LeRoy CARHART, M.D., on behalf of themselves and the patients they serve; William G. Fitzhugh, M.D., on behalf of themselves and the patients they serve; William H. Knorr, M.D., on behalf of themselves and the patients they serve; Jill L. Vibhakar, M.D., on behalf of themselves and the patients they serve, Appellees, v. Alberto GONZALES, in his official capacity as Attorney General of the United States, and his employees, agents, and successors in office, Appellant, and Susan Frietsche; David S. Cohen; Stacey I. Young, Interested Parties. Margie Riley, et al., Amici on Behalf of Appellee, 413 F.3d 791

BYE, Circuit Judge. This case presents a challenge to the federal Partial-Birth Abortion Ban Act of 2003, Pub.L. No. 108-105, 117 Stat. 1201 (codified at 18 U.S.C. § 1531). The day the President signed the Act into law, plaintiffs filed suit in the United States District Court for the District of Nebraska seeking an injunction against enforcement of the Act. After a trial, the district court held the Act unconstitutional on several grounds. The government appeals. We affirm the judgment of the district court. I A In 2000, the Supreme Court handed down its decision in Stenberg v.

Carhart, 530 U.S. 914, 120 S.Ct. 2597, 147 L.Ed.2d 743 (2000), which found Nebraska’s partial-birth abortion ban unconstitutional for two separate reasons. First, the Court determined the law was unconstitutional because it did not contain an exception to preserve the health of the mother. Second, the Court determined the law was worded so broadly it covered the vast majority of late-term abortions and thus imposed an undue burden on the right to abortion itself. In the eight years before the Court’s decision in Steinberg, at least thirty states passed laws banning partial-birth abortions.

See id. at 983, 120 S.Ct. 2597 (Thomas, J., dissenting). In 1996 and 1997, Congress enacted prohibitions on partial-birth abortions, however, President Clinton vetoed them. Id. at 994 n. 11, 120 S.Ct. 2597 (Thomas, J., dissenting). In 2003, Congress enacted, and President George W. Bush signed, the Partial-Birth Abortion Ban Act of 2003. The Act exposes “[a]ny physician who, in or affecting interstate or foreign commerce, knowingly performs a partial-birth abortion and thereby kills a human fetus” to up to two years of imprisonment. 18 U.S.C. § 1531(a).

The Act goes on to define a “partial-birth abortion” as an abortion in which the person performing the abortion: (

A) deliberately and intentionally vaginally delivers a living fetus until, in the case of a head first presentation, the entire fetal head is outside the body of the mother, or, in the case of a breech presentation, any part of the fetal trunk past the navel is outside the body of the mother, for the purpose of performing an overt act that the person knows will kill the partially delivered living fetus; and (

B) performs the overt act, other than completion of delivery, that kills the partially delivered living fetus .... Id. § 1531(b)(1). The Act contains an exception allowing the performance of “a partial-birth abortion that is necessary to save the life of the mother.” Id. § 1531(a). The Act does not, however, contain an exception for the preservation of the health of the mother. Presumably recognizing that the Act is similar (though not identical) to the Nebraska law found unconstitutional in Sten-berg, Congress made several findings and declarations in the Act.

Congress “f[ound] and declare[d]” that “under well-settled Supreme Court jurisprudence, the United States Congress is not bound to accept the same factual findings that the Supreme Court was bound to accept in Stenberg.” Partial-Birth Abortion Ban Act of 2003 § 2(8), 117 Stat. at 1202. Congress concluded that a “moral, medical, and ethical consensus exists that the practice of performing a partial-birth abortion ... is a gruesome and inhumane procedure that is never medically necessary and should be prohibited.” § 2(1), 117 Stat. at 1201.

In addition to determining there is “substantial evidence” that partial-birth abortions are never medically necessary, Congress also concluded partial-birth abortions “pose[ ] serious risks to the health of the mother undergoing the procedure.” §§ 2(13), 2(14), 117 Stat. at 1203-04. After a trial, the district court found the Act unconstitutional on two separate grounds. First, the district court concluded Congress’s finding regarding a medical consensus was unreasonable and thus the Act was unconstitutional due to its lack of health exception.

Second, the district court concluded the Act covered the most common late-term abortion procedure and thus imposed an undue burden on the right to an abortion. B The procedures in question in this case are used during late-term abortions and we therefore must, for context, present some basic information regarding these procedures. There are three primary methods of late-term abortions: medical induction; dilation and evacuation (D & E); and dilation and extraction (D & X). In a medical induction, formerly the most common method of second-trimester abortion, a physician uses medication to induce premature labor.

