Hmo Louisiana, Inc. v. Department of Health and Human Services

Opinion

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

HMO LOUISIANA, INC.,

Plaintiff,

v. Case No. 24-cv-2931 (CRC)

DEPARTMENT OF HEALTH AND HUMAN SERVICES,

Defendant.

MEMORANDUM OPINION

Plaintiff HMO Louisiana, Inc. (“HMOLA”) offers private insurance plans to Medicare

beneficiaries as an alternative to traditional government-managed plans. Each year, the Centers

for Medicare and Medicaid Services (“CMS”) issue “star ratings” evaluating the quality of these

plans. The ratings help beneficiaries compare plans in the marketplace.

In late 2024, after HMOLA combined two of its plans, CMS issued a 3.5 star rating to the

consolidated plan. Dissatisfied with this rating, HMOLA challenges CMS’s method of

calculating it under the Administrative Procedure Act (“APA”). Before the Court are the parties’

dueling motions for summary judgment. For the following reasons, the Court will grant the

government’s cross-motion for summary judgment and deny HMOLA’s motion.

I. Background

A. The Star Ratings Program

Medicare enrollees may elect to receive benefits under Part C of Medicare, commonly

known as the “Medicare Advantage” (“MA”) program. See 42 U.S.C. § 1395w-21. Under that

program, CMS contracts with private insurance companies—MA Plans—to provide Medicare-

covered benefits to enrollees. See 42 C.F.R. § 422.4; 42 U.S.C. § 1395w-23. MA Plans receive annual star ratings from CMS evaluating the quality of their services.

Am. Compl. ¶ 4. These ratings, in turn, are used by Medicare beneficiaries to shop for plans. Id.

The ratings are intended to be “a true reflection of plan quality and enrollee experience” based on

“complete, accurate, and reliable” data. 83 Fed. Reg. 16,440, 16,520–21 (Apr. 16, 2018). CMS

displays the star ratings in its online and print resources available to Medicare beneficiaries,

including the online Medicare Plan Finder tool. See, e.g., 42 C.F.R. § 422.166(h). CMS also

uses the star ratings to determine the bonuses paid to MA Plans. See 42 U.S.C. § 1395w-23(o);

42 C.F.R. § 422.160(b)(2). If a plan’s star ratings drop too low, CMS may terminate it from the

MA program altogether. Id. § 422.510(a)(4)(xi).

CMS regulations establish the methodology used to calculate annual star ratings. See 42

C.F.R. §§ 422.162(b), 422.166. The ratings strive to treat each contract “fairly and equally.” 83

Fed. Reg. at 16,521. CMS also publishes Technical Notes containing more granular detail on

how ratings are calculated, including those for the up to 40 performance measures that comprise

the overall rating for each plan. Joint Appendix (“JA”) 9; 42 C.F.R. § 422.164(a). Star ratings

primarily use measurement data from two years before their label year. For example, the 2025

star ratings were published in late 2024, using mostly 2023 measurement-year data. JA 36–113;

HHS Opp’n at 5–6.

MA Plans may consolidate two or more contracts—that is, combine them into a single

contract beginning the next contract year. 42 C.F.R. § 422.162(a). Combined contracts that no

longer exist at a contract year’s end are known as consumed contracts. JA 33. The contract that

continues after consolidation is known as the surviving contract. Id. Following consolidation,

all beneficiaries still enrolled in the consumed contract(s) are moved to the surviving contract.

Id.

2 For the first two years after consolidation, while the agency is still relying on pre-

consolidation measurement data, CMS averages the scores of the consolidated contracts. 42

C.F.R. § 422.162(b)(3); 42 U.S.C. 1395w-23(o)(4)(D). Specifically, it calculates the enrollment-

weighted mean of each contract’s scores using its July enrollment. 42 C.F.R. §

422.162(b)(3)(iv).

