Bedford County Memorial Hospital v. Health & Human Services

JAMES DICKSON PHILLIPS, Circuit Judge: In these consolidated appeals, the Secretary of Health and Human Services (Secretary) challenges the judgment of two district courts which independently, on the complaints of two sets of hospitals, declared invalid a new regulation, 42 C.F.R. § 405.452(a)(l)(ii) (1984), promulgated by the Secretary that changed the process by which hospitals are compensated for the portion of their malpractice costs attributable to Medicare patients. We agree with the district courts that the regulation is invalid because promulgated in violation of provisions of the Administrative Procedure Act (APA) and the Medicare Act and because, as promulgated, it is arbitrary and capricious. We therefore affirm the determinations of invalidity and remand with directions to enter summary judgment requiring the Secretary to compensate the complaining hospitals under the regulation sought to be superseded. I Two groups of hospitals, Bedford County Memorial Hospital, et al. (Bedford), and Alexandria Hospital, et al. (Alexandria) sued separately in different district courts challenging a regulation promulgated in 1979 by the Department of Health and [1019] Human Services (Secretary). The new regulation, 42 C.F.R. § 405.452(a)(l)(ii), changed the way that the Secretary compensates hospitals for the portion of their malpractice costs attributable to Medicare patients. Under a prior regulation, which had guided reimbursement since 1966, the Secretary treated malpractice costs as one of many overhead expenses that were pooled by Medicare services providers and reimbursed based on the provider’s “utilization ratio,” which is a ratio of usage of the hospital by Medicare patients to total patient usage. The new regulation segregated malpractice costs from general overhead expenses and established a scheme of reimbursement based on actual malpractice claim loss experience of Medicare providers arising from serving Medicare patients.1 The Secretary promulgated the new regulation by notice and comment rulemaking subject to the Administrative Procedure Act (APA), 5 U.S.C. § 553. Both hospital groups challenged the new regulation on three grounds in district court: (1) the Secretary failed to meet the APA, 5 U.S.C. § 553(c) requirement that an agency provide an adequate statement of basis and purpose for informal rules; (2) the rule is arbitrary and capricious; and (3) the rule is contrary to a provision of the Medicare Act, 42 U.S.C. § 1395x(v)(l)(A). The Bedford court entered judgment invalidating the rule for lacking an adequate statement of basis and purpose and as constituting arbitrary and capricious rulemaking. Bedford County Memorial Hospital v. Heckler, 583 F.Supp. 367 (W.D.Va.1984). The Bedford court indicated that it would have found the rule to be contrary to the Medicare Act as well had it needed to reach the issue. See id. at 378 & n. 16. The Alexandria court entered a similar judgment in Alexandria Hospital v. Heckler, 586 F.Supp. 581 (E.D.Va.1984). Both courts remanded the hospitals’ underlying claims for malpractice cost reimbursement to the Secretary without specifying whether the Secretary was to pay Bedford and Alexandria under the old rule, or instead, was to be permitted to attempt correction of the deficiencies in her rule and be allowed to apply it as corrected. These appeals by the Secretary followed and were consolidated by us for hearing and decision. II We note at the outset that the challenged regulation has spawned extensive litigation, and that to date, six other circuits have considered challenges to its validity. All but one have found the rule invalid; and the one remanded with expression of doubt as to validity. See DeSoto General Hospital v. Heckler, 766 F.2d 182 (5th Cir. 1985) (affirming district court judgment insofar as it invalidated the new rule); Lloyd Noland Hospital and Clinic v. Heckler, 762 F.2d 1561 (11th Cir.1985) (affirming district court judgment invalidating the [1020] new rule); St James Hospital v. Heckler, 760 F.2d 1460 (7th Cir.1985) (affirming district court judgment invalidating the new rule); Humana of Aurora, Inc. v. Heckler, 753 F.2d 1579 (10th Cir.1985) (reversing a district court judgment upholding the new rule); Abington Memorial Hospital v. Heckler, 750 F.2d 242 (3d Cir.1984) (affirming district court judgment invalidating the new rule); and Walter O. Boswell Memorial Hospital v. Heckler, 749 F.2d 788 (D.C. Cir.1984) (raising serious questions regarding the Secretary’s rulemaking, but ultimately remanding because the parties failed to submit a complete administrative record for consideration by the district court). We are in essential agreement with all the other circuits that have held the regulation invalid on a variety of grounds. On these appeals, we agree with the district courts that the regulation is invalid because of the Secretary’s failure to provide an adequate statement of basis and purpose in her rulemaking process. The APA, 5 U.S.C. § 553(c), requires that an agency promulgating an informal rule issue a concise statement of basis and purpose that identifies and addresses major issues raised during the comment period. St. James, 