The issue presented in this appeal is whether the district court erred by ordering the Arkansas Department of Human Services (ADHS) and the Arkansas Foundation for Medical Care (AFMC) to publish to Medicaid service beneficiaries and providers the identity of physician reviewers who make decisions to approve or deny medical care to Medicaid-eligible children. Because the district court's order is at odds with the relevant Medicaid regulations, we reverse.

The general background to this case is detailed in our prior decisions in Pediatric Specialty Care, Inc., v. Arkansas Department of Human Services, 293 F.3d 472 (8th Cir.2002); Pediatric Specialty Care, Inc. v. Arkansas Department of Human Services, 364 F.3d 925 (8th Cir.2004); and Pediatric Specialty Care, Inc. v. Arkansas Department of Human Services, 443 F.3d 1005 (8th Cir.2006), issued today. In short, this case stems from ADHS's attempts to curtail the services it provides to needy children in Arkansas pursuant to the Medicaid Act. ADHS contracts with AFMC to review applications for Medicaid services and to determine whether and to what extent applicants qualify for those services. The plaintiffs are applicants for Medicaid services, both as recipients and providers. They contend that ADHS and AFMC manipulate their system of prior authorization so as to deny medically necessary services to qualified, Medicaid-eligible children solely for the state's financial benefit.

As relevant to this appeal, one of the plaintiffs in this matter, Tomorrow's Child Learning Center, LLC, moved the district court for a temporary restraining order requiring AFMC (and, consequently, ADHS) to disclose to Medicaid service recipients and providers the names of the physician reviewers who made determinations regarding whether and to what extent children were eligible for treatment. By order dated May 28, 2004, the court treated this motion as one for a preliminary injunction and denied the motion, holding that such disclosure was prohibited by applicable Medicaid Act provisions and regulations. Several months later, the district court informed the parties that it was reconsidering its earlier decision and solicited letter briefs on the matter. On November 29, 2004, the court issued an order reversing its May 28, 2004 decision. In the decision, the court held that the plaintiffs "may discover the name, qualifications, and identity of each peer review physician and advisor who participated in the denial of medical benefits, based on the premise that the care did not meet medical necessity requirements," but that the plaintiffs "may not discover any patient records of non-parties, nor may Plaintiffs discover any material produced by [AFMC] that evaluated the failure of a particular medical provider to follow professional standards of care." (Dist. Ct. Order of Nov. 29, 2004 at 15.) This appeal followed.

"We review a district court's grant or denial of preliminary injunctive relief for an abuse of discretion or misplaced reliance on an erroneous legal principle." Goff v. Harper, 60 F.3d 518, 520 (8th Cir. 1995). This is a "deferential standard," only permitting us to reverse the district court if, "after a thorough review of the record, the proof unmistakably establishes clear error or an abuse of discretion." FTC v. Freeman Hosp., 69 F.3d 260, 267 (8th Cir.1995).

Ordinarily, the district court is to consider four factors in deciding whether to issue the preliminary injunction: (1) the threat of irreparable harm, (2) the balance between that harm and the injury to the non-moving party if the injunction is granted, (3) the probability of success on the merits of the claim, and (4) whether granting the injunction would benefit the public interest. Dataphase Sys., Inc. v. C L Sys., Inc., 640 F.2d 109, 114 (8th Cir. 1981) (en banc).

AFMC and ADHS assert that the district court's order must be reversed because it conflicts with provisions of the Medicaid Act, as well as regulations interpreting the Act that protect the confidentiality of peer reviewers. The plaintiffs argue that the district court's order is not a final appealable matter, because the district court did not explicitly go through a Dataphase analysis. We disagree, and thus undertake analysis of whether the order was properly issued. Accord Baker Elec. Co-Op, Inc. v. Chaske, 28 F.3d 1466, 1472 (8th Cir.1994) (characterizing a temporary restraining order as a preliminary injunction and undertaking a Dataphase analysis even though the district court failed to do so).