Stenberg, 530 U.S. at 924, 120 S.Ct. 2597. In a D & E, now the most common procedure, the physician causes dilation of the woman’s cervix and then “the physician reaches into the woman’s uterus with an instrument, grasps an extremity of the fetus, and pulls.” Women’s Med. Prof’l Corp. v. Taft, 353 F.3d 436, 439 (6th Cir.2003). “When the fetus lodges in the cervix, the traction between the grasping instrument and the cervix causes dismemberment and eventual death, although death may occur prior to dismemberment.” Id. This process is repeated until the entire fetus has been removed.

D & X and a process called intact D & E are what are “now widely known as partial birth abortion.” Id. In these procedures, the fetus is removed “intact” in a single pass. If the fetus presents head first, the physician collapses the skull of the fetus and then removes the “intact” fetus. Stenberg, 530 U.S. at 927, 120 S.Ct. 2597. This is what is known as an intact D & E. If the fetus presents feet first, the physician “pulls the fetal body through the cervix, collapses the skull, and extracts the fetus through the cervix.” Id.

This is the D & X procedure. “Despite the technical differences” between an intact D & E and a D & X, they are “sufficiently similar for us to use the terms interchangeably.” Id. at 928, 120 S.Ct. 2597. II As a preliminary matter, although the plaintiffs purported to bring a facial challenge to the Act, the district court expressed confusion over whether its judgment declared the Act facially unconstitutional or unconstitutional as applied to the plaintiffs. See Carhart v.

Ashcroft, 331 F.Supp.2d 805, 1042-47 (D.Neb.2004) (stating the district court “do[es] not know” if its ruling was facial or as applied and leaving “that for others to determine”). This is a question of law and we therefore review it de novo. See, e.g., United States v. Jeffries, 405 F.3d 682, 684 (8th Cir.2005). The traditional standard for evaluating a facial challenge was set forth in United States v. Salerno, 481 U.S. 739, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987).

In Salerno, the Supreme Court explained that a “facial challenge to a legislative Act is, of course, the most difficult challenge to mount successfully, since the challenger must establish that no set of circumstances exists under which the Act would be valid.” Id. at 745, 107 S.Ct. 2095. In Stenberg, however, the Supreme Court struck down Nebraska’s partial-birth abortion ban as facially unconstitutional without applying the Salerno standard.

In fact, the approach taken in Stenberg was fundamentally inconsistent with Salerno’s “no set of circumstances” test in that it regarded rarity of the need for a particular procedure as “not highly relevant.” Stenberg, 530 U.S. at 934, 120 S.Ct. 2597. The Salerno test is also inconsistent with the general undue burden analysis for abortion statutes set forth in Planned Parenthood v. Casey, 505 U.S. 833, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992). This has led the vast majority of circuit courts to apply these abortion-specific standards in place of Salerno. See Planned Parenthood of N. New England v.

Heed, 390 F.3d 53, 57-59 (1st Cir.2004) (collecting cases), cert. granted sub nom. Ayotte v. Planned Parenthood, — U.S. —, 125 S.Ct. 2294, 161 L.Ed.2d 1088 (May 23, 2005); Richmond Med. Ctr. for Women v. Hicks, 409 F.3d 619, 627-28 (4th Cir.2005) (same). We have previously declined to apply the “no set of circumstances” test in the context of facial challenges to abortion restrictions in Planned Parenthood, Sioux Falls Clinic v.

Miller, 63 F.3d 1452, 1458 (8th Cir.1995), where we explained we would “follow what the Supreme Court actually did — rather than what it failed to say” and thus. applied Casey’s undue burden test. We will again follow what the Supreme Court “actually did” and apply the test from Stenberg rather than the one from Salerno. We therefore join every circuit that has addressed the question. See Hicks, 409 F.3d at 628; Planned Parenthood of Idaho, Inc. v. Wasden, 376 F.3d 908, 921 n. 10 (9th Cir.2004); Planned Parenthood of the Rocky Mountains Servs., Corp. v. Owens, 287 F.3d 910, 919 (10th Cir.2002).

Thus, if the Act fails the Stenberg test, it must be held facially unconstitutional. Ill We begin our analysis with the Supreme Court’s decision in Stenberg. That case has engendered some disagreement as to the proper standard for evaluating the necessity of a health exception. The proper reading of Stenberg is a question of law and therefore is reviewed de novo. See, e.g., Jeffries, 405 F.3d at 684.

The government argues Stenberg merely examined the specific factual record before the Court, and thus a health exception is only required when a banned procedure is actually “necessary, in appropriate medical judgment, for the preservation of the health of the mother.” Stenberg, 530 U.S. at 930, 120 S.Ct. 2597 (internal quotations omitted).

