Normand F. Pizza, Nyda Brook Zelenka, Samson Livingstone Sempasa, Michael Allyn Stroud, Fred J. Cassibry, Brook, Morial, Cassibry, Pizza & Adcock, et al., New Orleans, LA, for plaintiffs-appellants.

S. William Livingston, Jr., Mark Elliott Plotkin, Covington & Burling, Washington, DC, P. Bruce Waters, TA Dept. of Health & Hospitals, Jerry L. Phillips, Mary Dozier O'Brien, Baton Rouge, LA, for defendants-appellees.

Appeal from the United States District Court for the Middle District of Louisiana.

Before EMILIO M. GARZA and DeMOSS, Circuit Judges and Zagel1, District Judge.

ZAGEL, District Judge.

A penny saved is a penny earned. That is the formula for federal Medicaid law--hospitals that save dollars by operating efficiently and economically earn state and federal dollars to cover all operating costs. The Medicaid Act,2 specifically the Boren Amendment, provides that hospitals in participating states that operate "efficiently and economically" are entitled to reimbursement of costs which must be incurred. 42 U.S.C. Sec. 1396a(a)(13)(A) (1991). Louisiana's Medicaid plan adopts the same formula since Louisiana elected to participate in the joint federal-state Medicaid program and receive matching federal funds. Louisiana's Department of Health and Hospitals ("LDHH"), under the direction of its secretary, administers its Medicaid plan.

As a participating state, Louisiana must comply with the Medicaid Act and implementing regulations promulgated by the Health Care Financing Administration (HCFA). Amisub, (PSL), Inc. v. Colorado Dep't of Social Servs., 879 F.2d 789, 794 (10th Cir.1989), cert. denied, 496 U.S. 935, 110 S. Ct. 3212, 110 L. Ed. 2d 660 (1990). The federal structure gives each state Medicare agency a certain degree of flexibility in developing its Medicaid plan. Each plan, however, must provide for the reimbursement of inpatient hospital services:

through the use of rates ... which the State finds, and makes assurances satisfactory to the Secretary, are reasonable and adequate to meet the costs which must be incurred by efficiently and economically operated facilities in order to provide care and services in conformity with applicable State and federal laws, regulations and quality and safety standards.

42 U.S.C. Sec. 1396a(a)(13)(A). In fact, the Supreme Court in Wilder v. Virginia Hosp. Assoc., 496 U.S. 498, 110 S. Ct. 2510, 110 L. Ed. 2d 455 (1990), held that the leeway in adopting a method of computing rates does not relieve States of their obligation to pay reasonable rates. Id. at 2520. Once developed, each state must submit its plan to HCFA for approval. 42 U.S.C. Sec. 1396. To secure HCFA approval, each state Medicare Agency must make findings and submit assurances to HCFA that: (1) the payment rates "are reasonable and adequate to meet the costs that must be incurred by efficiently and economically operated providers"; (2) the methods and standards employed "take into account the situation of hospitals which serve a disproportionate number of low income patients with special needs"; and (3) the payment rates "are adequate to assure that recipients have reasonable access, taking into account geographic location and reasonable travel time, to inpatient hospital services of adequate quality." 42 C.F.R. Sec. 447.253(b)(1)(i), (ii)(A), (ii)(C) (1992). Such findings must be made and assurances filed with every amendment to established plans; findings must be made at least annually. 42 C.F.R. Sec. 447.253(a), (b) (1992).

LDHH first developed its Medicaid plan for inpatient hospital services in 1983. Under the plan, LDHH reimburses hospitals 100 percent of all capital costs, educational expenses, and malpractice expenses. The remaining operating costs are reimbursed either on a 100 percent basis or at a maximum level predetermined by each hospital's "target rate." LDHH set each hospital's initial "target rate" as the higher of its 1980 and 1981 average operating costs per Medicaid discharge. The plan allowed LDHH to increase these target rates in 1982, 1983, and 1984 in accordance with HCFA's inflation index published periodically. In 1985 and 1986, LDHH submitted proposed amendments to freeze the target rates for cost reporting periods beginning July 1, 1985 through June 30, 1987. HCFA approved this freeze. In 1987, LDHH resumed its plan and increased target rates up to 2.3% under the HCFA index. In 1988, LDHH again froze target rates until July 1, 1990, despite HCFA's disapproval of the proposed amendment.3 Since then, target rates have increased annually by the amount of the applicable HCFA indices.

