concurring in part, dissenting in part.
I agree with the court that HIPPA does not adopt state privilege law in a federal question suit brought in federal court, but rather Rule 501 of the Federal Rules of Evidence governs the evidentiary privileges applicable in such suits. Opinion at 925. I also agree that it is not for us to create a federal common law physician-patient privilege where none exists, and that the redacted medical records are not privileged. Opinion at 926. However, for several reasons, I disagree with the court’s conclusion that enforcing the subpoena creates an undue burden under Fed.R.Civ.P. 45(c)(3)(A)(iv). In passing HIPPA, Congress recognized a privacy interest only in “individually identifiable medical records” and not redacted medical records, and HIPPA preempts state law in this regard. The “de-identification” (redaction) of all identifying information from the medical records and the extensive protective order in place also eliminates any privacy interest in the records. Additionally, not only are the records in this case relevant, as the court acknowledges, but they are highly probative of the underlying issue. Finally, contrary to the court’s conclusion that quashing the subpoena occurs “at no substantial cost to federal substantive and procedural policy,” both suffer greatly. This court should enforce the subpoena. I therefore concur in part and dissent in part.
As the court recognizes, in section 264 of HIPPA, Congress authorized the Secretary of Health and Human Services to promulgate regulations to protect the privacy of medical records. Opinion at 924 (■citing 42 U.S.C. § 1320d-2). Therefore, HIPPA and the related regulations determine the privacy interests at stake. While tediously detailed, these regulations appear to have thoroughly considered and resolved the privacy concerns expressed by the hospital and the court.
Section 164.502, which sets forth the general rules for the use and disclosure of “protected health information,” provides that “[a] covered entity may not use or disclose protected health information, except as permitted or required by this sub-part or by subpart C of part 160 of this subchapter.” 45 C.F.R. § 164.502(a). Before looking to the various exceptions, the initial question is whether the information sought in this case is “protected health information.” The regulations define “protected health information” as “individually identifiable health information.” 45 C.F.R. § 160.103. Both Congress and HHS define “individually identifiable health information” as information that “is created or received by a health care provider, health plan, employer, or health care clearinghouse; and relates to the past, present, or future physical or mental health or condition of an individual, the provision of health care to an individual, or the past, present, or future payment for the provision of health care to an individual, cmd-(i) identifies the individual; or (ii) with respect to which there is a reasonable basis to believe the information can be used to identify the individual.” 42 U.S.C. 1320d(6); 45 C.F.R. § 160.103 (emphasis added).
In this case, the government seeks only redacted medical records and agrees that all identifying information may be removed before Northwestern makes the records available for its review. Because the records will be redacted, they will not identify the individual. Nor is there a reasonable *934basis to believe that the information can be used to identify the individual. Section 164.514(b) confirms the latter conclusion. Section 164.514(b)(2)(i) sets forth specific identifiers which, if removed, “de-identify,” the health records:
(A) Names;
(B) All geographic subdivisions smaller than a State, including street address, city, county, precinct, zip code, and their equivalent geocodes, except for the initial. three digits of a zip code if, according to the current publicly available data from the Bureau of the Census:
(1) The geographic unit formed by combining all zip codes with the same three initial digits contains more than 20,000 people; and
(2) The initial three digits of a zip code for all such geographic units containing 20,000 or fewer people is changed to 000.
(C) All elements of dates (except year) for dates directly related to an individual, including birth date, admission date, discharge date, date of death; and all ages over 89 and all elements of dates (including year) indicative of such age, except that such ages and elements may be aggregated into a single category of age 90 or older;
(D) Telephone numbers;
(E) Fax numbers;
(F) Electronic mail addresses;
(G) Social security numbers;
(H) Medical record numbers;
(I) Health plan beneficiary numbers;
(J) Account numbers;
(K) Certificate/license numbers;
(L) Vehicle identifiers and serial numbers, including license plate numbers;
(M) Device identifiers and serial numbers;
(N) Web Universal Resource Locators (URLs);
(O) Internet Protocol (IP) address numbers;
(P) Biometric identifiers, including finger and voice prints;
(Q) Full face photographic images and any comparable images; and
(R) Any other unique identifying number, characteristic, or code, except as permitted by paragraph (c) of this section;