B. HMO Louisiana

HMOLA is one of the nation’s largest MA health plans, currently serving approximately

30,000 members in Louisiana. Am. Compl. ¶ 2. In 2024, HMOLA’s parent company, Louisiana

Health Service & Indemnity Company d/b/a Blue Cross and Blue Shield of Louisiana,

consolidated two of its contracts—H6453 and H5576. Id. ¶ 3. H6453 is the surviving contract.

During 2023, contract H5576 offered a Special Needs Plan (“SNP”). Id. ¶ 67. SNPs

provide benefits and care designed specifically for people with enumerated chronic diseases,

those living in institutions such as nursing homes, and those who also have Medicaid. JA 45,

196. Contract H6453 did not offer an SNP during that period. Am. Compl. ¶ 68. Star rating

measure C05, “Special Needs Plan Care Management,” assesses the quality of the SNP offered

by a contract. Am. Compl. ¶ 51. When a contract does not offer an SNP, or its SNP has fewer

than 30 enrollees, measure C05 is marked as “no data available” and excluded from the final

rating calculation. JA 46; HMOLA Mot. Summ. J. at 2. H6453, the contract that survived after

consolidation, offers an SNP for 2025. HHS Opp’n at 1.

When CMS initially calculated contract H6453’s 2025 star rating, it did not include a

score for measure C05, having interpreted its Technical Notes to exclude the consumed

contract’s data for that measure. JA 215, 217. HMOLA notified CMS of this potential error,

3 asking CMS to permit it to submit C05 data which it claimed would result in an increase of this

measure “from no star rating to 96% (5 Stars).” Id. HMOLA believed that if CMS included the

C05 data, the overall rating for its surviving contract would increase from 3.5 to 4 stars. Id.

CMS initially refused HMOLA’s request. JA 217. But after HMOLA put forth a strong

argument that excluding C05 conflicted with the applicable regulations and made no sense, see

JA 219–20, CMS agreed to accept HMOLA’s 2023 C05 data corresponding to the consumed

contract, H5576. JA 225–26. That data earned a three-star rating. Id. Since the surviving

contract, H6453, did not offer SNPs in 2023, it did not factor into the enrollment-weighted

average. Id. Therefore, CMS gave the consolidated contract a three-star rating on measure C05,

noting “a significant decline in the measure score from the prior year[.]” Id. Dismayed by the

unexpectedly low rating, HMOLA did an about face: It now argues that the C05 data it

explicitly asked CMS to consider should not have been factored in, after all. Am. Compl. ¶ 6.

II. Legal Standard

At summary judgment, the Court must determine whether the challenged agency action

complies with the APA and is supported by the administrative record. Richards v. INS, 554 F.2d

1173, 1177 (D.C. Cir. 1977). Under the APA, “[t]he reviewing court shall . . . hold unlawful and

set aside agency action, findings, and conclusions found to be arbitrary, capricious, an abuse of

discretion, or otherwise not in accordance with law[.]” 5 U.S.C. § 706(2)(A). Arbitrary and

capricious review is “narrow,” Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402,

416 (1971), and precludes the Court from “substitut[ing] its judgment for that of the agency,”

Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43

(1983). Rather, the Court must determine whether the agency “examine[d] the relevant data and

articulate[d] a satisfactory explanation for its action including a rational connection between the

4 facts found and the choice made.” Id. (internal quotation marks omitted). Even if the agency did

not fully explain its decision, the Court may uphold it “if the agency’s path may reasonably be

discerned.” Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286 (1974)

(citing Colo. Interstate Gas Co. v. Fed. Power Comm’n, 324 U.S. 581, 595 (1945)). The Court’s

review is limited to the administrative record, Holy Land Found. for Relief & Dev. v. Ashcroft,

333 F.3d 156, 160 (D.C. Cir. 2003), and the party challenging an agency’s action bears the

burden of proof, City of Olmsted Falls v. FAA, 292 F.3d 261, 271 (D.C. Cir. 2002).