760 F.2d at 1469; Boswell, 749 F.2d at 794. The hospitals contend and we agree that the Secretary’s statement failed to address major criticisms of the accuracy and relevance of the Westat study,2 the primary source of malpractice claim data relied on by the Secretary in formulating the new rule, and failed to address alternatives suggested by commenters. The Secretary contends that it gave an adequate statement in noting that malpractice premium expense had inflated much more rapidly than other overhead items since the beginning of the original practice of pooling malpractice costs with other overhead items. Further, she cites the Westat study as establishing that because of his relatively advanced age, lower life expectancy and reduced income potential, the average Medicare patient represented a lower risk of malpractice claim losses than the average patient in the overall patient population. From this the Secretary had derived the regulation’s stated basis that rapidly escalating malpractice costs had now made associated expenses too significant to lump into a crudely allocated overhead pool in which malpractice costs were necessarily then over-allocated to Medicare patients.3 See Bedford, 583 F.Supp. at 379 app. We agree with the district court that this statement of basis was inadequate in view of the complete failure to address criticisms of the Westat study and alternatives raised by commentators. Accord Lloyd Noland Hospital, 762 F.2d at 1566-67; St. James, 760 F.2d at 1469-70. One criticism raised by the hospitals’ comments went to the heart of the reliability of the Westat study’s data and its applicability to the hospitals affected by the new regulation.4 The Westat study includ [1021] ed malpractice claims against both doctors and hospitals whereas only claims against hospitals would be relevant in determining whether the old overhead allocation scheme over-allocated malpractice costs to Medicare patients. The hospitals specifically questioned the over-inclusiveness of the data base by pointing out with supporting data that younger patients tend to sue their doctors more than do older patients and that doctor-defendant cases typically produce larger recoveries than do hospital-defendant cases. If that criticism were valid, the inclusion of doctor-defendant cases in the Westat study overstated any existing difference in malpractice risk for hospitals between older Medicare patients and patients in general, and undermined the stated basis for the new regulation segregating malpractice costs from general overhead.5 Because the accuracy and relevancy of the Westat study was thereby made a critical issue in deciding whether to promulgate the new regulation largely on the study’s basis, it was essential that the criticism be addressed in the statement of basis; but it was not. In the absence of a valid basis for change of regulation, the Secretary was not free to depart from the longstanding policy of including malpractice costs in overhead. See Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29, 41-42, 103 S.Ct. 2856, 2865-66, 77 L.Ed.2d 443 (1983); Humana, 753 F.2d at 1582. Hence, at least the major criticisms of the information forming the basis for the Secretary’s rule should have been addressed in the Secretary’s statement of basis and purpose.6 See Home Box Office, Inc. v. FCC, 567 F.2d 9, 35-36 (D.C.Cir. 1977). A second major flaw in the Secretary’s statement of basis and purpose was the failure to address the alternative advanced by the hospitals of creating separate insurance pools for Medicare and non-Medicare patients. Normally, the failure to discount alternatives would not invalidate a rule because an agency need not pick a theoretical best solution for a problem, but rather, need only implement a reasonable solution. See Mourning v. Family Publications Service, Inc., 411 U.S. 356, 369, 93 S.Ct. 1652, 1660, 36 L.Ed.2d 318 (1973). However, the Medicare statute, 42 U.S.C. § 1395x(v)(l)(A), expressly requires the Secretary to consider the practices of national organizations in determining how to reimburse costs. It is undisputed that the Secretary’s new reimbursement system based on five year loss history at each hospital is not used by any national organization, yet the hospitals’ alternative represents a typical risk pooling scheme used by insurers. In view of the statute, rejection of an alternative that comports with usual practices is a major issue that should have been addressed in the Secretary’s statement.7 Hence we find the Secretary’s rule to be invalid for failure to adhere to the state [1022] ment of basis and purpose requirement of the APA, 5 U.S.C. § 553(c). See APA, 5 U.S.C. § 706(2)(D). However, in order fully to assess the regulation’s impact and determine appropriate relief, we continue into the issues of whether the rule is arbitrary and capricious and contrary to law. Ill Agency action is arbitrary and capricious if the agency relies on factors that Congress did not intend for it to consider, entirely ignores important aspects of the problem, explains its decision in a manner contrary to the evidence before it, or reaches a decision that is so implausible that it cannot be ascribed to a difference in view. Motor