The dispute in Louisiana concerns whether LDHH made findings and submitted assurances as required by the Boren Amendment. The Hospitals4 here complain that LDHH did not apply the "penny saved, penny earned" formula outlined in the Medicaid Act in deriving the reimbursement rates set under Louisiana's initial Medicaid plan and amendments for the years 1985, 1986, 1988, 1989 and 1990. The Hospitals filed a Sec. 1983 action against the Secretary of LDHH and other agency officials, claiming their actions deprived them of rights secured under the Boren Amendment. The Hospitals eventually moved for partial summary judgment declaring that LDHH failed, as a matter of law, to comply with the Boren Amendment when it established reimbursement rates and other payment schedules under Louisiana's Medicaid plan. In the motion, the Hospitals challenged LDHH's assurances submitted to HCFA, insisting that LDHH failed to make any "findings" that the rates set were reasonable and adequate to meet the costs that must be incurred by efficiently and economically operated hospitals.

LDHH followed suit and filed a cross motion for partial summary judgment declaring that it complied with the findings process mandated in the Medicaid Act and its regulations. The district judge granted LDHH's motion for partial summary judgment, concomitantly denied the Hospitals' motion for summary judgement, dismissed the case in its entirety, and subsequently denied the Hospital's motion for a new trial, but amended his prior ruling.5 The Hospitals now appeal. This Court has jurisdiction to hear their appeal under 28 U.S.C. Sec. 1291.

On appeal, the Hospitals seek review of the district court's grant of summary judgment and other adverse rulings. The Hospitals maintain that the district court erred when it: (1) applied the highly deferential "arbitrary and capricious" standard of judicial review to the procedural issue of whether the LDHH complied with federal law; (2) determined that, as a matter of law, LDHH complied with the Boren Amendment and was entitled to summary judgment; and (3) dismissed the entire case, including the substantive issues on the reasonableness and adequacy of the reimbursement rates, after ruling only on the preliminary issue of procedural compliance with the Boren Amendment.

We review a district court's grant of summary judgment de novo, employing the same standard as a district court would employ under Federal Rule of Civil Procedure 56(c). Harbor Ins. Co. v. Urban Constr. Co., 990 F.2d 195, 199 (5th Cir.1993). Summary judgment is proper only if no genuine issue of material fact exists and the moving party is entitled to judgment as a matter of law. FED.R.CIV.P. 56(c); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250, 106 S. Ct. 2505, 2511, 91 L. Ed. 2d 202 (1986). All reasonable inferences are drawn in favor of the nonmoving party. Harbor, 990 F.2d at 199. The parties here do not dispute the material facts. They argue whether the facts of record support a judgment as a matter of law that LDHH made appropriate findings and assurances in compliance with the Boren Amendment. LDHH says "yes"; the Hospitals say "no."LDHH has made its findings and submitted assurances to HCFA for approval of its initial Medicaid plan in 1984 and all subsequent amendments to the plan's reimbursement rates. HCFA approved the plan and amendments to "freeze" reimbursement rates until 1988. LDHH contends that this federal agency action entitles Louisiana's plan to a presumption of regularity and warrants application of the arbitrary and capricious standard for reviewing the record to determine whether LDHH complied with federal law. LDHH cites a litany of cases for the indisputable proposition that a state agency's rate-setting action is entitled to considerable deference and is reviewable only under the arbitrary and capricious standard.6 But this rule does not resolve the specific question presented here.

There is a fundamental difference under the Medicaid Act between an agency's discretion to set reimbursement rates and an agency's mandatory compliance with the findings and assurances requirements. It is LDHH's compliance or noncompliance with the findings requirement that is subject to cross motions for partial summary judgment. The findings requirement is both a procedural and a substantive requirement--LDHH must find that the rates are reasonable and adequate and the plan must adopt rates that are actually reasonable and adequate. Wilder, 496 U.S. at 512, 110 S.Ct. at 2519; Illinois Health Care Assoc. v. Bradley, 983 F.2d 1460, 1463 (7th Cir.1993) (distinguishing between the Boren Amendment's procedural component that the agency make findings and assurances and its substantive component that the plan implemented result in adequate payments). The Hospitals complained below of LDHH's noncompliance in practice with the findings requirement and the plan's noncompliance with the substantive findings requirement. 42 C.F.R. Sec. 430.35(C) (1992) ("A question of noncompliance in practice may arise from the State's failure to actually comply with a Federal requirement, regardless of whether the plan itself complies with that requirement.")).