45 C.F.R. § 164.514(b)(2)®.
Once these identifiers are redacted, the medical records are no longer “individually identifiable health information.” 45 C.F.R. § 164.514(a).1 Under HIPPA and the implementing regulations, there is no protected privacy interest in non-identifiable health information. Again, the regulations confirm this conclusion. 45 C.F.R. § 164.502(d)(2) provides:
Uses and disclosures of de-identified information. Health information that meets the standard and implementation specifications for de-identification under § 164.514(a) and (b) is considered not to be individually identifiable health information, i.e., de-identified. The requirements of this subpart do not apply to information that has been de-identified in accordance with the applicable requirements of § 164.514 ....
45 C.F.R. § 164.502(d)(2) (emphasis added).
*935Because the government seeks only redacted records that are not individually identifiable, under HIPPA there is no privacy interest in those records. However, even if the records were “individually identifiable,” they would still be subject to the general privacy rules governing use and disclosure of protected health information set forth in § 164.502. As noted above, the privacy protection afforded in that section provides several exceptions. 45 C.F.R. § 164.502(a) (“A covered entity may not use or disclose protected health information, except as permitted or required by this subpart or by subpart C of part 160 of this subchapter.”) (emphasis added). Of relevance here is 45 C.F.R. 164.512(e)(l)(i), which authorizes the disclosure of protected health information pursuant to a court order. In this case, the government obtained a court order authorizing the disclosure of the medical records. Under the regulations, such an order negates any need to redact identifying information.2 45 C.F.R. § 164.512(e)(l)(i). Yet, as the government stressed at oral argument, it has no need for, nor desire to know, the individual identities of the patients. Therefore, it is only seeking the relevant redacted medical records. Such redacted records are afforded no privacy protection under HIPPA, logically so because the redacted records have no identifiably private information to expose. And although Illinois law has adopted an expansive view of privilege that includes redacted medical records, as the court recognizes, Illinois law does not govern this question.
That should end the inquiry. But instead the court resurrects the privacy question through the “undue burden” language of Fed.R.Civ.P. 45(c)(3)(A)(iv). Rule 45(c)(3)(A)(iv) provides that a court may quash or modify a subpoena if it “subjects a person to undue burden.” Fed.R.Civ.P. 45(c)(3)(A)(iv). In the court’s view, compliance with the subpoena would impose an undue burden (i.e. “potential psychological cost”) on the women whose redacted records were subpoenaed. Such an undue burden exists, according to the court, because the potential loss of privacy outweighs the probative value of the medical records. See opinion at 927 (stating that the Illinois district court’s finding that the “potential loss of privacy that would ensue were these medical records used in a case in which the patient was not a party ... outweighs the loss to the government through non-disclosure” is “solidly based”). This conclusion is wrong on several levels.
Initially, to reiterate, HIPPA and the implementing regulations recognize that there is no loss of privacy where the medical records are redacted (or in HIPPA jargon, “de-identified”). Nor is it reasonable to believe that the unidentified 45 women have “acquaintances ... who will put two and two together, ‘out’ the 45 women, and thereby expose them to threats, humiliation, and obloquy.” Opinion at 929. In fact, there is no reason to believe that the women themselves have any idea that their records are among the few sought by the government in this case.3 But even if they knew,4 no one else *936ever would, because all of the information that could reasonably be used to identify them will be redacted, see 45 C.F.R. § 164.514(b)(2)®, and none of the information — not even the redacted non-identifying information — will ever be made public, much less paraded in court or placed on the Internet within the reach of “skillful Googlers.” Opinion at 929. That is guaranteed by the additional security of the protective order entered in this case in the Southern District of New York. See, e.g., Reproductive Serv., Inc. v. Walker, 439 U.S. 1307, 1308, 99 S.Ct. 1, 58 L.Ed.2d 16 (1978) (Brennan, J., in- chambers) (dissolving stay of subpoena seeking abortion records of non-party patients on condition that patient names were redacted and parties agreed to a protective order to ensure privacy of all patients).
