In plan year 2018, HHS introduced the high-cost risk pool (HCRP) program to augment the existing ACA risk adjustment program, helping mitigate the costs associated with high-cost claimants. HCRP issuer charge trends exceed medical trends and are expected to continue at this level. As high-cost events and treatments increase in frequency, ACA issuers must consider how the HCRP interplays with their strategy.

Background

A fundamental component of the Affordable Care Act (ACA) since full implementation in 2014, risk adjustment transfers funds among issuers based on the risk level of their populations compared to the overall market, focusing competition on quality and plan offerings rather than on issuer enrollees’ morbidity levels.¹ However, the ACA risk adjustment model has historically underpredicted the costs for very high-cost enrollees,² which can incentivize issuers to find ways to avoid these enrollees.

In plan year (PY) 2018, the U.S. Department of Health and Human Services (HHS) established the HCRP program.³ While technically part of the ACA risk adjustment program, the HCRP functions more like specific stop-loss insurance. HHS pools a percentage of each enrollee’s medical and pharmacy claims above a certain threshold. For the first five years of the HCRP, HHS established and maintained a consistent threshold of $1 million and a coinsurance rate of 60%, meaning issuers recover 60% of costs above $1 million. HHS assesses the total nationwide cost as a market-specific percentage of premium for all ACA issuers nationwide.⁴ Consistent with this new program, since PY 2018 HHS has dampened plan liability to reflect only 40% of costs above $1 million per enrollee when calibrating the HHS-hierarchical condition category (HCC) risk score model.

Historical results

The table in Figure 1 summarizes historical HCRP charges assessed to issuers by market.⁵

While initially low as a percentage of premium, the HCRP charge per member per month (PMPM) average annual trend is 15% for both individual and small group, which is significantly higher than premium trends. If HCRP recoveries trended at the same rate as premiums, we would expect the HCRP charge as a percentage of premium to remain constant. As the HCRP parameters remain fixed at a $1 million threshold and 60% coinsurance rate, the recovery trends are leveraged. There are two reasons for this effect, best shown by examples:

• Assume an enrollee’s claim costs in year 1 are $1,100,000, with $100,000 over the HCRP threshold. All else equal, if medical trend is 10%, then costs for the same care the following year would be $1,210,000. The amount over the HCRP threshold would then increase 110% ($210,000 / $100,000 - 1), greatly exceeding the underlying trend of 10%.
• Assume another enrollee’s claim costs in year 1 are $950,000, which are less than the $1 million HCRP threshold. All else equal, if medical trend is 10%, then costs for the same care the following year would be $1,045,000, with $45,000 newly exceeding the fixed $1 million HCRP threshold.

HHS has not provided commentary on whether it plans to raise the HCRP thresholds and/or adjust the coinsurance in future years. Assuming the parameters continue to remain unchanged, the HCRP charges will likely become a larger component of premiums each year going forward.

Considerations for ACA issuers

State-level considerations

Unlike risk adjustment transfers, which are zero-sum at the state and market level, HCRP is zero-sum across the entire nation for each of the two ACA markets (individual and small group). This can lead to an imbalance at the overall state level.

Over the first five years of the HCRP program, two-thirds of states’ recoveries are less than 0.5% of premium in both the individual and small group markets, with the number of such states decreasing over time to under half in 2022. However, some states have significantly higher HCRP recovery levels almost every year, such as Alaska for individual and Wyoming for small group. Other states have significantly lower (and sometimes zero) HCRP recoveries almost every year, such as Rhode Island for individual and Mississippi for small group. This variability among states is expected, given differences in reimbursement rates (e.g., high-cost hospital rates), morbidity, and general volatility for smaller states.

HCRP recovery consistency

We reviewed historical HCRP recovery patterns by Health Insurance Oversight System (HIOS) issuer ID and observed the following, which may be helpful in estimating future HCRP recoveries within ACA rate filings when the experience year is known but the projection year (two years after the experience year) is not known. Please see the table in Figure 2 for additional details. Note that this table excludes issuers that exit the market before each time period.

• 85% of issuers that have an HCRP recovery in a given year (for example, 2019) will have an HCRP recovery two years later (2021 in this same example). However, given the volatility of HCRP recoveries in magnitude, more than half of issuer-level HCRP recoveries will increase or decrease by at least 50% over a two-year period.
• 78% of issuers that do not have an HCRP recovery in a given year will also not have an HCRP recovery two years later.

Figure 2 shows that issuers with an HCRP recovery in a given year are likely to have an HCRP recovery again in future years, and issuers without an HCRP recovery are likely to continue without a recovery.

