On Wednesday, August 8, CMS filed a proposed rule clearing the way for the federal government to continue making payments under the ACA’s risk adjustment program for the benefit year 2018. The 2018 proposed rule is unsurprising. It essentially mirrors the final rule CMS issued two weeks ago for benefit year 2017, including the same technical fix to the risk adjustment methodology meant to satisfy a decision handed down in February by a federal district court judge in New Mexico, who had ruled that the use of certain risk management formulas in the program was arbitrary and capricious. Both the 2017 final rule (which will presumably allow CMS to make billions of dollars in risk adjustment payments to plans this fall for benefit year 2017) and the 2018 proposed rule (which is designed to allow payments for benefit year 2018, to be paid out during the fall of 2018) will still need to be approved by the district court before any risk adjustment payments can be made, meaning the future structure of the risk adjustment program is far from set in stone. The New Mexico CO-OP which had challenged the risk adjustment methodology in the first place made clear in a filing on August 1 that it will continue to fight the risk adjustment methodology. However, CMS’s decision to continue administering the risk adjustment program is noteworthy because it represents a stark reversal from the agency’s announcement in early June that the federal government would freeze the program in response to the district court ruling.
The risk adjustment program is but one component of the ACA’s premium stabilization program, which also included reinsurance and risk corridors. All three programs have been the subject of many lawsuits (we wrote about the risk corridors litigation here and here). However, the litigation surrounding the risk adjustment program is perhaps more significant because unlike the other programs, the risk adjustment program is permanent.
The Structure of the Risk Adjustment Program
The structure of the ACA’s risk adjustment program is tedious and complicated, but it is impossible to understand the significance of CMS’s decision, and the litigation that served as its catalyst, without some basic understanding of the program’s structure. The risk adjustment program was designed to transfer funds from plans with lower-risk enrollees (i.e., younger and healthier individuals) to plans with higher-risk enrollees (i.e., older and sicker individuals). HHS created a methodology to estimate the financial risk of a plan’s enrollees, known as an individual risk score, using demographic data, claims for medical diagnoses, and other factors. A plan’s individual risk scores are then averaged across all its enrollees to calculate an average risk score. Plans with low average risk scores are supposed to pay into the program, and plans with high average risk scores receive payments from the program.
The purpose of risk adjustment is to stop insurers from structuring benefits, cost sharing, and other plan components in order to only attract healthy individuals with low expected medical claims. However, the actual administration of the program has been the source of significant scrutiny, particularly in the government’s use of statewide average premium, an important variable in the risk adjustment methodology. Statewide average premium is basically, true to its name, the calculation of the average premium in the applicable exchange marketplaces. The actual calculation of transfer payments to and from plans is calculated by comparing a plan’s average risk score to the statewide average premium. In administering the program, HHS also decided to make the program budget-neutral, restricting payments to insurers from exceeding payments into the program.
Several small CO-OP plans have argued that the use of a statewide average premium penalizes plans that keep premiums low through smaller payments to providers, management of enrollees’ medical care, and reduced administrative costs. As one actuary showed in an example, if there were two silver plans with identical risk scores (and other factors) of .9 in a state with a statewide average premium of $275, each plan would have to transfer the same $18.97 per enrollee into the risk adjustment program. However, if one of the plans had an actual premium of $250 and the other had an actual premium of $275, the plan with the $250 premium would pay a higher percentage of their plan premium into the program (8%) compared to the plan with the $275 premium (7%). HHS’s use of statewide average premium (as opposed to, say, the actual premium a plan charges) to set risk adjustment transfers has the illogical effect of increasing a plans’ risk adjustment liabilities if, all other factors staying the same, that plan has a lower premium.
