Appeal from a judgment of the Supreme Court (Klein, J.), entered June 17, 1988 in Albany County, which, inter alia, partially granted petitioner’s application, in a proceeding pursuant to CPLR article 78, to annul determinations of respondent Department of Health establishing petitioner’s 1983 Medicaid and Medicare reimbursement rates.
Petitioner, a not-for-profit hospital in the Village of Cooperstown, Otsego County, is reimbursed for services rendered to eligible Medicaid and Medicare patients at rates determined by respondent Commissioner of Health pursuant to Public Health Law § 2807. Reimbursement rates are computed for a given period on the basis of actual costs experienced by a facility in a previous "base” year. However, the extent to which such costs are utilized for rate setting is limited by the imposition of "ceilings” on various categories of costs (see, 10 NYCRR 86-1.14) and a facility’s costs in excess of these ceilings are "disallowed” in setting its reimbursement rates. Among these costs are "ancillary costs” which relate to specific medical procedures such as operations, X rays and laboratory services. Petitioner’s ancillary cost ceiling was established for 1983 by developing a "peer group” of hospitals having *827similar cost-related characteristics, through a grouping methodology known as "seed clustering” (see, 10 NYCRR 86-1.13 [b]), and determining the group’s base year average ancillary cost per patient discharge.
For 1983, 1981 was utilized as the base year and petitioner was grouped with four other upstate teaching hospitals: Ellis Hospital in Schenectady, St. Joseph’s Hospital in Syracuse, and St. Mary’s and Genesee Hospitals in Rochester. A methodology known as "the case mix formula” was also used to measure and adjust for the cost-related differences in patient populations between peer group hospitals, recognizing that even basically similar hospitals are not identical in the kind of patients they care for and that such differences may have a significant impact on comparative costs. Here, application of the case mix formula resulted in an increase of petitioner’s ancillary cost ceiling above the peer group average by 10.7% for Medicaid and 12.4% for Medicare. Nevertheless, petitioner’s ancillary costs were still substantially "over ceiling”, resulting in disallowance of a significant portion of the actual cost of ancillary services to Medicare and Medicaid patients.
Petitioner contested the 1983 ancillary services reimbursement rates and, after unsuccessful administrative appeal, commenced this CPLR article 78 proceeding, contending that respondent Department of Health acted arbitrarily and illegally by classifying it in a peer group with hospitals whose size, cost elements and geographic areas were vastly dissimilar. Supreme Court granted the petition, annulled respondents’ 1983 rate determinations and directed respondents to recompute the rates, with petitioner in a peer group of comparable hospitals. Respondents now appeal.
At the outset, respondents contend that petitioner did not raise the issue of noncomparability due to its smaller size in its rate appeal and, thus, this court should not consider the issue. We disagree. The record demonstrates that petitioner’s claim that it was not grouped with comparable facilities was raised at the administrative level and in the pleadings and is properly before us for review (cf., Matter of Estate of Manno v State of New York Tax Commn., 147 AD2d 805, Iv denied 74 NY2d 610).
Turning to the merits, the Public Health Law requires the Commissioner of Health to determine that reimbursement rates are "reasonable and adequate to meet the costs which must be incurred by efficiently and economically operated facilities” (Public Health Law § 2807 [3]). 10 NYCRR 86-1.13 (a) requires that in developing a peer group, the Commissioner *828take into account, inter alia, the general criteria of case mix, service mix, patient mix, size of facility, teaching activity and geographic location. Petitioner does not challenge the methodology utilized by respondents in establishing its ancillary cost ceilings but, rather, claims that the application of the methodology here resulted in its improper and illegal grouping with vastly dissimilar hospitals. Thus, the argument continues, respondents’ actions were not in accord with the requirements of Public Health Law § 2807 (3) and 10 NYCRR 86-1.13.
Central to petitioner’s argument is the claim that it, a teaching hospital with 180 beds in a village of approximately 2,500 residents, is unique and readily distinguishable from the other members of its peer group, who have from 304 to 478 beds and are in cities with populations ranging from approximately 65,000 to 240,000. Petitioner asserts that "Euclidian distance measures between [it] and [its] reference group are extremely large, indicating a rather dispersed group”. Thus, petitioner claims, the grouping data under respondents’ seed clustering methodology supports its claim of noncomparability.