The court’s erroneous conclusion that a privacy interest exists in the redacted documents leads to the unnecessary attempt to assess the probative value of the evidence. Notably, the district court (Judge Koeoras) did not reach the undue burden of compliance issue of Fed. R.Civ.P. 45(c)(3)(A)(iv). In the interest of time, with the trial date at hand, the court bypassed a remand and accepted the district court’s findings on the privilege issue and applied them to the undue burden question. It then in effect agrees with the district court that there is little if any probative value in the requested documents. Based on the complaint, Dr. Hammond’s declaration, the congressional findings when it passed the law, and the arguments made by the government and the hospital (both very limited since privilege, not probative value, was the issue argued below), there is significant probative value. But that is not for us to .decide, as the probative value of the evidence ■ has already been determined. District Court Judge Casey, who is presiding over the underlying case, believes the information is relevant, so much so, that he has indicated that if it is not produced, he would consider lifting the stay and dismissing the case (or at least dismissing Dr. Hammond from the case). This should also make clear that Judge Casey believes the evidence is not just relevant “in the attenuated sense,” opinion at 930, but highly probative , to the difficult question he will face starting on March 29. If any deference is owed, it is to the presiding judge — the judge who handled this case pre-trial and who knows the arguments presented by both sides, and the judge who will need all (non-privileged) relevant evidence available to allow him to make the necessary factual findings to determine this difficult and contentious constitutional case.
However, while recognizing that “[t]he merits of the dispute are for determination at trial,” opinion at 927, the court nonetheless interjects its own theory of the case and its own judgment of the probative value of the evidence. For instance, the court states: “What the government would like to show, in refutation of Dr. Hammond’s impending testimony, is that D & E is always an adequate alternative, from the standpoint of a pregnant woman’s health, to the D & X procedure. The government has failed to explain how the record of a D .& X abortion would show this.” Opinion at 930. But the government’s document request was not so structured: The government did not ask for the records of the D & X abortions identified by Dr. Hammond, -but rather requested the redacted medical records of patients who had abortions — both the D & E and D *937& X variety — for the reasons asserted by Dr. Hammond as justifying a partial-birth abortion. For instance, Dr. Hammond stated that he sometimes performed abortions for women to protect their health after they learned that “their fetuses have anomalies that are often quite severe.” Declaration ¶ 4. The government requested the patient records for 2003 of any women who had an abortion during their 19th or 20th week of pregnancy, (whether partial-birth or D & E) for that reason. Interrogatories 1 At 3; Document Request at 7. As the government explained at oral argument, those records are highly relevant to the question of medical necessity because, if they show that Dr. Hammond did not regularly perform partial-birth abortions under those circumstances, that would demonstrate that Dr. Hammond does not believe a partial-birth abortion is necessary to protect the women’s health. Of course, there could be some variations in the medical conditions of the individual cases that explain why Dr. Hammond used a different method, but Dr. Hammond remembers few, if any, of the circumstances surrounding the abortions. Opinion at 928. Thus, the only way the government (and the trial judge) can assess Dr. Hammond’s contention that partial-birth abortions are medically necessary to protect the women’s health is to review the medical records of the patients with the conditions that Dr. Hammond referenced.
The court rejects this theory, stating: “But such information would be unlikely to be found in Hammond’s records in view of his strongly expressed preference for using the D & X method on patients in fragile health. It would be much more likely to be found in the records, not sought by the government, of physicians who perform D & E rather than D & X abortions on such women.” Opinion at 927. But that is exactly the point: The government does not know what is to be found in Dr. Hammond’s medical records. It only knows what could be found there— evidence that, notwithstanding Dr. Hammond’s declaration that he strongly prefers using the D & X method of abortion on patients in fragile health, in practice, he does not use that procedure. Such evidence would be highly probative, as the court itself implies by recognizing it “would be unlikely to be found in Hammond’s records in view of his strongly expressed preference for using the D & X method.”
In fact, the relevance here cannot be overstated: Congress made explicit findings that a partial-birth abortion is never medically necessary to protect a women’s health. Yet, Dr. Hammond claims Congress is wrong. The court concisely lays out Dr. Hammond’s argument: In a D & X (partial-birth) abortion, “the fetus is destroyed after the lower extremities, and sometimes the torso, have emerged from the womb and only the head remains inside,” and this, according to Dr. Hammond is safer then the D & E procedure, where “the fetus is destroyed while it is still entirely within the womb .... Opinion at 927. Dr. Hammond seeks to testify accordingly, and it is therefore imperative that the government be able to determine the veracity of his testimony. There is no better way than by determining if Dr. Hammond’s actual practice supports his testimony. And this is not a question only of impeachment, but rather concerns the heart of this case.