What standard does a federal court use to determine whether LDHH complied with the procedural requirements of federal law, i.e., whether LDHH, in fact, made the "findings" stipulated in the Boren Amendment? Whether LDHH complied with the procedural requirements of the Boren Amendment is a question of law, subject to de novo review. Amisub, 879 F.2d at 795. On this point, most opinions are clear.7 By conducting such a review of the agency's actions, federal courts do not usurp the agency's permissible authority to balance political and financial interests underlying the Medicaid plan.

The opinions are not so clear on the proper standard for reviewing substantive compliance with the findings requirement.8 The opinions generally recite the boiler plate language--de novo review of procedural and substantive compliance with federal law and arbitrary and capricious review of nonadjudicatory agency actions. Some opinions then proceed, we think, to conflate the issues reviewable for substantive compliance with the Boren Amendment with issues reviewable as otherwise nonadjudicatory agency action.9 Both involve a review of the underlying factual foundation and a substantive determination regarding the adequacy of the payment rates. Both implicate a balancing of policy and financial considerations. Both fall within the auspices of the state agency's exercise of discretion. Finally, both are reviewed to some degree by the Secretary prior to approving a plan or amendment. See Illinois Health Care, 983 F.2d at 1465 ("rate-setting and the identification of efficiently and economically operated facilities are all part of the same process") (quoting Folden v. Washington State Dep't of Social & Health Servs., 744 F. Supp. 1507, 1533 (W.D.Wash.1990)).

It is precisely the agency's exercise of discretion and the Secretary's approval that warrant application of the arbitrary and capricious standard of review. Illinois Health Care, 983 F.2d at 1462-63 (reimbursement plan, approved by Secretary and product of federal-state agency action, must be reviewed with the same deference accorded federal agency actions); Pinnacle Nursing Home v. Axelrod, 928 F.2d 1306, 1313 (2d Cir.1991). The reason for the policy of deferential review of a federal agency's interpretation of federal law is its "expertise and familiarity ... with subject matter of its mandate and the need for coherent and uniform construction of federal law nationwide." Turner v. Perales, 869 F.2d 140 (2d Cir.1989). The joint federal-state Medicaid program and the rate-setting flexibility mandated by the Boren Amendment evoke the same policy. This two-step review process--de novo review of the state's factfinding process and arbitrary and capricious review of the findings and rates--provides the "minimum necessary to assure proper accountability." S.Rep. No. 139, 97th Cong., 1st Sess. 478 (1981), reprinted in 1981 U.S.Code Cong. & Admin.News 396, 744.10 It also strikes a balance between Congress's view of the federal role under the Medicaid Act and general principles of federalism, which do not permit states to be final arbiters of their compliance with federal law. Accordingly, a presumption of regularity and deferential standard attaches to LDHH's exercise of discretion in setting reimbursement rates, but only after a reviewing court determines that LDHH made bona fide findings.

The first question then is whether LDHH made findings in compliance with the Boren Amendment procedural requirements. If yes, then and only then will we need to inquire into the substantive adequacy and reasonableness of these reimbursement rates using the arbitrary and capricious standard of review. Wilder, 496 U.S. at 520 n. 18, 110 S. Ct. at 2523 n. 18; Nebraska Health Care Assoc. v. Dunning, 778 F.2d 1291, 1294 (8th Cir.1985), cert. denied, 479 U.S. 1063, 107 S. Ct. 947, 93 L. Ed. 2d 996 (1987). Even this standard and the presumption of validity do not shield LDHH from a "thorough, probing, in-depth review" of the Medicaid plan. Illinois Health Care, 983 F.2d at 1463 (quoting Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 415, 91 S. Ct. 814, 823, 28 L. Ed. 2d 136 (1971) (reviewing under arbitrary and capricious standard the Secretary's approval of highway construction through park)).11

The Hospitals contend that LDHH never engaged in a findings process to substantiate its compliance with the Boren Amendment and its implementing regulations. In their view, LDHH first erred by not adopting an objective profile of what constitutes an economically and efficiently operated hospital and, instead, arbitrarily defined an economic and efficient hospital as one whose costs do not exceed the assigned target rate. Second, LDHH improperly relied on "subjective, generalized and unsupported assumptions about the general state of the economy in Louisiana" in deviating from the state Medicaid plan and imposing target rate "freezes." Third, LDHH violated federal law because it implemented and continued the rate freezes in 1989 and 1990, after HCFA disapproved the TN 88-12 proposed freeze, and because LDHH did not submit assurances to HCFA for the years 1989 and 1990. LDHH responds that it made the appropriate findings and points to the affidavits of Carolyn O. Maggio (director of LDHH Bureau of Health Services Financing) and LDHH employee Helene Robinson (policy and program manager of Louisiana Medical Assistance Program).