In our view, petitioner has sufficiently established that respondents failed to fully comply with statutory and regulatory mandates in establishing petitioner’s peer group for purposes of setting ancillary cost ceilings. The statutory mandate provides that, in determining whether reimbursement rates are "reasonable and adequate to meet the costs which must be incurred by efficiently and economically operated facilities” (Public Health Law § 2807 [3]), the Commissioner shall take into consideration, inter alia, "costs of hospitals of comparable size” (Public Health Law § 2807 [3] [emphasis supplied]). To this end, in establishing a peer group, the Commissioner is required to consider the size of the facility and its geographic location (see, 10 NYCRR 86-1.13 [a] [4], [6]). "While it is true that great deference is to be given to the Commissioner’s interpretation of a regulation * * * it is also axiomatic that an agency is bound by the language of its own regulation and cannot construe it in such a manner that the plain language on the face of the regulation is rendered meaningless” (Matter of Grace Plaza v Axelrod, 121 AD2d 799, 801). Accordingly, the placing of petitioner, a small rural hospital, in a peer group of much larger metropolitan hospitals for the purpose of "seed clustering” manifests a patent disregard for the required criteria of size and location.
Moreover, petitioner has established that it was disadvantaged by the Commissioner’s peer grouping. Its smaller size *829and lower patient volume have resulted in higher per unit costs than its peer group due to the relatively high "fixed” costs of the nonphysician support personnel needed for ancillary services. Additionally, and of persuasive significance, is the affidavit of a health systems management engineer who found that petitioner "sustains extra costs in equipment servicing contracts” and "in obtaining and transporting costs relative to, e.g., plasma and drugs”. The expert further noted that petitioner "cannot, as may the other members of its [peer] group, arrange for the sharing of costly services with neighboring institutions”. We find these to be some of the very considerations contemplated in the regulations that require the Commissioner to take into account size and geography in peer grouping. The Commissioner has failed to do so and, in this instance, we should not countenance such a deviation from the statutory and regulatory scheme.
Judgment affirmed, without costs. Kane, J. P., Yesawich, Jr., and Levine, JJ., concur.
Casey and Mercure, JJ., dissent and vote to modify in a memorandum by Mercure, J. Mercure, J. (dissenting). In our view, there should be a modification. We disagree with the majority’s conclusion that respondents violated applicable statutory and regulatory provisions in establishing petitioner’s peer group. Although Public Health Law § 2807 (3) does require respondent Commissioner of Health, in certifying over-all rates, to take into consideration a number of factors, the costs of hospitals of comparable size among them, it does not necessarily follow that each factor must be directly applied in every aspect of the rate-making process. The statutory mandate is satisfied if any component of the rate-making process makes appropriate adjustment for differences in size and geographic location. Significantly, 10 NYCRR 86-1.13 (a), which, unlike Public Health Law § 2807 (3), applies specifically to the formation of peer groups, requires only that consideration be given to the "general criteria” of size and geographic location, with no stated requirement of comparability.
Further, we conclude that petitioner has failed to satisfy its burden of establishing that the rate calculations were arbitrary and capricious (see, Matter of Severino v Ingraham, 44 NY2d 763; Tioga Nursing Home v Axelrod, 90 AD2d 570, affd 60 NY2d 717). Although petitioner asserts that its smaller size disadvantaged it because of its relatively high fixed costs, the record discloses that the most truly "fixed” of the costs associated with petitioner’s ancillary services are its equipment and installation costs, which are deemed "capital costs” for reim*830bursement purposes and were not held to the ancillary cost ceilings in 1983. Moreover, the fixed costs for salary and fringe benefit costs of physicians, including interns and residents assigned to provide ancillary services, were held to separate ceilings and the cost of supplies was directly related to the volume of use.
Nor does the record support petitioner’s contention that its higher costs resulted from specific aspects of its operation that were noncomparable with the other facilities in its peer group. We are not persuaded by petitioner’s assertion that its patients were "sicker” than those of other peer group hospitals. Notably, petitioner’s per-patient routine costs and average patient length of stay were within the average for its peer group and, in any event, the case mix methodology adjusted for cost-related differences caused by more seriously ill patients. Petitioner’s higher ratio of interns and residents and a fully salaried medical staff did not render the rate-fixing methodology arbitrary or capricious. First, as noted, the salary and fringe benefits of interns and residents were not covered by ancillary ceilings. Second, a higher ratio of interns and residents does not justify the provision of ancillary services to patients where there is no legitimate medical need. Further, there is no support in the record for petitioner’s contention that its rural location inflates its ancillary costs in comparison to the urban hospitals in its peer group. Notably, the average Euclidian distance (the mathematical measure of how close a hospital is to its peer group members, based on the total impact of all the grouping variables) of six other upstate teaching hospitals was greater than petitioner’s. Finally, although petitioner functions as a rural referral center, all of the facilities in petitioner’s peer group are teaching hospitals which commonly act as referral centers for patients requiring more sophisticated treatment.
Peer grouping has been accepted as a method of determining effective cost control (see, Matter of Jewish Mem. Hosp. v Whalen, 47 NY2d 331, 341) and, in our view, petitioner has failed to establish that it was improperly grouped. Thus, the judgment of Supreme Court should be modified accordingly and the petition dismissed in its entirety.