Vehicle, 463 U.S. at 43, 103 S.Ct. at 2866. We find that the rule in issue is arbitrary and capricious because the Secretary failed to consider important aspects of the malpractice cost reimbursement problem and implemented a system that is so implausible that it does not represent reasonable administration of the Medicare program. Accord Lloyd Noland Hospital, 762 F.2d at 1567-68; St. James, 760 F.2d at 1465-69. The first flaw in the Secretary’s rulemaking was the failure to consider problems regarding the accuracy and relevancy of the Westat study. Concededly, nothing in the Medicare Act requires the Secretary to treat malpractice costs as a part of overhead; however, the practice of reimbursing the costs under the overhead formula was entrenched at the time of the Secretary’s rulemaking and could not be changed without justification. See Motor Vehicle, 463 U.S. at 41-42, 103 S.Ct. at 2865-2866. The Westat study, as discussed supra, was the primary factual basis for the Secretary’s justification for changing the reimbursement rule, and in addition, was used in setting the actual reimbursement rate for hospitals that lack a five year malpractice loss history. Hence, the accuracy and relevancy of the Westat study was critical to the rulemaking, and the Secretary ignored an important aspect of the reimbursement problem in ignoring problems of over-inclusiveness of the Westat data base.8 Even had the Secretary justified the departure from prior practice, the rule in issue would be arbitrary and capricious because we find it contrary to the evidence and an implausible strategy for more fairly allocating malpractice expenses. The rule bases reimbursement solely on the loss history of the providing hospital, or if the individual hospital lacks a five year loss history, on a national figure derived from the Westat study.9 However, for hospitals protecting themselves with malpractice insurance, approximately forty percent of every premium dollar reflects the administrative expenses incurred by insurance companies rather than dollars paid out on malpractice claims. The Secretary gives no explanation of how a reimbursement formula based solely on loss experience fairly accounts for the portion of premium expenses that represent fixed costs of insurance coverage unrelated to the loss history of particular groups of patients.10 [1023] Since the Secretary’s purpose in promulgating the rule was to more fairly allocate malpractice costs, the adopted system based solely on loss history is contrary to evidence that a large part of premium expense reflects administrative costs. The system therefore cannot be expected to result in fair allocation of malpractice costs, and is arbitrary and capricious. IV In addition to finding the rule arbitrary and capricious, we find it contrary to law. Accord St. James, 760 F.2d at 1470-73. The Medicare Act requires that the Secretary administer the program so that costs of treating Medicare patients will not be borne by non-Medicare patients. 42 U.S.C. § 1395x(v)(l)(A). By basing reimbursement of malpractice costs solely on loss history, the Secretary has failed to take account of administrative expenses, a disproportionate share of which will be borne by non-Medicare patients under the Secretary’s system.11 Hence, the Secretary’s failure to provide for the problem of cross subsidization of administrative costs is contrary to law as well as arbitrary and capricious. The rule is also contrary to law in that the Secretary did not consider alternatives more closely paralleling the practices of other national organizations.12 No national organization determines the cost of malpractice coverage solely on individual hospital loss experience. Rather, similar hospitals are pooled as risks, and other factors such as administrative costs are factored in. Moreover, the Secretary’s failure to consider national practice has led to implementation of a system that appears likely to lead to unfair and illogical results. For instance, a hospital with no loss experience for five years would receive less reimbursement on insurance premium cost than a hospital with a loss experience if many of the claims involved Medicare patients.13 Two hospitals with precisely the same level of claims against them could be subject to widely varying reimbursement rates depending upon whether the hospital’s claims happened to be skewed towards Medicare or non-Medicare patients.14 It thus appears clear that this regulation does not reward safe hospitals, treat similarly situated hospitals similarly, nor allow hospital planners to anticipate the level of reimbursement that they will receive from year to year. The Medicare Act, 42 U.S.C. § 1395x(v)(l)(A) expressly requires the Secretary to take national practices into account, and the failure to do so is contrary to law and arbitrary and capricious, espe [1024] dally in view of the unfair and illogical results to which it obviously might lead. Hence, we agree with the district courts that the Secretary's rulemaking suffered from procedural as well as serious substantive flaws. In our view, the Secretary did not comport with explicit directives of the Medicare Act, proceeded in the face of grave questions regarding the accuracy and