Courts generally agree that a state can develop its own methodology for arriving at the required findings. Amisub, 879 F.2d at 797; Illinois Health Care, 983 F.2d at 1464. Regardless of the methodology, an agency still must make the required findings.12 What constitutes a Medicaid Act "finding" has been defined in different ways in the last decades. Some opinions suggest that a finding simply consists of a "reasonably principled analysis." Folden, 744 F. Supp. at 1532. Others, ostensibly in the spirit of the Boren Amendment's flexible rate-setting scheme, pose that a finding is any showing of a "nexus" between reimbursement rates and efficiently and economically operated hospitals. Pinnacle Nursing Home, 928 F.2d at 1314. The most detailed definition of a finding appears in Amisub.

[T]he plain language of federal Medicaid law mandates the State Medicaid Agency, at a minimum, to make "findings" which identify and determine 1) efficiently and economically operated hospitals; 2) the costs that must be incurred by such hospitals; and 3) payment rates which are reasonable and adequate to meet the reasonable costs of the state's efficiently and economically operated hospitals.

Amisub, 879 F.2d at 796 (emphasis in original).

Whether a court chooses to require a "reasonably principled analysis" or a "nexus" or a profile of efficiently and economically operated hospitals is not crucial to determining compliance with the findings requirement. All three of these cases adopt their own terminology to answer the same question. That is, what is the minimum quantum of evidence that an agency must possess in its cognition to substantiate its assurances that the reimbursement rates in the Medicaid plan and any proposed amendments (1) reasonably and adequately meet the costs that must be incurred by efficiently and economically operated hospitals,13 (2) accommodate hospitals serving a disproportionate number of low income patients with special needs, and (3) adequately ensure that Medicaid recipients have reasonable access to inpatient hospital services of adequate quality. 42 C.F.R. Sec. 447.253(b)(1)(i), (ii)(A), (ii)(C) (1992).

The evidence clearly need not consist of the state agency's own comprehensive study of all state hospitals. In Illinois Health Care, the Seventh Circuit aptly observed that a sample of nursing homes as "paradigms of efficiency" may be impossible, waste money better spent on patients, and lead to more controversy. Illinois Health Care, 983 F.2d at 1464-65. On the other hand, the findings requirement is not a mere formality that can be satisfied simply by having a state officer think a bit about hospital costs and then copy out the statutory language on a piece of paper, put the heading "assurances" on that piece of paper, and send it to HCFA. Wilder, 496 U.S. 515, 110 S.Ct. at 2520; Amisub, 879 F.2d at 797; see Pinnacle Nursing Home, 928 F.2d at 1313-14 (procedural requirements are not "mere surplusage" but restrict the state's flexibility in formulating its reimbursement plan). The state agency must show it conducted an objective analysis, evaluation, or some type of factfinding process to determine the effects of the rates on the level of care Medicaid patients receive. Nebraska Health Care, 778 F.2d at 1294. As part of the factfinding process, the state agency must "judge the reasonableness of its rates against the objective benchmark of an 'efficiently and economically operated facility' providing care in compliance with federal and state standards while at the same time ensuring 'reasonable access' to eligible participants." Wilder, 496 U.S. at 519, 110 S.Ct. at 2522. This objective benchmark can be a "relative rather than an absolute concept." Illinois Health Care, 983 F.2d at 1467. It also can be implicit in a rate-setting methodology. Id.

LDHH admits, with some hesitation, that it conducted no studies and made no efforts to determine which state hospitals are efficiently and economically run. Instead, LDHH functioned under the premise that every hospital was economically and efficiently operated in 1981 and used the available cost reports for that base year to calculate the target rates for each hospital. Whether a particular hospital remains efficient and economical is gauged by whether the hospital stays within the designated target rate. Hospitals that exceed the designated target rate are deemed not efficient or economical and are not reimbursed actual costs. Rather, the hospitals receive the maximum payment determined by their respective target rates.14

Clearly, the method by which LDHH promulgated the initial target rates fails the Boren Amendment test. LDHH did not make any finding that its plan complied with the three substantive requirements outlined in 42 C.F.R. Sec. 447.253(b). Other than the cost reports, LDHH gathered no information and conducted no empirical analysis to ascertain whether the target rates "reasonably and adequately" compensated efficient and economical hospitals and hospitals servicing a disproportionate number of low income patients. "Federal law is not satisfied if a state merely makes conceptual policy decisions. A policy predicated upon provincialism and self-interest, not upon findings of reasonableness and adequacy, is unacceptable." West Virginia Univ. Hosps., Inc. v. Casey, 885 F.2d 11, 30 (3rd Cir.1989).