High-cost enrollee characteristics

We reviewed a large sample of historical enrollment and claims for ACA individual and small group enrollees to better understand the characteristics of high-cost enrollees. While a wide range of factors can lead to high claims, we note the following:

• The typical high-cost enrollee has many conditions, evidenced by HCCs and/or prescription drug categories (RXCs). However, none of the high-cost enrollees reviewed would receive enough risk adjustment transfer collections and HCRP recoveries on their own to cover their claims. This is likely due to HCCs and RXCs representing the average cost of enrollees with the given condition, rather than enrollees with higher-than-average claim levels.
• In our sample of enrollees with claim levels above the HCRP threshold, almost all meet the $1 million claim threshold on either medical claims or pharmacy claims alone.
  • The majority of high-cost enrollees have very large medical claims associated with severe hospitalizations. These enrollees have many conditions, including acute conditions, severe/trauma conditions, and complications.
  • Other high-cost enrollees have over $1 million in pharmacy claims. These enrollees have fewer HCCs and lower risk adjustment receipts than high-cost enrollees with high medical claims. With the rise of high-cost specialty drugs, ACA issuers should review new and emerging treatments as they contract with providers and pharmacy benefit managers (PBMs) to design benefits, formularies, and clinical guidelines.

We note the following limitations on the above characteristics:

• Filtering to enrollees above $1 million paid claims reduces the sample considerably, so our conclusions may not be representative of all high-cost enrollees.
• We calculated risk adjustment transfers based on nationwide average risk adjustment parameters, so results are likely to vary by state. For example, an enrollee with a very high risk score in a state with a low statewide average risk score would receive higher risk adjustment transfer collections than we calculated in the results above.
• Our analysis of revenue sufficiency did not consider prescription drug rebates or administrative costs.

Incorporating the HCRP into ACA strategy

• Prudent issuers should consider the dynamics and interplay among the incidence and characteristics of high-cost events, the HCRP, and risk adjustment to mitigate financial risk while ensuring proper quality and care coordination:
• HCRP recoveries and risk adjustment receivables are unlikely to cover costs for high-cost enrollees individually. While this is a normal risk of insurance, ACA issuers enrolling populations with higher medical needs need to understand their risk properly. Even an ACA issuer with a relatively healthy population could have catastrophic claims unrelated to general morbidity.
• Methods that can control spending for potential high-cost enrollees are more effective than relying on ACA programs alone. Care management, disease management, and hospital contracting are essential, both before and after the HCRP threshold. A good starting point is to evaluate whether any current enrollees incurred high claims recently (e.g., greater than $1 million in the past year) and, if so, to incorporate measures to manage the enrollee’s care and costs.
• Prescription drugs can also push enrollees over the HCRP threshold. While the HCRP currently does not consider prescription drug rebates, issuers should incorporate expected rebates into their analyses. Additional pharmacy management programs may mitigate catastrophic pharmacy costs.
• A well-structured commercial ACA reinsurance contract should incorporate the HCRP. For example, issuers may target a different coinsurance before and after the HCRP attachment threshold. Issuers ignoring the HCRP may inadvertently purchase more coverage than they need, paying unnecessary reinsurance premiums.
• Issuers need to incorporate appropriate HCRP projections into their premium rates and ensure that all valid HCRP-eligible claims are accepted to the EDGE server prior to the deadline.

Conclusion

The HCRP is becoming a larger component of ACA premiums over time, as recoveries and charges trend high. Consider historical state-specific and issuer-specific results when projecting future HCRP results. Also consider the HCRP when developing ACA pricing and strategy to understand the dynamics of high-cost claimants, mitigate risk, and maximize the chance of success.

Caveats and Limitations

The analysis in this paper is based on data from a variety of sources, including publicly available data and our interpretation of guidance from the Centers for Medicare and Medicaid Services (CMS). We have not audited or verified this data and other information. If the underlying data or information is inaccurate or incomplete, then the results of our analysis may likewise be inaccurate or incomplete.

Our findings will change if CMS changes the current HCRP parameters. We do not have access to detailed claims data used by CMS in determining HCRP recoveries and charges. We used a sample from a large nationwide dataset of ACA membership to develop all summaries at the claimant level.

Guidelines issued by the American Academy of Actuaries require actuaries to include their professional qualifications in all actuarial communications. Peter Fielek and Cameron Gleed are members of the American Academy of Actuaries and meet the qualification standards for performing the analyses in this report.

---

¹ The full text of the regulation is available at https://www.govinfo.gov/content/pkg/FR-2013-03-11/pdf/2013-04902.pdf.

² CMS (October 26, 2021). HHS-Operated Risk Adjustment Technical Paper on Possible Model Changes. Retrieved August 30, 2023, from https://www.cms.gov/files/document/2021-ra-technical-paper.pdf.

³ The full text of the regulation is available at https://www.govinfo.gov/content/pkg/FR-2016-12-22/pdf/2016-30433.pdf. The HCRP program was announced in 2016 and was implemented in 2018.

⁴ Premium charges are calculated separately for individual (including catastrophic, non-catastrophic, and merged market plans) and small group markets.

⁵ CMS. Premium Stabilization Programs. Retrieved August 30, 2023, from https://www.cms.gov/CCIIO/Programs-and-Initiatives/Premium-Stabilization-Programs.