Plaintiffs, in contrast, contend that “where substantial medical authority supports the proposition that banning a particular abortion procedure could endanger women’s health, Casey requires the statute to include a health exception when the procedure is “ ‘necessary, in appropriate medical judgment, for the preservation of the life or health of the mother.’ ” ” Stenberg, 530 U.S. at 938, 120 S.Ct. 2597 (quoting Casey, 505 U.S. at 879, 112 S.Ct. 2791 (quoting Roe v. Wade, 410 U.S. 113, 165, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973))).

The government argues that Stenberg embodies a -lenient standard, and further urges that congressional factfinding must be afforded deference under Turner Broadcasting v. FCC, 512 U.S. 622, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) (Turner I), and Turner Broadcasting v. FCC, 520 U.S. 180, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997) (Turner II). The government contends that because (in its opinion) Congress is. afforded deference in factfinding as a general proposition, the district court’s adoption of the “substantial medical authority” standard amounts to an implicit overruling of the Turner line of cases.

According to the government, the “substantial medical authority” standard “must [therefore] be understood as[,] at most[,] a rule of decision in the absence of congressional findings, not as a basis for disregarding such findings.” Br. of Appellant at 33. The government’s argument, however, fundamentally misconstrues the threshold issue, for our task lies not in identifying who gets to decide, but rather in identifying the precise question that must be answered.

The other end of the spectrum on potential readings of Stenberg is exemplified by a recent decision in which the Fourth Circuit addressed Stenberg’s health exception requirement standard in a ease involving a state partial-birth abortion statute. Hicks, 409 F.3d at 625-26. The Fourth Circuit held that Stenberg “established the health exception requirement as a per se constitutional rule.” Id. at 625.

The court explained that “[t]his rule is based on substantial medical authority (from a broad array of sources) recognized by the Supreme Court, and this body of medical authority does not have to be reproduced in every subsequent challenge to a ‘partial birth abortion’ statute lacking a health exception,” and therefore all statutes regulating partial-birth abortion must contain a health exception. Id. Several district courts have, at least implicitly, taken this position as well. See, e.g., Reproductive Health Servs. of Planned Parenthood v.

Nixon, 325 F.Supp.2d 991, 994-95 (W.D.Mo.2004); WomanCare of Southfield, P.C. v. Granholm, 143 F.Supp.2d 849, 855 (E.D.Mich.2001); Summit Med. Assocs. v. Siegelman, 130 F.Supp.2d 1307, 1314 (M.D.Ala.2001); Daniel v. Underwood, 102 F.Supp.2d 680, 684 (S.D.W.Va.2000). We agree with the Fourth Circuit that Stenberg establishes a per se constitutional rule in that the constitutional requirement of a health exception applies to all abortion statutes, without regard to precisely how the statute regulates abortion. See Heed, 390 F.3d at 59 (applying Stenberg to parental notification law).

As the Ninth Circuit recently explained: “Any abortion regulation must contain adequate provision for a woman to terminate her pregnancy if it poses a threat to her life or her health.” Wasden, 376 F.3d at 922. While Stenberg’s health exception rule undoubtedly applies to all abortion statutes, such a proposition does not explain how to evaluate whether a given restriction poses a constitutionally significant threat to the mother’s health. We believe the appropriate question is whether “substantial medical authority” supports the medical necessity of the banned procedure.

See Stenberg, 530 U.S. at 938, 120 S.Ct. 2597; id. at 948, 120 S.Ct. 2597 (O’Connor, J., concurring); see also Planned Parenthood Fed’n of Am. v. Ashcroft, 320 F.Supp.2d 957, 1033 (N.D.Cal.2004); Nat’l Abortion Fed’n v. Ashcroft, 330 F.Supp.2d 436, 487-90 (S.D.N.Y.2004); Carhart, 331 F.Supp.2d at 1008. The Stenberg Court determined medical necessity (as that term was used in Casey) does not refer to “an absolute necessity or to absolute proof.” Stenberg, 530 U.S. at 937, 120 S.Ct. 2597.

Rather, “appropriate medical judgment” must “embody the judicial need to tolerate responsible differences of medical opinion.” Id. Recognition of this principle was driven by the Court’s concern that “the division of medical opinion about the matter at most means uncertainty, a factor that signals the presence of risk, not its absence.” Id. Thus, when “substantial medical authority” supports the medical necessity of a procedure in some instances, a health exception is constitutionally required.

In effect, we believe when a lack of consensus exists in the medical community, the Constitution requires legislatures to err on the side of protecting women’s health by including a health exception. In dissent, both Justice Kennedy and Justice Thomas criticized the Stenberg majority for imposing what they believed was a high burden on legislatures. Justice Kennedy commented that by disagreeing with Nebraska, the Court was effectively “[rjequiring Nebraska to defer to Dr. Car-hart’s judgment[, which was] no different from forbidding Nebraska from enacting a ban at all; for it is now Dr.