III. Analysis

HMOLA asserts that CMS acted arbitrarily and capriciously by including data on

measure C05 in its star rating, in violation of CMS regulations, the applicable statute, and the

agency’s technical guidance. HMOLA Mot. for Summ. J. at 9–10. HMOLA further contends

that CMS’s explanation for accepting its request to include the data was insufficient. Am.

Compl. ¶ 96. Because none of CMS’s rules, the relevant statute, or the Technical Notes supports

C05’s exclusion, HMOLA’s argument fails.

A. CMS’s Statutes and Rules

HMOLA first contends that CMS’s inclusion of measure C05 data from the consolidated

contract conflicts with its governing statute and regulations. Am. Compl. ¶ 74. To the contrary,

CMS’s inclusion of measure C05 in its calculation is entirely consistent with both.

The statute provides that, following consolidation, star ratings shall be adjusted “to reflect

an enrollment-weighted average of scores or ratings for the continuing and closed contracts.” 42

U.S.C. 1395w-23(o)(4)(D). The regulations add further specificity, mandating that, “[i]n the

case of contract consolidations involving two or more contracts for health or drug services of the

same plan type under the same parent organization, CMS assigns Star Ratings for the first and

5 second years following the consolidation based on the enrollment-weighted mean of the measure

scores of the surviving and consumed contract(s)[.]” 42 C.F.R. § 422.162(b)(3)(i). As relevant

here, for the second year after consolidation, “CMS uses the enrollment-weighted measure scores

using the July enrollment of the measurement year of the consumed and surviving contracts for

all measures[.]” 42 C.F.R. § 422.162(b)(3)(iv)(B)(1).

Here, the consumed contract earned three stars for measure C05 based on the applicable

enrollment data, while the surviving contract did not offer SNPs and accordingly had no data

available for that measure. JA 217. Thus, the “enrollment-weighted mean” of the two scores is

three stars—three stars for the consumed contract, and no score for the surviving contract. That

is precisely what the regulation requires.

In the face of this plain language, HMOLA argues that it is somehow inconsistent with

CMS regulations to “issue[] Star Ratings for a plan that did not actually provide the services

supposedly reflected in the ratings.” HMOLA Mot. for Summ. J. at 10. But as just explained,

the regulations explicitly contemplate the use of data from a consumed contract in calculating a

consolidated contract’s scores for the first two years. 42 C.F.R. §§ 422.162(b)(3)(i), (iv)(B)(1).

Broadening its argument, HMOLA complains that CMS’s calculation is “inconsistent

with [its] statutory mandate” because it does not represent a “true reflection” of HMOLA’s

“services and care.” HMOLA Mot. for Summ. J. at 10–11. Once again, HMOLA’s argument

relies on the fact that the surviving contract “did not even offer [SNPs] during the measuring

period[.]” Id. at 11 (emphasis in original). But the consolidated contract is made up of two

components. One of them did offer SNPs in 2023. Not scoring C05 would therefore render the

overall rating less reflective of the contract’s quality, not more.

6 Worse, HMOLA’s methodology would encourage MA Plans to artificially inflate their

scores by strategically consolidating contracts. The relevant statute appears designed squarely to

prevent such gamesmanship. It is titled: “Special rule to prevent the artificial inflation of star

ratings after the consolidation of Medicare Advantage plans offered by a single organization.”

42 U.S.C. 1395w-23(o)(4)(D). Without at all suggesting that HMOLA combined the two

contracts to whitewash the performance of contract H5576’s legacy SNP, accepting HMOLA’s

current litigation position would bring about precisely that result.

B. The Technical Notes

HMOLA claims that the consideration of measure C05 also violated the agency’s 2025

Technical Notes. HMOLA predicates this argument on the assertion that the Technical Notes

bind CMS. Am. Compl. ¶ 50. CMS does not disagree. Rough Tr. at 27:7–21. Even assuming

the Technical Notes are binding, however, CMS did not violate them here.

