US residents) and again in August to 6.0 new long-term care cases per 100,000 US residents.

New reported long-term care COVID-19 deaths per week were the highest in April, decreased through May, June, and July, before increasing modestly in August (Figure 3 and Table 2). New weekly long-term care deaths decreased from April to May (3.2 deaths to 2.3 deaths per 100,000 US residents), to June (0.8 deaths per 100,000 US residents), to July (to 0.6 deaths per 100,000 US residents) before increasing slightly in August to 0.8 deaths per 100,000 US residents. 

The number of new LTC cases and deaths per 100,000 US residents was overall lower in July and August than in April and May. This pattern of new LTC cases differs from national new case trends, while the pattern of new LTC deaths is similar to national new death trends (Figure 3 and Table 2). New LTC outbreaks were more severe in the earliest months of the outbreak (primarily in the Northeast) than in more recent months, based on the lower number of new LTC cases and deaths per week (Figure 3 and Table 2). This pattern mirrors new deaths nationally, but is in contrast to national new case trends where, new cases in the general population were higher in July and August than in April and May (Figure 3). It is important to continue to track new LTC cases and deaths in the coming months to understand if we are at the beginning of a continued increase in new LTC cases and deaths that will outpace the rate of new cases and deaths identified in April and May.

Trends in long-term care facilities may mirror trends in community outbreaks, but may also be affected by measures that have been put in place to mitigate the impact of the pandemic on residents and staff. Trends of new cases and deaths, as well as recent analysis of cases and deaths by hotspot state status, support the idea that increased community transmission plays a role in increasing cases and deaths in long-term care facilities. However, nationally, long-term care facilities have seen less severe outbreaks now than they did earlier in the pandemic. While the severity of long-term care cases and outbreaks have decreased, potentially as a result of policies restricting visitors, implementing universal testing of staff and residents, and greater social distancing in communities, it will be important to look at burden of cases/deaths and new cases/deaths again if states experience a rise in cases and deaths in the community.

This analysis is based on data from 38 states plus Washington DC, for a total of 39 states. Within these 39 states, we were able to trend long-term care cases in 35 states and long-term care deaths in 36 states. Data was trended as far back as internal records and publicly available historical data allowed. States were chosen based on where we could reliably trend data. States were excluded from this analysis if they do not report data on cases and deaths in long-term care facilities, if their data is sourced from sporadically released media reports, or if there were data quality issues. The 12 states excluded from the analysis were excluded for the following reasons:

Alaska, Arizona, Hawaii, Missouri, Montana, New Mexico, South Dakota – Not reporting cases and deaths in long-term care facilities

Maine, Nebraska, New Hampshire, Wyoming – Data on cases and deaths in long-term care facilities are sourced from sporadic media reports

West Virginia – State-reported data has severe data quality issues

For all states, we trended the subset of data that would give us the longest reliable trend line. Notable examples of this include Louisiana, where data from non-nursing home long-term care facilities were excluded because they were not consistently reported. In Delaware, data excludes staff cases because that data was not reported consistently. For this reason, this analysis should not be used to identify state-level or national data on total long-term care cases and deaths. The most recent data on total cases and deaths in long-term care facilities can be located here.

Table 1: Long-Term Care Coronavirus Cases and Deaths Per 100,000 US and State Residents As of August 20th, 2020

This table presents the burden of long-term care cases and deaths that each state has experienced as of August 20^th, 2020. Total population data was taken from 2019 state population estimates from the US Census Bureau. The latest long-term care cases/deaths data available was used to calculate the burden of cases and deaths experienced by each state. National long-term care case burden was calculated by summing US population from 35 states that reported case data, dividing total cases by total population in those 35 states, and multiplying by 100,000 to find the value per 100,000 US residents. National long-term care deaths burden was calculated similarly by using data from 36 states that report LTC deaths.

Table 2: Average New LTC Cases and Deaths Per Week Per 100,000 US and State Residents, By Month

Total population data was taken from 2019 state population estimates from the US Census Bureau. The first week of available data for each state was not included in this analysis since the first week of data does not reflect a single week of cases/deaths, but rather all cases and deaths that have occurred up to that point. New cases and deaths were calculated for each week thereafter, and then averaged for all of the weeks within the month. April, June, and July reflect 4 weeks of data. May reflects 5 weeks of data. August reflects 3 weeks of data. These average new cases were converted to represent cases and deaths per 100,000 state residents to allow for easier comparison across states. National new cases and national new deaths were calculated by averaging new cases and new deaths across states. See limitations for more details on this process.

Limitations

There were several possible approaches to this analysis, all of which posed major limitations. This analysis could be limited to the time period where most major states are reporting, which would limit the time period of this analysis to mid-June to present day. However, this approach would miss the major peaks in states such as Massachusetts, Connecticut, and New Jersey. Another option was to limit the analysis to states where we had the earliest weeks of data available. However, this approach would exclude Texas, Michigan, Maryland, and several other states where data was not available until at least several weeks after other states began reporting. Deaths in the states that would have been excluded in this approach make up 30% of all long-term care deaths due to COVID-19. This analysis could have also used the federally reported data, but this would have limited the time period of analysis as well.

Due to data availability and quality issues, we were unable to include all states in this analysis. Thus, all national calculations in this analysis are subject to data availability. In particular, national calculations of new cases and deaths in Table 2 are limited by varying numbers of states included in each week and month, the differences in policies across states, and the lack of reporting comparability. However, given the data limitations, this is the best approximation of new cases and deaths per month.

We conducted a sensitivity analysis for the analysis in Table 2 to see how much using different states in each month affected both the raw Ns and the direction of the trend. Our sensitivity analysis found no notable differences. When looking at the same set of 23 states for new cases and 19 states for new deaths from April to August, we found similar trends as when looking at all available states in respective months. As mentioned above, this method excludes several major states, such as Texas, Michigan, and Maryland.

Another key limitation is related to data on cases. Two large states – Florida and New York – do not report data on cumulative cases. Thus, the national numbers we present for case data are not truly nationally representative. These states report a large number of deaths, so they have likely experienced similarly large numbers of cases.