Leroy Carhart who sets abortion policy ....” Id. at 965, 120 S.Ct. 2597 (Kennedy, J., dissenting). Justice Thomas characterized the majority opinion as requiring a health exception “because there is a ‘division of opinion among some medical experts Id. at 1009, 120 S.Ct. 2597 (Thomas, J., dissenting) (quoting id. at 936-37, 120 S.Ct. 2597). “In other words, unless a State can conclusively establish that an abortion procedure is no safer than other procedures, the State cannot regulate that procedure without including a health exception.” Id. (Thomas, J., dissenting).

Although the Stenberg majority did not believe the rule it announced gave individual doctors an absolute veto over legislatures, it emphasized that a health exception is required where “substantial medical authority” supports the medical necessity of a procedure. Id. at 938, 120 S.Ct. 2597. Such language would be rendered essentially meaningless if we accepted the government’s reading of the case, a reading that would conform to neither the majority’s reasoning nor to the dissenters’ concerns.

In sum, we conclude Stenberg requires the inclusion of a health exception whenever “substantial medical authority” supports the medical necessity of the prohibited procedure. TV A Having identified the proper question, we now turn to determining how this question should be answered.

The government argues the Turner line of cases requires courts to “ ‘accord substantial deference to the predictive judgments of Congress,’ ” and the “sole obligation” of reviewing courts “is ‘to assure that, in formulating its judgments, Congress has drawn reasonable inferences based on substantial evidence.’ ” Turner II, 520 U.S. at 195, 117 S.Ct. 1174 (quoting Turner I, 512 U.S. at 665-66, 114 S.Ct. 2445).

Thus, under the government’s formulation, we would be bound by Congress’s determination that a “moral, medical, and ethical consensus exists that the practice of performing a partial-birth abortion” is never medically necessary, so long as this apparent factual determination is reasonable and supported by substantial evidence. The government’s argument is predicated on an erroneous assumption: that the “substantial medical authority” standard is a question of fact.

While questions of law and questions of fact sometimes can be neatly separated, such questions are often intermingled and identified as so-called mixed questions of fact and law. See, e.g., Ornelas v. United States, 517 U.S. 690, 696, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996). Whether a partial-birth abortion is medically necessary in a given instance would be a question of fact; for in any given instance it would be either true or false that a partial-birth abortion is medically necessary. There may be conflicting expert opinions, but only one can actually be 'right in any given set of medical circumstances.

In contrast, whether the record in a particular lawsuit reflects the existence of “substantial medical authority” supporting the medical necessity of such procedures is a question that is different in kind; it asks only whether there is a certain quantum of evidence to support a particular answer, not which of the divergent opinions is ultimately correct. Reviewing the record to determine if the evidence presented suffices to support the conclusion reached by the lower court is typically treated as a matter of law. See, e.g., Howard v.

Massanari, 255 F.3d 577, 580 (8th Cir.2001) (applying de novo review of the Social Security Commissioner’s conclusion despite prior district court review); United States v. Thompson, 285 F.3d 731, 733 (8th Cir.2002) (reviewing the sufficiency of the evidence de novo). We must, of course, examine the evidence, but the legal question inherent in this inquiry is whether such record evidence constitutes “substantial medical authority” in a given case. This case differs slightly from the typical case in which we review the evidence to determine if the record is sufficient to support the lower court’s conclusion.

Under the “substantial medical authority” standard, our review of the record is effectively limited to determining whether substantial evidence exists to support the medical necessity of partial-birth abortions without regard to the factual conclusions drawn from the record by the lower court (or, in this case, Congress). Thus, Stenberg created a standard in which the ultimate factual conclusion is irrelevant. Under this standard, we must examine the record to determine if “substantial medical authority” supports the medical necessity of the banned procedures.

If it does, then a health exception is constitutionally required. If the need for a health exception is not supported by “substantial medical authority,” by contrast, then the state is free to impose the restriction without providing a health exception. We believe an example from the Supreme Court’s First Amendment jurisprudence is instructive here. In New York Times v.

Sullivan, 376 U.S. 254, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964), the Court held that the First Amendment “prohibits a public official from recovering damages for a defamatory falsehood relating to his official conduct unless he proves that the statement was made with ‘actual malice’— that is, with knowledge that it was false or with reckless disregard of whether it was false or not.” Id. at 279-80, 84 S.Ct. 710. To meet this burden, the public official must show actual malice by clear and convincing evidence. See, e.g., Campbell v. Citizens for an Honest Gov’t, Inc., 255 F.3d 560, 569 (8th Cir.2001). In Bose Corp. v.

