The Secretary of Health and Human Services appeals a summary judgment invalidating the 1979 medicare regulation known as the Malpractice Rule, now 42 C.F.R. § 405.452(a)(l)(ii) (1984). The district court found the rule arbitrary and capricious, contrary to 5 U.S.C. § 706(2)(A). We modify the judgment and, as modified, affirm.
Facts and Procedural Background
The Malpractice Rule was promulgated by the Secretary in performance of her statutory duty to reimburse hospitals for the reasonable costs of services furnished to medicare beneficiaries.1 Health care providers formerly were reimbursed on a straight utilization basis, with indirect costs such as malpractice insurance premiums collated in a general and administrative (G & A) cost account. By way of example, if Medicare patients occupied 40% of a hospital’s bed-days in a given year, the Medicare program reimbursed the hospital for 40% of its G & A costs, including malpractice insurance premiums.2
In 1976 the Secretary authorized a study which had as its goal the reduction of incidents of medical malpractice. Pursuant to a commission Westat, Inc., a private contractor, prepared the 1976 Medical Malpractice Closed Claim Study Report (“Westat Study”). The report reached the general conclusion that malpractice awards decrease with age and medicare patients tend to receive smaller awards than the general population. The Secretary received this report in May of 1978 at a time when there was obvious concern about the skyrocketing increase in malpractice insurance premiums.
Apparently convinced that the federal government was paying a disproportionate share of hospital malpractice insurance costs the Secretary prepared a regulation tying reimbursement to the hospital’s actual malpractice claims experience.3 The pro*184posed Rule was announced in the Federal Register on March 15, 1979. 44 Fed.Reg. 15744. The comment period extended to April 30 and the Rule became effective July 1, 1979. 44 Fed.Reg. 31641.
Appellees, five Louisiana hospitals, filed timely appeal requests with the Provider Reimbursement Review Board (PRRB), the administrative body established to resolve medicare reimbursement disputes. Because the PRRB is bound by the Secretary’s regulations, 42 C.F.R. § 405.1867, appellees asked the PRRB to note its lack of authority, thus clearing the way for immediate judicial review. 42 U.S.C. § 1395oo (f)(1). The PRRB granted appel-lees’ request for expedited judicial review. The instant suit, covering cost reporting years 1980 and 1981, challenged the Malpractice Rule as: (1) arbitrary and capricious, (2) promulgated in violation of the notice and comment provisions of the Administrative Procedures Act, 5 U.S.C. § 553(b), and (3) contrary to the Social Security Act because it did not compensate hospitals for the reasonable costs of providing health care services to medicare patients. The district court considered only the first challenge, finding the Secretary’s action arbitrary and capricious. We do likewise.
Analysis
We join many colleagues who have addressed the validity of the Malpractice Rule. Four circuits have concluded that the Rule is invalid. Abington Memorial Hosp. v. Heckler, 750 F.2d 242 (3d Cir.1984); Humana of Aurora, Inc. v. Heckler, 753 F.2d 1579 (10th Cir.1985); St. James Hosp. v. Heckler, 760 F.2d 1460 (7th Cir.1985);4 and Lloyd Noland Hosp. & Clinic v. Heckler, 762 F.2d 1561, (11th Cir.1985). The District of Columbia Circuit, clearly signaling its view, remanded a district court decision upholding the Rule. Walter O. Boswell Memorial Hosp. v. Heckler, 749 F.2d 788 (D.C.Cir.1984). Three district courts have upheld the Rule, Athens Community Hosp. v. Heckler, 565 F.Supp. 695 (E.D.Tenn.1983); Cumberland Medical Center v. Heckler, 578 F.Supp. 39 (M.D.Tenn.1983); and Normandy Osteopathic-North Hospital v. Heckler, No. 83-1687-C(1) (E.D.Mo. Dec. 21, 1984). Fifteen district courts have found the Rule invalid.5 Mt. Carmel Mercy Hosp. v. Heckler, 581 F.Supp. 1311 (E.D.Mich.1984); Bedford County Memorial Hosp. v. Heckler, 583 F.Supp. 367 (W.D.Va.1984); Chelsea Community Hosp. v. Heckler, No. 83-CV-6126-AA (E.D.Mich. Dec. 20, 1983); Albany General Hosp. v. Heckler, 584 F.Supp. 614 (D.Ore.1984); St. Joseph’s Hosp. v. Heckler, 583 F.Supp. 1545 (D.Ariz.1984); Alexandria Hosp. v. Heckler, 586 F.Supp. 581 (E.D.Va.1984); Metropolitan Hosp. v. Heckler, No. C-83-502A (N.D.Ga. June 25, 1984); Menorah Medical Center v. Heckler, No. C-83-0822 (W.D.Mo. July 26, 1984); Mercy Medical Center v. Heckler, No. 3-82-CIV-1724 (D.Minn. Aug. 17, 1984); Parkway Medical Center v. Heckler, 614 F.Supp. 564 (S.D.Fla.1984); St. Anthony Regional Hosp. v. Heckler, 613 F.Supp. 23 (N.D.Iowa 1984); East Jefferson General Hosp. v. Heckler, No. 83-4107 (E.D.La. Oct. 18, 1984); Arkansas Methodist Hosp. v. Heckler, 597 F.Supp. 238 (E.D.Ark.1984); Sisters of St. Mary v. Heckler, No. 83-0789-C(4) (E.D.Mo. Nov. 6, 1984); Charter Medical Corp. v. Heckler, 604 F.Supp. 638 (M.D.Ga.1985).
Standard of Review
The APA directs that a court shall “hold unlawful and set aside agency action, findings, and conclusions found to be — (A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). A reviewing court must determine “whether the decision was based on a consideration of the *185relevant factors and whether there has been a clear error of judgment.” Motor Vehicle Mfrs. Assn. v. State Farm Mut., 463 U.S. 29, 103 S.Ct. 2856, 2867, 77 L.Ed.2d 443, 458 (1983). The scope of review is narrow and we may not substitute our judgment for that of the agency. Notwithstanding, a rule or regulation may be deemed arbitrary and capricious where the agency “offer[s] an explanation for its decision that runs counter to the evidence before the agency____” Id.
Appellees maintain that the Secretary erred by relying on inadequate emperical information. Appellees contend that the explanation given for the Secretary’s decision runs counter to the evidence in the administrative record, including the Westat Study. This argument posits two questions for our resolution: (1) is the Westat Study flawed, and (2) did the Secretary rely entirely or primarily on that report in the promulgation of the Malpractice Rule?
The Westat Study6
As evidence of the proposition that the government pays a disproportionate share of the cost of hospital malpractice insurance, the Westat Study is woefully inadequate. This is so, in no small measure, because the study was commissioned to analyze the causes of malpractice-related injuries and was designed to gather information that could be used to develop injury-prevention programs. The study was not intended to furnish a critical analysis of the medical malpractice insurance premium calculus. We perforce must note the cautionary statement of the authors: “It seems proper, since this is a census, to make inferences only within the time frame for which the data were collected and only to the universe of claims closed by the nine participating companies.” The survey canvassed nine carriers during a four-month period in 1976. The authors’ caveat undercuts the Secretary’s decision to use the findings of the Westat Study to establish the “national average” ratio of 5.1% which applied to four of the five hospitals in the case sub judice.
We are convinced that the most significant flaw in the Westat Study is its failure to distinguish between claims against doctors and those against hospitals. That view is shared. Boswell Memorial Hospital; St. James Hospital. Thirty-two percent of the claims examined by Westat were against hospitals, 63% were against physicians. This imbalance skewed the data and manifestly impaired its reliability. The section of the report comparing the malpractice experience of medicare patients against other patients used as its critical data base claims for malpractice occurring in hospitals, not just claims for malpractice against hospitals. This unrefined data is of little relevance and grossly insufficient to ground a nationwide rule.