HMOLA first maintains that the inclusion of C05 data contravenes a provision of the

Technical Notes stating that “[c]ontracts and [plan benefit packages] with an effective

termination date on or before [June 15, 2024] are excluded and [the C05 measure is] listed as

‘No data available.’” JA 45. Because H5576 “effectively terminated at its merger on January 1,

2024,” says HMOLA, including its C05 data violates the provision. HMOLA Reply at 3;

HMOLA Mot. for Summ. J. at 6.

Contrary to HMOLA’s argument, contracts cannot possibly “terminate” for purposes of

this provision when consumed. As just explained, the regulatory scheme requires CMS to

consider data from a consumed contract for two years post-consolidation. 42 C.F.R.

§ 422.162(b)(3)(i). Contract consolidations are always effective January 1. See JA 220.

7 Accordingly, if this provision of the Technical Notes applied to consumed contracts, their data

could never be considered—in violation of CMS’s regulations.

HMOLA itself made this point before its change in position. It observed that termination

upon consumption “would negate the entire regulatory framework for calculating Star Ratings

following a contract consolidation, which aims to prevent [MA organizations] from

consolidating a low performing contract into a higher performing contract in order to shed the

low scores.” JA 220. That position makes good sense. As already noted, a consumed contract

does not stop providing services upon consolidation—it continues to offer them through the

consolidated contract.

At oral argument, HMOLA countered that while the regulations generally require

consideration of data from consumed and surviving contracts, each measure’s specific exclusions

control. Rough Tr. at 10:8–24. Here, per the Technical Notes, measure C05 excludes data from

contracts terminated before June 15, 2024. HMOLA says that trumps the general requirement to

base star ratings on the consolidated and surviving contracts’ enrollment-weighted average.

HMOLA misapplies the specific-governs-the-general canon, however. The regulations

governing the calculation of ratings following a contract consolidation cover the specific

situation at issue here. The Technical Notes’ C05 termination provision, on the other hand,

applies across the board, irrespective of consolidation. JA 45. And another provision of the

Technical Notes clarifies exactly which measures require consideration of data from both the

consumed and surviving contract following a consolidation:

In the second year following a consolidation, the measure values for the surviving contract of a consolidation are as reported for CAHPS, call center, HOS, and HEDIS measures. For all other measures, the measure values for the surviving contract of a consolidation are calculated as the enrollment weighted mean of all contracts in the consolidation.

8 JA 122. The provision explicitly lists the measures to be excluded from ordinary post-

consolidation calculations, then reiterates that the general enrollment-weighted mean rule applies

to all other measures. Given that C05 is not excluded, the general consolidation formula applies

to that measure like any other.

Taking a different tact to argue that contract H5576 was “terminated,” HMOLA insists

that the definition of “consumed contract” implies termination. Rough Tr. at 9:16–23. The

regulatory definition of “consumed contract” is “a contract that will no longer exist after a

contract year’s end as a result of a consolidation.” 42 C.F.R. § 422.162(a); Rough Tr. at 9:16–

23. But conflating consumed and terminated contracts is inconsistent with CMS’s regulations,

which set forth only three types of termination: (1) by “written mutual consent” between CMS

and the MA organization; (2) by CMS’s sole determination; or (3) by the MA organization if

“CMS fails to substantially carry out the terms of the contract.” 42 C.F.R. § 422.508–12.

Consolidation is not among them. And HMOLA has given no indication that any of the listed

scenarios applies here. Moreover, all contemplated terminations carry sanctions or conditions on

the MA organization, none of which appear to apply to HMOLA. See, e.g., 42 C.F.R. §

422.508(c), (d); id. § 422.510(e); id. § 422.512(e).

Finally, HMOLA argues that the Technical Notes prohibit CMS from considering data

for services that the surviving contract did not offer pre-consolidation. Am. Compl. ¶ 59. But as

already explained, HMOLA’s consolidated contract effectively offered SNPs during the

measurement period through one of the two component contracts, H5576. Nowhere do the

Technical Notes distinguish between whether a service was offered by the surviving or

consumed contract pre-consolidation. Accordingly, the Technical Notes offer no additional

support for HMOLA’s claims.