Consumers Union of the United States, Inc., 466 U.S. 485, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984), the Court faced the question of whether Federal Rule of Civil Procedure 52(a), which makes facts subject only to review for clear error, was the appropriate standard for reviewing a finding of actual malice. Id. at 487, 104 S.Ct. 1949. An individual’s state of mind is a question of historical fact and would thus normally be reviewed only for clear error. See, e.g., Hickey v.

Reeder, 12 F.3d 754, 756-57 (8th Cir.1993) (holding that state of mind is a question of fact that is reviewed for clear error); see also Bose, 466 U.S. at 498 n. 15, 104 S.Ct. 1949 (noting that in Herbert v. Lando, 441 U.S. 153, 170, 99 S.Ct. 1635, 60 L.Ed.2d 115 (1979), the Court had referred “in passing” to actual malice as “ultimate fact”). The Court concluded, however, that the First Amendment requires independent appellate review.

The Bose Court explained that “[jjudges, as expositors of the Constitution, must independently decide whether the evidence in the record is sufficient to cross the constitutional threshold.” Id. at 511, 104 S.Ct. 1949.

The Court further stated that “independent inquiries of this kind are familiar under the settled principle that in cases in which there is a claim of denial of rights under the Federal Constitution, this Court is not bound by the conclusions of lower courts, but will reexamine the evi-dentiary basis on which those conclusions are founded.” Id. at 510, 104 S.Ct. 1949 (internal quotations and alterations omitted). Thus, despite the fact that an individual’s mental state is a question of pure historical fact, a determination of whether the record supports the finding of actual malice is a question of law.

See, e.g., Harte-Hanks Communications, Inc. v. Connaughton, 491 U.S. 657, 685, 109 S.Ct. 2678, 105 L.Ed.2d 562 (1989); Mercer v. City of Cedar Rapids, 308 F.3d 840, 849 (8th Cir.2002); see also Bose, 466 U.S. at 499, 104 S.Ct. 1949 (explaining the “New York Times rule emphasizes the need for an appellate court to make an independent examination of the entire record”). The same reasoning applies here.

While judges under Bose must determine whether clear and convincing evidence of an individual’s state of mind exists in an effort to protect that individual’s First Amendment rights, here we must examine the record to determine whether “substantial medical authority” supports the need for a health exception so as to guard against the denial of another constitutional right. As a result, the government’s argument regarding Turner deference is irrelevant to the case at hand. Our review is based on the record and is guided, as described below, by the legal conclusions reached by the Supreme Court in prior cases.

Therefore, we need not address the government’s assertions that federal courts must defer to congressional factfinding. B Courts engage in different types of fact-finding, as the facts that they find can be either of an adjudicatory or legislative nature. See Qualley v. Clo-Tex Int’l, Inc., 212 F.3d 1123, 1128 (8th Cir.2000). Adjudicatory facts are those relevant only to the particular parties involved in the case. United States v. Gould, 536 F.2d 216, 219 (8th Cir.1976).

Classic examples are “ ‘who did what, when, where, how and with what motive or intent.’ ” Id. (quoting 2 Kenneth Davis, Administrative Law Treatise § 15.03, at 353 (1958)). In contrast, legislative facts are those that have salience beyond the specific parties to the suit. Qualley, 212 F.3d at 1128. The medical necessity of particular abortion procedures clearly falls into this latter category, as such procedures are either sometimes medically necessary or they are not: the answer to this question does not vary from place to place or party to party.

While lower court conclusions drawn from the same body of evidence may vary from individual case to individual case, appellate courts can impose uniformity within their jurisdictions by according no deference to a lower court’s record-based conclusions.

Indeed, adopting a deferential posture in such circumstances could lead to the absurd result where two district courts within the same circuit (perhaps even within the same state) might examine the same body of evidence and reach different conclusions as to the medical necessity of the partial-birth abortion procedures, but we would be forced to affirm both because the question is a close one. See Hope Clinic v. Ryan, 195 F.3d 857, 883-84 (7th Cir.1999) (en banc) (Posner, J., dissenting), vacated and remanded, 530 U.S. 1271, 120 S.Ct. 2738, 147 L.Ed.2d 1001 (2000); see also Lockhart v.

McCree, 476 U.S. 162, 169 n. 3, 106 S.Ct. 1758, 90 L.Ed.2d 137 (1986) (expressing doubt that “legislative facts” are reviewed deferentially because different courts can come to different conclusions from the same evidence). As Judge East-erbrook has cogently explained for the Seventh Circuit, the medical necessity of partial-birth abortion “must be assessed at the level of legislative fact, rather than adjudicative fact determined by more than 650 district judges. Only treating the matter as one of legislative fact produces the nationally uniform approach that Sten-berg demands.” A Woman’s Choice-E.