Risk Adjustment Litigation
As a result, the practical impact of the risk adjustment program is attenuated from the program’s statutory purpose (to make transfers based on a plans actual risk, and reduce cherry-picking healthy enrollees). In the two most high-profile risk adjustment cases to date, Minuteman Health, Inc v. U.S. Department of Health and Human Services et al.and New Mexico Health Connections et al. v. U.S. Department of Health and Human Services et al., two CO-OP plans have anchored their legal arguments around this disparity. In both cases, the plaintiff CO-OPs contended that because the ACA requires that the risk adjustment program assess a charge on insurers based on actuarial risk, the use of statewide average premium to calculate transfer payments violated the plain wording of the ACA. Further, the Plaintiffs charged that the use of statewide average premiums was arbitrary and capricious. In the implementing regulations, HHS had justified the use of statewide average premium based on the programs’ budget-neutrality. However, as the CO-OPs both pointed out, there was nothing in the ACA requiring the program to be budget-neutral in the first place.
After engaging in the typical steps of a Chevron analysis, the court in Minuteman concluded that the ACA did not prohibit HHS from using a statewide average premium in the transfer calculation. The court noted that even the plaintiff in that case never suggested that the ACA required HHS to rely on risk-scores alone. The court also noted that HHS had reasonably considered alternatives (including the plan’s suggestion of using its actual premium), which it had justified as more predictable given the fact that the program was budget-neutral. The court noted that, while nothing in the ACA required that risk adjustment transfers be budget-neutral, the ACA did not prohibit such budget neutrality.
While the New Mexico court also found that HHS’s use of statewide average premium complied with the statutory wording of the ACA, it nonetheless concluded that application of the statewide average premium was arbitrary and capricious. As the court explained in the decision, HHS had justified its use of statewide average premium in the implementing regulations based on the program being budget-neutral. However, HHS had failed to provide any justification for why the program had to be budget-neutral in the first place. The closest HHS ever came to an explanation was a statement in a 2011 White Paper it published in which HHS wrote that the ACA’s “risk adjustment program is designed to be budget-neutral.” The New Mexico court found that explanation insufficient, while the Massachusetts court appeared to take that statement as evidence that HHS had sufficiently considered the alternative of applying the program not as budget-neutral.
The CMS Weathervane
On June 7, CMS announced that it would freeze risk adjustment payments for 2017, which totaled $10.4 billion, in response to the New Mexico Health Connections decision. The immediate and overwhelming response, which came from healthcare policy experts and insurance trade organizations, was perplexed outrage. As many pointed out, CMS still had several options for administering the program, despite the judge temporarily halting the program. The court had ruled that the agency had not properly justified the choice to administer the program as budget-neutral, but CMS could quickly remedy the situation by issuing an interim final rule to provide a rationale. On July 24, CMS apparently took stock of the uproar, and decided to release such a rule, followed by an analogous proposed rule for the 2018.
CMS’s reversal seems to make practical sense in the short term. The amount of money at stake is significant, and many plans had anticipated receiving the payments, which are scheduled to be made during the fall of 2018. The initial decision to halt the program has also rekindled the accusations made by supporters of the ACA that the current Administration is attempting to undermine the program. However, the discussion about the actual structural issues with the risk adjustment program have been somewhat obfuscated by the noise. The arguments made by CO-OPs that the program punished them for offering low-premium plans, along with several other criticisms not addressed in this post, are compelling and legitimate.
Unfortunately, much of the damage to the ACA’s CO-OP program has already occurred—one of the plaintiff’s to the above cases went into receivership in 2017. Maybe the biggest irony is the fact that many of the legal issues and operational challenges with the risk adjustment program developed during the Obama Administration, which was committed to making the ACA work.
 See also http://www.mass.gov/anf/budget-taxes-and-procurement/oversight-agencies/health-policy-commission/public-meetings/annual-cost-trends-hearing/2015/2015-pre-filed-testimony-minuteman.pdf
 Minuteman Health, Inc v. United States Department of Health and Human Services et al., No. 1:16-cv-11570 (D. Mass. Jan. 30, 2018).
 New Mexico Health Connections et al. v. U.S. Department of Health and Human Services et al., No. 1:16-cv-00878-JB-JHR (D.N.M. Feb. 28, 2018).