relevancy of critical data, and neglected even to consider the obvious flaw in its proposed system for the allocation of administrative costs. With the seriousness of the deficiencies of the Secretary’s rulemaking in mind, we turn to the issue of appropriate relief for the hospital petitioners in this case. V Both district courts found the Secretary’s rule invalid and remanded to the Secretary, but neither made clear whether the remand was to be for further rulemaking procedures or payment based on the old formula. We hold that the proper relief is to remand all claims properly before the district courts for payment under the old rule. Accord, Desoto General, 766 F.2d at 186; Lloyd Noland Hospital, 762 F.2d at 1569. The Secretary contends that she should have the opportunity to correct any deficiencies in the rule by further rulemaking.15 The hospitals contend that since the Secretary’s new rule is invalid, reimbursement should be ordered under the old scheme, which utilized the general overhead formula. While retroactive rulemaking designed to cure defects in regulatory schemes should sometimes be permitted, this should only occur when the public interest in retroactively curing the defect outweighs the harm to private interests resulting from invalid first instance rule-making. See Daughters of Miriam Center for the Aged v. Mathews, 590 F.2d 1250, 1260 & n. 27 (3d Cir.1978). In balancing the interests in this case, the relatively serious deficiencies in the rulemaking and extensive subsequent proceedings that would be required weigh against allowing a retroactive rulemaking. Moreover, only a few years of payments are in question. The hospitals have a legitimate interest in a timely resolution of reimbursement issues that, in view of the deficiencies noted above, outweighs the public interests that could be advanced by a remand for further rulemaking with all reimbursement suspended in the interval. Hence we find that the appropriate relief for claims properly before the district courts is to remand for entry of decrees directing payment forthwith under the old overhead formula. On remand, the district courts should determine which claims have been properly presented in accordance with 42 U.S.C. § 1395oo and remand to the Secretary with instructions consistent with this opinion. VI Having determined that the district courts correctly found the Secretary’s regulation, 42 C.F.R. § 405.452(a)(l)(ii) invalid, we affirm the judgments below. We remand to the district courts for issuance of orders clarifying the relief to be granted to the hospital petitioners. SO ORDERED. Footnotes [1] The new regulation effectively segregated malpractice costs by making malpractice costs an exception to costs reimbursed under the overhead formula: (ii) Exception: Malpractice insurance. For cost reporting periods beginning on or after July 1, 1979, costs of malpractice insurance premiums and self-insurance fund contributions must be separately accumulated and directly apportioned to Medicare. The apportionment must be based on the dollar ratio of the provider's Medicare paid malpractice losses to its total paid malpractice losses for the current cost reporting period and the preceding 4-year period. If a provider has no malpractice loss experience for the 5-year period, the costs of malpractice insurance premiums of self-insurance fund contributions must be apportioned to Medicare based on the national ratio of malpractice awards paid to Medicare beneficiaries to malpractice awards paid to all patients. The Health Care Financing Administration will calculate this ratio periodically based on the most recent departmental closed claim study. If a provider pays allowable uninsured malpractice losses incurred by Medicare beneficiaries, either through allowable deductible or coinsurance provisions, or as a result of an award in excess of reasonable coverage limits, or as a governmental provider, such losses and related direct costs must be directly assigned to Medicare for reimbursement. 42 C.F.R. § 405.452(a)(l)(ii) (1984). [2] The Westat study is a report regarding malpractice claims paid by nine major insurance companies during a four month period of 1976. [3] By statute, the Secretary must establish reimbursement guidelines that do not over or under reimburse providers for services rendered to Medicare patients. 42 U.S.C. § 1395x(v)(l)(A). The Secretary has rejected all previous attempts to segregate expenses from the overhead pool because she believed that although providers were under-compensated for some items, the system worked equitably because providers were over-compensated for other items. Rather than attempt the unjustifiably costly task of precisely allocating each overhead item, the Secretary presumed that overall, no cross-subsidization occurred. The new rule regarding malpractice premiums was to be a “unique exception” due to the great significance and high degree of misallocation of malpractice costs. See Bedford, 583 F.Supp. at 379 app. [4] The Secretary contends that the hospitals failed to raise some of the specific challenges to the Westat data base during the comment period. However, commenters at least raised general questions regarding the Westat study, and