Side Women’s Clinic v. Newman, 305 F.3d 684, 688 (7th Cir.2002). The Neioman court recognized that “[fjindings based on new evidence could produce a new understanding, and thus a different legal outcome .... But if the issue is one of legislative rather than adjudicative fact, it is unsound to say that, on records similar in nature, Wisconsin’s law could be valid ... and Indiana’s law invalid, just because different district judges reached different conclusions about the inferences to be drawn from the same body of statistical work.” Id.; see also Hope Clinic, 195 F.3d at 884 (en banc) (Posner, J., dissenting).

Thus, although the Seventh Circuit prior to Stenberg had affirmed a trial court’s decision upholding a partial-birth abortion ban based on the trial court’s conclusion that partial-birth abortions are never medically necessary, the Supreme Court vacated the decision without regard to the specific facts found by that particular trial court. See Hope Clinic, 530 U.S. at 1271, 120 S.Ct. 2738. On remand, the Seventh Circuit held the state bans unconstitutional (in agreement with the parties). See Hope Clinic v.

Ryan, 249 F.3d 603, 604 (2001) (en banc) (decision on remand) (“[Bjoth Illinois and Wisconsin have conceded that their partial-birth-abortion statutes are unconstitutional under the approach the Court adopted in Stenberg. We agree with this assessment of Sten-berg ’s significance.”). While we are hesitant to read too much into the Supreme Court’s decision to vacate and remand Hope Clinic, its decision, along with the Seventh Circuit’s comments regarding Stenberg’s significance, is suggestive of a need to achieve constitutional uniformity through treatment of the issue as one of legislative fact.

In the specific context of a ban on partial-birth abortions, we join the reasoning of the Fourth Circuit and some of the district courts that have treated Stenberg as a per se constitutional rule.

In Sten-berg, the Court surveyed all of the available medical evidence (including the formal district court record, the district court records from other partial-birth abortion cases, amicus submissions, and some congressional records) and determined that “substantial medical authority” supported the need for a health exception. “[Tjhis body of medical authority does not have to be reproduced in every subsequent challenge to a ‘partial birth abortion’ statute lacking a health exception.” Hicks, 409 F.3d at 625.

Neither we, nor Congress, are free to disagree with the Supreme Court’s determination because the Court’s conclusions are final on matters of constitutional law. See, e.g., Dickerson v. United States, 530 U.S. 428, 437, 120 S.Ct. 2326, 147 L.Ed.2d 405 (2000) (“Congress may not legislatively supersede our decisions interpreting and applying the Constitution.”); City of Boerne v. Flores, 521 U.S. 507, 517-21, 117 S.Ct. 2157, 138 L.Ed.2d 624 (1997); Stell v. Savannah-Chatham County Bd. of Educ., 333 F.2d 55, 61 (5th Cir.1964) (“[N]o inferior federal court may refrain from acting as required by [Brown v.

Board of Education ] even if such a court should conclude that the Supreme Court erred as to its facts or as to the law.”). And because the medical necessity of a health exception is a question of legislative fact, subsequent litigants need not relitigate questions the Supreme Court has already addressed. See, e.g., Hicks, 409 F.3d at 625; N.J. Citizen Action v.

Edison Township, 797 F.2d 1250, 1268 (3d Cir.1986) (Weis, J., dissenting) (“The constitutional facts supporting a rule or doctrine must necessarily carry prece-dential weight so that government will be able to predict the validity of their regulatory actions. Thus, in large part the longevity of constitutional facts may be attributed to the doctrine of stare decisis and the important purposes that principle serves.”); Matthews v. Launius, 134 F.Supp. 684, 686-87 (D.Ark.1955) (recognizing that to succeed in a suit under Brovm, a plaintiff need not reprove Brown’s factual predicates).

This is not to say, however, that because the Supreme Court concluded “substantial medical authority” supported the need for a health exception in 2000, legislatures are forever constitutionally barred from enacting partial-birth abortion bans. Rather, the “substantial medical authority” test allows for the possibility that the evidentiary support underlying the need for a health exception might be reevaluated under appropriate circumstances.

Medical technology and knowledge is constantly advancing, and it remains theoretically possible that at some point (either through an advance in knowledge or the development of new techniques, for example), the procedures prohibited by the Act will be rendered obsolete. Should that day ever come, legislatures might then be able to rely on this new evidence to prohibit partial-birth abortions without providing a health exception.