LDHH insists that it engaged in "ongoing analyses of adequacies of the rates."15 According to Robinson, in deciding to freeze the target rates, she "looked at how costs had increased from the base period to the current audited period versus how the average target rate had increased in that same period of time. And that the increases in the target rate were greater than the actual increases in costs." The "costs" compared were the hospitals' total cost per discharge, without any separation of labor costs or other cost components. Robinson retrieved this cost data from audited cost reports available for the latest time period. There were hospitals whose data was not considered. She did not employ a random statistical sampling or scientific methodology to determine that the hospitals reviewed were representative. Rather, Robinson admits in her deposition that she didn't do an actual analysis but based the "process" on what she thought was right based on her experience. Robinson further says that no analysis was done to determine if the uncounted hospitals skewed the results until 1987 or 1988.

Robinson says she reviewed other data available from the "Department of Labor and Statistics" and other publications reflecting general economic conditions. Robinson recalls in deposition she "had difficulty obtaining data on Louisiana-specific" information. She did manage to review "like weekly things in the newspaper.... things in magazines and so on regarding, you know, unemployment, drops in personal income, and so on for the State." However, this analyses could not be reproduced because it was an "ongoing process" not reduced "necessarily [to] a written bound copy of something."

Furthermore, Robinson neither recommended to her staff nor initiated any efforts to obtain information regarding the economic conditions in the State as they affect hospitals, hospital labor costs, or any other costs that would be reflected in the hospital cost reports. Nor did Robinson recall having access to or using any studies or data base concerning such economic conditions and cost of care, or medical care, throughout the State. Finally, Robinson did not investigate the effects of case mix changes on Louisiana providers, such as increased outpatient services and longer hospital stays for inpatient services. What she says is this:

Pizza: [W]as any study done by DHH during the period 1982 through 1989 to determine whether or not those factors influenced inflation for hospitals in Louisiana? [Those factors enumerated were the "increase in the hiring of RNs as opposed to LPNs, the increasing complexity of patient treatments, the trend of treating more outpatients [,] ... that wage rates for hospital employees had climbed higher than for other kinds of employees, and finally that insurance costs had climbed at a higher rate for hospitals than it had for other industries nationwide."]

Robinson: Studies were done in regard to hospital cost increases that would have reflected some of those factors.

Pizza: I know, but was a study done, though, looking at those factors themselves and their effect on hospitals in Louisiana, germane just to those factors?

Robinson: We look at the factors in the aggregate.

Pizza: So your answer then is no particular study was done to look at those individual factors; is that right?

Robinson: As I stated, we look at the factors and the increases in costs in the aggregate.

Pizza: Okay. Now when you say you looked at them in the aggregate, what did you do to look at them in the aggregate?

Robinson: We review cost data--

Pizza: Cost reports?

Robinson: --from the cost reports.

Pizza: So you examine cost reports to see how that data affected hospitals; is that right?

Robinson: We examine cost data to review--to determine the adequacy of the rates.

LDHH's purported "ongoing analysis" suffers from the same faulty logic as its initial rate-setting scheme. Findings were not made, instead assumptions were made. It is circular for LDHH to set target rates under the assumption that all hospitals are efficiently and economically operated and then identify efficiently and economically operated facilities as those whose costs fall below their own reimbursement rate. Under these circumstances, a hospital's ability to keep costs below the target rate is not a reflection of its efficiency or economy of operations. Even Robinson admits this fallacy in LDHH's rate-setting methodology. She testified at her deposition that "if [the hospitals] were within their target rate, they were deemed efficient and economic, even though they might have had some continued inefficiencies." She also acknowledges in her affidavit that each hospital's target rate initially included all allowable base year costs, without "limit[ ] or cap[ ] in any respect to remove inefficient, uneconomic or unnecessary costs." By enacting the Boren Amendment, Congress intended states to abandon such "reasonable cost" schemes that paid actual hospital costs, despite obvious disparities in efficiencies and economies, in favor of reimbursement systems that encourage hospital efficiency and cost containment. H.R.Rep. No. 158, 97th Cong., 1st Sess. 293 (1981)16; Temple University v. White, 941 F.2d 201, 207 (3rd Cir.1991); West Virginia Univ. Hosp., 885 F.2d at 23.