Moreover, as the government explained during oral argument, the medical records are highly relevant to its case because its experts must be able to review Dr. Hammond’s files to determine whether, in their expert opinion, a D & X procedure was the most appropriate procedure, as Dr. Hammond claims. The court recognizes that “[t]he need for a health exception to the ban in the Partial-Birth Abortion Ban Act will pivot on the clash of expert witnesses *938in the New York trial.” Opinion at 931. Yet, the court refuses to recognize the importance of the redacted records to the government’s case, even after the government explained the need for its experts to review the files to form independent expert opinions.
The medical records are also highly relevant to a second congressional finding, namely, that a “partial-birth abortion poses serious risks to the health of a woman undergoing the procedure.” 117 Stat. 1201. Congress detailed numerous risks it found posed by partial-birth abortions. Although the government did not point this out during oral argument, Northwestern’s attorney alerted the court to the fact that the medical records will show whether there were any complications from the abortion, and this evidence is highly probative to the underlying constitutional challenge.5
The court also questions whether the government sincerely wants to determine “whether D & X abortions are ever medically indicated,” because the government did not seek summary statistics of all circumstances in which such abortions are performed. Opinion at 929. But as the government pointed out at oral argument, it was trying to limit the burden on Northwestern by confining its document request to those specific situations where Dr. Hammond claimed a partial birth abortion was necessary to preserve the mother’s health. See Fed.R.Civ.P. 45(c)(1) (“A party or attorney responsible for the issuance and service of a subpoena shall take reasonable steps to avoid imposing undue burden or expense on a person subject to that subpoena.”). And it succeeded, maybe even better than the government had hoped: During oral argument, the government learned for the first time that there are only 45 records that satisfy its document request. Given that Dr. Hammond stated in his declaration that he performs, teaches or supervises about 300 abortions a year, and that the government sought the records for a two- to three-year time frame, it probably surprised the government to learn that there were only 45 relevant records, with the rest apparently unrelated to the mother’s or fetus’s health.
In any event, the limited scope of the document request, and the government’s agreement to redact the records — something not required by HIPPA — if anything, refutes any questioning of the government’s motives or the court’s implication that the government is on a fishing expedition. Opinion at 930. Although contradictory, the court also chastises the government for not asking for enough records, implying that since the government did not ask for all relevant documents, the documents it did request were somehow less than relevant. Granted, there were many more relevant records that the government did not seek, but the government should not be impugned for prudently limiting its document request to those few medical records Dr. Hammond directly referenced.6
*939That brings us back to the question of undue burden, which, along with HIPPA, should have been the focus of the narrow question before the district court and this court in this case. Under Rule 45, a court may quash a subpoena where it creates an undue burden. There is no such burden in this case because HIPPA establishes that there is no privacy interest in redacted records and those records are highly relevant to the constitutional challenge to the Partial Birth Abortion Ban Act. The only burden identified by the court seems to be a “potential psychological cost.” Opinion at 927. Even assuming that is the kind of “burden” Rule 45 contemplates, reliance on that as a burden in effect creates a privilege where none exists.7
Finally, contrary to the court’s conclusion, quashing the subpoena in this case does come at a “substantial cost to federal substantive and procedural policy.” The court’s ruling may well be the death knell for Dr. Hammond’s claim, as the district court made clear that it believed the records relevant and that it would consider dismissing the case if the records were not produced. Given that the government cannot adequately cross-examine Dr. Hammond, the district court would be well within its rights to bar Dr. Hammond’s testimony, which will not only harm his case, but also the other plaintiffs’. The court’s decision also comes at a substantial cost to the federal policy adopted by HIP-PA. Lastly, and most significantly, it comes at a cost to the truth of Congress’ *940findings that a partial-birth abortion is never necessary to protect a woman’s health and poses significant health risks, and to the constitutionality of such a law. For these and the foregoing reasons, I would enforce the subpoena to produce the designated records.