V Stenberg identified what some refer to as “evidentiary circumstances” upon which the Court purportedly relied in determining whether “substantial medical authority” supported the need for a health exception.

The Stenberg Court noted (1) the district court’s conclusion that D & X significantly obviates health risks in certain circumstances and a highly plausible record-based explanation of why that might be so; (2) a division of opinion among medical experts regarding the procedure; and (3) an absence of controlled medical studies that address the safety and medical necessity of the banned procedures. 530 U.S. at 936-37, 120 S.Ct. 2597.

In evaluating the government’s case, we take Stenberg as the baseline and then determine if the government has proffered evidence sufficient to distinguish the present situation from Stenberg’s “evidentiary circumstances.” If the government marshals such evidence, we must then determine whether the evidence on the other side remains “substantial medical authority.” Because we conclude the government has not adduced evidence distinguishing this case from Stenberg, we need not attempt to define the precise contours of “substantial medical authority.” We know from Stenberg that “substantial medical authority” supports the conclusion that the banned procedures obviate health risks in certain situations.

For example, there is “substantial medical authority” (in the form of expert testimony and amici submissions) that these procedures reduce the risk of uterine perforation and cervical laceration because they avoid significant instrumentation and the presence of sharp fetal bone fragments. Stenberg, 530 U.S. at 930-34, 120 S.Ct. 2597. There is also evidence the procedure takes less time and thus reduces blood loss and prolonged exposure to anesthesia. Id. The banned procedure may also eliminate the risk posed by retained fetal tissue and embolism of cerebral tissue into the woman’s bloodstream. Id.

Moreover, there is evidence regarding the health advantages the banned procedures provide when the woman has prior uterine scarring or when the fetus is nonviable due to hydrocephaly. Id. There is some evidence in the present record indicating each of the advantages discussed in Stenberg are incorrect and the banned procedures are never medically necessary. See Carhart, 331 F.Supp.2d at 822-51. There were, however, such assertions in Stenberg as well. See Stenberg, 530 U.S. at 933-34, 120 S.Ct. 2597; id. at 964-66, 120 S.Ct. 2597 (Kennedy, J., dissenting).

Though the contrary evidence now comes from (some) different doctors, the substance of this evidence does not distinguish this case from Stenberg in any meaningful way. To avoid Stenberg, the government cannot simply claim Stenberg was wrongly decided, for we are bound by the Supreme Court’s conclusions. The facts in Stenberg were hotly contested, and simply asserting that the other side should have prevailed accomplishes nothing.

Rather, to succeed, the government must demonstrate that relevant evidentiary circumstances (such as the presence of a newfound medical consensus or medical studies) have in fact changed over time. If one thing is clear from the record in this case, it is that no consensus exists in the medical community. The record is rife with disagreement on this point, just as in Stenberg.

In fact, one of the government’s witnesses himself testified that no consensus exists in the medical community and further stated that there exists a “body of medical opinion,” including the “position[s] taken by [the] American College of Obstetrics and Gynecologists” (ACOG) and “a responsible group of physicians,” indicating that the procedures are indeed sometimes medically necessary. Carhart, 331 F.Supp.2d at 1012. The lack of consensus also extends to medical organizations.

The American Medical Association believes the banned procedures to be medically unnecessary while ACOG believes these procedures can be the most appropriate in certain situations. Id. at 843, 997. The Supreme Court relied on the ACOG view in particular in Stenberg, 530 U.S. at 935-36, 120 S.Ct. 2597. Moreover, the congressional findings quote “a prominent medical association’s” conclusion that “there is no consensus among obstetricians about its use.” Partial Birth Abortion Ban Act of 2003 § 2(14)(C), 117 Stat. at 1204 (internal quotations omitted).

In short, no medical consensus has developed to support a different outcome. See, e.g., Carhart, 331 F.Supp.2d at 1009 (concluding Congress’s determination that a consensus against the banned procedures existed is unreasonable and not supported by substantial evidence); Nat’l Abortion Fed’n, 330 F.Supp.2d at 488-89 (same); Planned Parenthood Fed’n of Am., 320 F.Supp.2d at 1025 (same). While the existence of disagreement among medical experts has not changed, there has been one new study on the safety of the banned procedures. A recent study by Dr.

Stephen Chasen addressed the comparative health effects of the D & X and D & E procedures. Stephen T. Chasen et al., Dilation and evacuation at > 20 weeks; Comparison of operative techniques, 190 Am. J. of Obstetrics and Gynecology 1180 (2004). The study found no significant difference in blood loss, procedure time, or short-term complication rates between the procedures.