We conclude that the Westat Study is not a legally sufficient basis for the proposition that the federal government is paying a disproportionate share of hospital malpractice insurance premiums.7 This leads to the inquiry whether there is any other evidence in the administrative record supportive of the Rule.
The Administrative Record
The Secretary submitted the entire administrative record upon which she based the Malpractice Rule. The record contains 11 volumes. The Westat Study and related correspondence comprise three volumes, miscellaneous agency memoranda two volumes, and public comments 6 volumes. After reviewing this record, we are convinced that the Secretary relied primarily on the Westat Study in promulgating the Malpractice Rule. This is evident from the basis and'purpose statement accompanying the Rule.
This [utilization] method results in Medicare paying a disproportionate amount of malpractice costs. A national study conducted by a DHEW consultant indicates that malpractice awards for Medicare patients are significantly lower in amount than losses for other patient population. Lower awards for Medicare patients result because the income potential and life expectancy of these patients are less *186than the non-Medicare population. Thus, the use of overall Medicare utilization to allocate malpractice costs results in Medicare’s paying a disproportionate amount of malpractice costs.
44 Fed.Reg. 31641. We find arbitrary and capricious the adoption of the Malpractice Rule based on “a single study criticized extensively by its authors in its particular application....” Boswell Memorial Hosp., 749 F.2d at 803. We join the Third, Seventh, Tenth and Eleventh Circuits and declare the Malpractice Rule invalid.
We affirm the district court’s ruling that the Malpractice Rule is invalid. We modify the judgment to the extent that we remand to the PRRB for payment of the appellees’ hospital insurance costs pursuant to 42 C.F.R. § 405.452(b)(1) (1979), without reference to the abrogated malpractice rule.
MODIFIED and AFFIRMED.
. Sections 1814(b) and 1833(a)(2) of the Social Security Act require reimbursement to health care providers of the lesser of the reasonable cost of services furnished to beneficiaries or the customary charges made by the provider for the same services. Section 1861(v)(l) of the Act defines reasonable cost and authorizes the Secretary to issue regulations establishing the method of determining reasonable cost. See Pub.L. No. 89-97, 79 Stat. 291 (1965), codified as amended at 42 U.S.C. § 1395 et seq.
. Other costs frequently included in the typical G & A account include: data processing cost, business office overhead, purchasing, credit and collection costs, and pastoral care.
. The regulation isolated malpractice insurance costs and removed them from the G & A pool. After July 1, 1979, reimbursement was 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.” 42 C.F.R. § 405.452(a)(l)(ii). If a hospital had no malpractice loss experience for the preceding five-year period, reimbursement was based on a national ratio of malpractice awards paid to *184medicare beneficiaries to malpractice awards paid to all patients. Id. Four of the five hospitals involved in this litigation were in this latter category, and the national ratio in effect for the years at issue was 5.1%, based on the Westat Study. The medicare utilization rates of these hospitals for the years in question ranged from 36.7% to 61.7%.
. The Seventh Circuit consolidated the appeals of St. James Hosp. v. Heckler, 579 F.Supp. 757 (N.D.Ill.1984), and Humana of Illinois, Inc. v. Heckler, 584 F.Supp. 618 (C.D.Ill.1984).
. The judge a quo adopted the reasoning of his district court colleagues in Lloyd Noland, Bed-ford County, St. James, Abington, and Mt. Car-mel. In each case the court found that the only evidence supporting the Secretary’s rule was the Westat Study. All found this reliance ill-placed; the study was deemed flawed and basing a rule thereon was considered arbitrary and capricious.
. The Westat Study’s deficiencies are brought into sharp focus in Walter O. Boswell Memorial Hosp. v. Heckler, 749 F.2d at 795-97. We need not repeat the litany here.
. We do not wish to discourage the Secretary’s efforts to avoid inappropriate payments to health care providers in any given area. We are convinced that the Secretary simply needs better data and a more workable alternative. As this opinion is prepared for publication, we are informed of the Secretary’s latest efforts, a proposed rule to amend the regulation at issue in this case. 50 Fed.Reg. 116.