apparently, the Secretary did not release the data base until late in the rulemaking process. We find that the comments questioning the Westat study were adequate to put the Secretary on notice that the accuracy and relevancy of the study was a major issue in the rulemaking. The lack of precision in the comments is attributable [1021] to the Secretary’s delay in making the data base available. [5] The hospitals raised other criticisms of the Westat data base that we find are less important. The study was under-inclusive in that it did not include the malpractice experience of self-insured hospitals. However, no reason exists to believe that self-insured hospitals tend to have different loss ratios than insured hospitals, and therefore, exclusion of this data does not appear to have skewed results. The hospitals also note that some responses during the Westat survey did not indicate whether the patient was covered by Medicare. However, the study’s authors adjusted for the incomplete responses by allocating a percentage to Medicare based on statistics concerning Medicare admissions. Finally, Westat surveyed only nine of the largest insurance companies. Nevertheless, the survey achieved broad coverage since these companies carry the great majority of malpractice insurance. [6] A high degree of accuracy and relevancy of the Westat study was essential, for the Secretary relied on Westat data to set the national reimbursement ratio for hospitals without a five year loss history. See supra note 1. [7] The Secretary was not bound to address alternatives proposed by commenters that failed to come closer to comporting with the practices of national organizations than did the Secretary’s new system. [8] The hospitals suggest that the Secretary could not segregate malpractice costs from overhead items without determining whether the remaining items balanced each other so that no cross subsidization would occur. We disagree. Had the Secretary established with reliable data that malpractice costs had grown substantially and were being over-allocated to Medicare patients, the Motor Vehicle requirement of justifying the departure from prior agency practice would have been met. Nothing in the statute requires that malpractice costs be pooled into overhead, and rapidly inflating costs coupled with over-allocation to Medicare patients would justify a new reimbursement system. The Secretary’s error was in failing to consider what appear to be basic problems in its source of information used to establish misallocation. [9] See supra note 1. [10] The Secretary contended that Medicare patients are a lower risk because dollar recoveries tend to be lower due to lower life expectancy and income potential. However, the Secretary presents no evidence that the number of Medicare patient claims is lower on a per capita basis than for patients generally. The administrative costs of processing a claim for Medicare as opposed to non-Medicare patients may be equivalent even if loss history shows that Medicare patients are a lower risk. If Medicare [1023] patient claims are equal in number on a per capita basis to other patients, administrative costs should be equally allocated among all patients even if average dollar recoveries vary according to patient groups. [11] For example, if a hospital’s loss history demonstrates that Medicare patients win seven percent of malpractice claim dollars paid, and the hospital pays $100,000 in malpractice premiums, its reimbursement would be $7,000. If Medicare patients file the same number of claims, but simply win lower dollar recoveries, administrative costs could well be the same for each patient. Assuming administrative costs total forty percent of premiums and risk factors sixty percent, and that Medicare patients account for twenty percent of hospital utilization the hospital should receive reimbursement of $12,200 instead of $7,000 (the hospital should get twenty percent of $40,000 paid for administrative costs and seven percent of $60,000 paid for undertaking risk). [12] The Secretary contends that she considered and rejected alternatives, such as creating separate risk pools for Medicare and non-Medicare patients, in a subsequent procedure. However, no part of that procedure has been made part of the record, and we decline to consider the subsequent procedure in deciding this appeal. [13] The hospital with no loss history would receive reimbursement of approximately seven percent, the national figure established by the Secretary. As an extreme example, a hospital with one loss in its loss history that by luck happened to be to a Medicare patient would receive 100 percent reimbursement. [14] Again, as an extreme example, if both hospitals had one claim in its loss history, but hospital A’s loss was to a Medicare patient whereas hospital B's loss was to a non-Medicare patient, hospital A would receive 100 percent reimbursement whereas hospital B would receive nothing. [15] The Secretary has given notice of its intent to promulgate a new rule retroactive to 1979. See Medicare Program; Payment for the Costs of Malpractice Insurance for Hospitals and Skilled Nursing Facilities; Proposed Rule, 50 Fed.Reg. 25,175 (1985) (to be codified at 42 C.F.R. Part 405).