The government argues that these conclusions reinforce Congress’s finding that the banned procedures are not safer than other methods (while also conceding that the conclusions militate against Congress’s finding that the banned procedures have “serious” health risks). In drawing its conclusions, however, the government ignores the study’s methodology. The choice of procedure in each case was not random, but was rather “based on cervical dilation and fetal position.” Id. at 1181.

Thus, the only real conclusion that can be drawn from this new study is that D & X is not inherently more dangerous than D & E in situations where the medical professional believes D & X to be the most appropriate procedure. No general conclusion regarding the medical necessity of the banned procedures in any given situation can be drawn from the study, which neither conclusively supports the position that the banned procedures are sometimes medically necessary, nor does it conclusively support the position that they are never medically necessary.

The Chasen study therefore detracts in no way from the Supreme Court’s prior conclusion, as there are still no medical studies addressing the medical necessity of the banned procedures. We need not belabor the point. The record in this case and the record in Sten-berg are similar in all significant respects. See Nat’l Abortion Fed’n, 330 F.Supp.2d at 492 (explaining that the government’s arguments “all fail to meaningfully distinguish the evidentiary circumstances present here from those that Stenberg held required a health exception to a ban on partial-birth abortion”).

There remains no consensus in the medical community as to the safety and medical necessity of the banned procedures. There is a dearth of studies on the medical necessity of the banned procedures.

In the absence of new evidence which would serve to distinguish this record from the record reviewed by the Supreme Court in Stenberg, we are bound by the Supreme Court’s conclusion that “substantial medical authority” supports the medical necessity of a health exception. “As a court of law, [our responsibility] is neither to devise ways in which to circumvent the opinion of the Supreme Court nor to indulge delay in the full implementation of the Court’s opinions.

Rather, our responsibility is to faithfully follow its opinions, because that court is, by constitutional design, vested with the ultimate authority to interpret the Constitution.” Richmond Med. Ctr. for Women v. Gilmore, 219 F.3d 376, 378 (4th Cir.2000) (Luttig, J., concurring). Because the Act does not contain a health exception exception, it is unconstitutional. We therefore do not reach the district court’s conclusion of the Act imposing an undue burden on a woman’s right to have an abortion. VI For the reasons stated above, the judgment of the district court is affirmed. . The Honorable Richard G.

Kopf, Chief Judge, United States District Court for the District of Nebraska. . Amici have argued Stenberg does not apply for several reasons. To the extent their arguments suggest we disregard or overrule Supreme Court precedent, such a course of action is beyond our power. One amicus suggests Stenberg does not control because that case was decided under the Fourteenth Amendment, which, of course, does not apply to the federal government.

While Sten-berg was indeed a Fourteenth Amendment case, the Due Process Clause of the Fifth Amendment is textually identical to the Due Process Clause of the Fourteenth Amendment, and both proscribe virtually identical governmental conduct. See, e.g., Malloy v. Hogan, 378 U.S. 1, 8, 84 S.Ct. 1489, 12 L.Ed.2d 653 (1964). If anything, the Fifth Amendment’s Due Process Clause has a broader reach in that it has been interpreted to apply the principles of 'the Fourteenth Amendment's Equal Protection Clause to the federal government. See, e.g., Adarand Constructors, Inc. v.

Pena, 515 U.S. 200, 253 n. 8, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995); Bolling v. Sharpe, 347 U.S. 497, 74 S.Ct. 693, 98 L.Ed. 884 (1954). . Of course, this may not be true of all abortion-related restrictions. . Though the government argues at length that substantial evidence supports Congress’s conclusion, it at no point engages the analysis undertaken by all three district courts to have addressed the constitutionality of the Act and one of the major points raised by the Appel-lees: that Congress’s conclusion that a consensus has formed against the medical necessity of the procedures was unreasonable.

The government has argued the district court adopted an erroneous reading of Stenberg by focusing on "substantial medical authority” and a lack of consensus against the procedures. Despite the fact that every federal court to have addressed the issue has rejected the government’s position, the government never challenges the district court’s conclusion that "substantial medical authority" supports the medical necessity of the banned procedures. By virtue of the government’s failure to argue the issue in either its opening brief or in its reply, we could consider the issue waived. See, e.g., Chay-Velasquez v.

Ashcroft, 367 F.3d 751, 756 (8th Cir.2004) (failure to raise issue in opening brief constitutes waiver). However, we decline to do so and will address the issue nonetheless. . The government argues the district court erred for various reasons in discounting the testimony of experts. We need not address this issue because giving full value to the government’s witnesses would in no way alter our conclusion that no consensus has been reached by the medical community. . The variations in long-term health effects noted in the study were not statistically significant and we therefore will not address them.

See Br. of Appellant at 43 (study cannot support "meaningful conclusions" about long-term complication rates due to small sample size).