Pursuant to 42 U.S.C. § 1320c-9(a), data and information collected by a peer review organization such as AFMC "shall be held in confidence and shall not be disclosed," except in very limited circumstances. "No patient record in the possession of [a peer review organization] shall be subject to subpoena or discovery proceedings in a civil action." § 1320c-9(d). The statute states more specifically that information reviewed by a peer review organization in order to determine if the provider's quality of care meets professional standards is privileged from discovery. Id. (referencing §§ 1320c-3(a)(1)(B) and 1320c-5(a)(2)).

The Department of Health & Human Services (HHS) is charged by statute with the authority to promulgate regulations regarding the confidentiality of patient records held by peer review organizations. Those regulations must "assure adequate protection of the rights and interests of patients, health care practitioners, or providers of health care." 42 U.S.C. § 1320c-9(a)(2). HHS has issued such regulations, which interpret what information is subject to disclosure and what information is confidential. The regulations deem information confidential if it "explicitly or implicitly identifies an individual patient, practitioner, or reviewer," and all peer review deliberations. 42 C.F.R. § 480.101(b) (emphasis added). The regulation defines peer review deliberations to include "review notes, minutes of meetings and any other records of discussions and judgments involving matters regarding [peer review] responsibilities and appeals from [peer review organization] determinations, in which the opinions of, or judgment about, a particular individual or institution can be discerned." § 480.101(b)(4). Peer review organizations generally may not disclose confidential information, subject to limited exceptions. See generally 42 C.F.R. §§ 480.130-480.143 (regulations regarding disclosure of confidential information).

Within the regulations concerning disclosure of confidential information, it is stated that a peer review organization "must not disclose its deliberations," except to federal governmental agencies such as the Centers for Medicare and Medicaid Services, the Office of the Inspector General, or the General Accounting Office. 42 C.F.R. § 480.139(a)(1). The regulations further state that a peer review organization may only disclose "information on a particular practitioner or reviewer at the written request of or with the written consent of that practitioner or reviewer." 42 C.F.R. § 480.133(a)(2)(iii).

Taken together, these regulations do not permit the disclosure ordered by the district court. They indicate clearly that: (1) nearly all peer review organization information is confidential; (2) confidential information related to deliberations must not be disclosed, except in circumstances not presented here; and (3) physician reviewer identities may be disclosed, but only with the reviewer's written consent. Thus, it seems clear under the regulations that the information plaintiffs seek-the identities of physician reviewers in specific cases-is not subject to discovery.

The plaintiffs have not argued, nor do we see any reason why, these regulations are not entitled to deference. See Chevron U.S.A. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). The district court cited Public Citizen, Inc. v. United States Department of Health & Human Services, 332 F.3d 654 (D.C.Cir.2003), in support of the view that we need not follow the agency's confidentiality regulations. There, the court refused to grant Chevron deference to the confidentiality regulations for a section of the Medicaid Act that was enacted after the regulations were promulgated. Pub. Citizen, 332 F.3d at 659.

We agree with the proposition that where regulations precede enactment of the statute, they cannot possibly be read as interpreting the statute. That is not the situation before us, however. Since § 42 U.S.C. § 1320c-9's most recent amendment, HHS has reconsidered and amended the confidentiality provisions of its regulations, such that information could now be disclosed with the written consent of a medical practitioner or peer reviewer. See 69 Fed.Reg. 48916, 49266-67 (Aug. 11, 2004) (amending 42 C.F.R. §§ 480.133, 480.140). Thus, unlike Public Citizen, the confidentiality regulations at issue here have been significantly and materially changed subsequent to the statute's amendment.

According the regulations the deference required by Chevron, we find the interpretation to be "based on a permissible construction of the statute." Chevron, 467 U.S. at 843, 104 S.Ct. 2778. The statute governing disclosure specifically charges HHS with the duty to ensure peer review information is kept confidential through regulations aimed toward protecting "the rights and interests of patients, health care practitioners, [and] providers of health care." 42 U.S.C. § 1320c-9(a)(2). The regulations effectuate that purpose, as they allow disclosure of the identities of peer reviewers only after obtaining written consent of the peer reviewer. 42 C.F.R. § 480.133(a)(2)(iii). While the plaintiffs may disagree with HHS's regulation, it is permissible and thus must stand.

The district court's order compelling disclosure of the identities of peer reviewers conflicts with the applicable enforceable regulations. For this reason, we have no alternative but to reverse.