. The government does not object to the removal of these identifiers and in fact has consented to redaction beyond that required by Section 164.514(b)(2)(i), for instance by agreeing that Northwestern may delete the state of residence. The fact that the regulations allow the disclosure of the patient’s state disproves Northwestern's assertion that, because the Hospital is located in Chicago, the patients could be identified since they would be assumed to be from Illinois. Such an assumption is unreasonable given that HIPPA allows for that very disclosure, while still treating the records as de-identified. But in any event, the government does not request that information.
.As the court also recognizes, the government did not need a court order in this case because it obtained a protective order securing the confidentiality of the redacted records. Opinion at 925. Thus, the government complied with the privacy protections established by HIPPA in three independent ways: by obtaining a court order; by obtaining a protective order; and by seeking only redacted records.
. Even if some are aware of the subpoena, there is nothing in the record to support the conclusion that “[t]he women whose records these are do not want them collected and examined by the Department of Justice and presented in evidence in the New York trial.” Opinion at 927.
. Notwithstanding the court's discussion of the notice procedures of HIPPA, see opinion *936at 926, HIPPA does not require notice where a court order authorizes disclosure, 45 C.F.R. § 164.512(e)(1), where there is a protective order in . place, 45 C.F.R. § 164.512(e)(l)(ii)(B), or where the records are redacted, 45 C.F.R. § 164.502(d)(2).
. Northwestern also acknowledged another point of relevancy during questioning: When asked whether the records could possibly demonstrate that the woman's life — and not just her health — was at risk, Northwestern's attorney responded, "yes, but that would help the other side.” This case is not about sides, but about the document request, and providing the district court with the evidence it needs to resolve the constitutional question before it.
. The court also charges the government with being evasive on the question of the probative value of the medical documents. Opinion at 927. It is true that the government’s main focus was not on the probative value of the medical records, but that is neither surprising nor nefarious, given the arguments below and the district court’s ruling. The district court in this case ruled that Illinois privilege law governed and not HIPPA. The question of the relevance and probative value of the *939documents was not central to the question of whether Illinois privilege law applied. This court appropriately reversed the district court on the privilege issue without delving into the question of relevance. Thus, it is not surprising that the government’s opening brief did not focus on the relevancy of the documents. Moreover, although Northwestern argued below that the documents were not relevant, it did so in the context of arguing for a federal common law doctor-patient privilege. On appeal, the government did not need to argue relevancy to address that legal issue, and in fact, this court again appropriately rejected the idea of a federal common law privilege without addressing the question of relevance. Relevance only became relevant once the court discounted the import of HIPPA de-identification and looked to a balancing test under Rule 45(c)(3)(A)(iv). Again, that the government's opening brief did not focus on this question is not surprising given that Northwestern’s Rule 45(c)(3)(A)(iv) undue burden argument below was limited to three short paragraphs, and the only tangential reference to relevance in its opening brief came from this sentence: “The Attorney General’s subpoena is an unacceptable intrusion into the privacy of the Hospital’s patients, promising no significant contribution to the ascertainment of truth in NAF v. Ashcroft." Memorandum in Support of Northwestern Memorial Hospital’s Motion to Quash Subpoena at 20. After Northwestern changed direction on appeal, to argue that the production of the records constituted an undue burden because the records were not relevant, the government responded at length in its reply brief. See Appellant's Reply Brief at 8 — 11. At oral argument, the government also elaborated on the relevance of the documents, not in a vague or evasive way, but by specifically demonstrating that the medical records are both relevant and highly probative of the issues in the underlying case. See supra at 8 — 10.
. Northwestern does not claim that it is an undue burden to comply with the subpoena because it is too costly, difficult or time consuming to produce the redacted records, only that it may negatively impact its reputation with past and future patients. The court agrees, calling it a “potential cost in lost goodwill,” opinion at 930, because Northwestern “will lose the confidence of its patients, and persons with sensitive medical conditions may be inclined to turn elsewhere for medical treatment.” Opinion at 929. However, this is not an authentic "cost,” because the same federal regulations apply equally to all hospitals. These regulations put all hospitals on the same footing, thus negating any basis for a patient rejecting a hospital’s care because a federal court orders the production of redacted records pursuant to a federal regulatory standard.