9 C. CMS’s Explanation for Changing Course

In a last-ditch effort, HMOLA shifts gears, focusing not just on CMS’s supposedly

incorrect methodology, but the contention that it insufficiently explained its change in approach.

Rough Tr. at 30:4–17; id. at 31:4–10. As an initial matter, is not entirely clear that the change-

in-position doctrine applies to an agency’s preliminary position during an informal adjudication

of the sort that CMS engaged in to calculate HMOLA’s star rating. That doctrine traditionally

applies “when an agency shifts from a position expressed in a more formal setting.” Food &

Drug Admin. v. Wages & White Lion Invs., L.L.C., 145 S. Ct. 898, 918 n.5 (2025) (citing FCC

v. Fox Television Stations, Inc., 556 U.S. 502, 517 (2009)). Nevertheless, the Supreme Court

recently assumed without deciding “that the change-in-position doctrine applies to an agency’s

divergence from a position articulated in nonbinding guidance documents.” Id. But HMOLA’s

challenge goes a step further than that. HMOLA is not arguing that CMS diverted from a

position laid out explicitly in the Technical Notes, merely that the agency changed its

interpretation over the course of a single informal adjudication. HMOLA cites no evidence

indicating that CMS had an “existing policy” of excluding the consumed contract’s C05 data

after consolidation. Indeed, CMS’s initial calculation here appears to be best understood as a

one-off error 1—not an “existing precedent” as is typically at issue in change-in-position

challenges. E.g., Fox Television, 556 U.S. at 510. Accordingly, the Court doubts that the

change-in-position doctrine applies under these circumstances.

1 Although not part of the administrative record, the government indicated at oral argument that permitting HMOLA to submit the data required CMS to make changes to its online portal. That may either support that the agency changed its policy or indicate that the agency has not confronted this issue before. Rough Tr. at 28:6–9. Either way, it does not change the Court’s conclusion.

10 Still, even if HMOLA’s shift is viewed as a change in an established policy, CMS has

done just enough to explain its shift. “[A]gencies are free to change their existing policies as

long as they provide a reasoned explanation for the change, display awareness that they are

changing position, and consider serious reliance interests.” White Lion, 145 S. Ct. at 917

(cleaned up). After HMOLA requested that CMS recalculate its scores using the C05 data, the

agency agreed: “At Louisiana Blue’s request and after further consideration, CMS agreed to

accept H5576’s 2023 SNP [data].” JA 225; Rough Tr. at 18:12–20. To be sure, this explanation

standing alone does not shed much light on the agency’s reasoning. In context, however, CMS

essentially adopted HMOLA’s rationale for requesting the inclusion of C05 data. In the same

exchange, just prior, HMOLA explained that under CMS’s regulations, H5576 had not been

terminated and its data was not subject to the Technical Note requiring exclusion. JA 219–20.

Although CMS could have provided a more fulsome explanation for its actions, the Court “may

reasonably [] discern[]” that CMS acquiesced to HMOLA’s request for the reasons it gave the

agency. 2 Bowman, 419 U.S. at 286.

2 Further supporting the Court’s holding is the practical reality that, if CMS’s explanation were found to be insufficient, the Court would remand for further explanation, not vacate the rating. See Massachusetts v. U.S. Nuclear Regul. Comm’n, 924 F.2d 311, 336 (D.C. Cir. 1991) (“In appropriate cases, we will remand without vacating an agency's order where the reason for the remand is a lack of reasoned decisionmaking.”). At that point, CMS would likely be able to easily cure any defect by explaining that it adopted HMOLA’s reasoning in making the requested change.

11 IV. Conclusion

For these reasons, the Court will grant the government’s cross-motion for summary

judgment and deny HMOLA’s motion. A separate order accompanies this Opinion.

CHRISTOPHER R. COOPER United States District Judge

Date: July 9, 2025