LDHH has reduced its findings process to a simple exercise of compilation and assumption, completely ignoring the Congressional mandate that state agencies consider relevant factors such as efficiency, economy, quality of care, and reasonable access. LDHH emphasizes that it has received relatively few complaints regarding hospital quality of care. This fact, however, says nothing about the reasonableness or adequacy of rates to meet the needs of efficiently and economically operated hospitals where rates initially set assume that every hospital is efficient and economic. At most, LDHH's "findings" process consisted of reviewing general information regarding the state of the economy and the available cost reports. LDHH can point to nothing in the federal Medicaid scheme that permits it to use the general state of the economy as the sole justification for setting rates.17 Nor is there anything that condones a findings procedure that employs only historical cost data and projected inflation figures. To the contrary, courts have condemned similar schemes.

In Amisub, the state Medicaid agency presented as evidence of its "findings" process only a "consistency between the current expenditure for Medicaid provider reimbursement and the amount of money historically appropriated by the Colorado legislature, and HCFA's acceptance of the previous Colorado Medicaid Plan." Amisub, 879 F.2d at 796. The Tenth Circuit rejected this evidence as proof of "findings" for three reasons. First, the state agency's reliance on HCFA's approval of a previous plan belies the statutory requirement that agency's make at least annual findings. Id. at 797. Second, there is nothing in the Medicaid Act from which to infer that an agency may rely solely on the "historical trends concept" that the new plan is adequate because the monetary appropriations are identical to the old plan. Id. at 797, 799. Finally, a "record is blatantly devoid of any effort ... to make the federally mandated findings" where the assurances are based solely on budgetary constraints. Id. at 800. The Tenth Circuit ultimately remanded the case and ordered the state agency "to comply with the procedural and substantive requirements of the federal Medicaid Act and its implementing regulations, and to engage in a bona fide findings process before submitting any new plan and/or assurances to HCFA." Id. at 801.18 We too remand this case and order LDHH to engage in a bona fide findings process.

LDHH likens its case to that in Mississippi Hosp. Assoc., Inc. v. Heckler, 701 F.2d 511 (5th Cir.1983), where we affirmed summary judgment declaring that Mississippi's Medicaid agency complied with the federal procedural requirements. Mississippi Hosp., however, is inapposite for two reasons. First, in Mississippi Hosp., we did not review whether the state agency engaged in a bona fide findings process. Rather, Mississippi's Medicaid agency was accused of not complying with federal procedural requirements based on its failure to consult with the Medical Care Advisory Committee, its publication of an allegedly inadequate public notice, and its failure to submit proper assurances. Id. at 520. It is evident from our discussion that Mississippi engaged in a bona fide findings process and that we confined our review to the sufficiency of these findings under the arbitrary and capricious standard. Id. at 516-18. We have not reached this second stage of analysis in this case.

Second, we find no similarities between the findings processes adopted by LDHH and those adopted by Mississippi's agency. Mississippi based its rate ceiling on two-years and over 300 hours of "careful and objective studies of cost data filed by Mississippi hospitals ...[,] methods used by other states and the federal government....[,] an incredible rage of cost incurred by hospitals," and statewide occupancy rates. Id. at 517, 520. Here, Robinson's review of cost reports was limited to comparing the available aggregate costs to the hospital's target rate to determine reimbursement.

This Court cannot endorse such a pro forma compliance with the findings requirement of the Boren Amendment. LDHH's inability to articulate an orderly process of evaluation or to identify specific documents reviewed renders suspect its "findings" and renders meritless its argument that it engaged in a bona fide findings process. Because we find that LDHH, as a matter of law, failed to comply in practice with the Boren Amendment findings requirement, this Court REVERSES the district court's grant of LDHH's summary judgment motion and denial of the Hospitals' summary judgment motion. This case is REMANDED for proceedings not inconsistent with this opinion.19

Nov. 26, 1993.

IT IS ORDERED that the petition for rehearing filed in the above entitled and numbered cause be and the same is hereby DENIED. However, the panel opinion reported at 3 F.3d 797 is amended, by the addition of footnote 19 to the last paragraph of the opinion, as follows: