United States v. Skrmetti, 605 U.S. ___ (2025)

NOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, pio@supremecourt.gov, of any typographical or other formal errors.

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Chief Justice Roberts delivered the opinion of the Court.

In this case, we consider whether a Tennessee law banning certain medical care for transgender minors violates the Equal Protection Clause of the Fourteenth Amendment.

I

A

An estimated 1.6 million Americans over the age of 13 identify as transgender, meaning that their gender identity does not align with their biological sex. See 1 App. 257–259; 2 id., at 827. Some transgender individuals suffer from gender dysphoria, a medical condition characterized by persistent, clinically significant distress resulting from an incongruence between gender identity and biological sex. Left untreated, gender dysphoria may result in severe physical and psychological harms.

In 1979, the World Professional Association for Transgender Health (WPATH) (then known as the Harry Benjamin International Gender Dysphoria Association) published one of the first sets of clinical guidelines for treating gender dysphoria with sex transition treatments. See P. Walker et al., Standards of Care: The Hormonal and Surgical Sex Reassignment of Gender Dysphoric Persons (1979), reprinted in 14 Archives of Sexual Behavior 79 (1985). The standards addressed two treatments in particular: hormonal sex reassignment (the use of hormones to induce the development of physical characteristics of the opposite sex) and surgical sex reassignment (surgery of the genitalia and/or chest to approximate the physical appearance of the opposite sex). See id., at 81, §§3.2–3.3. They recognized the extensive and sometimes irreversible consequences of hormonal therapy and sex reassignment surgery and acknowledged that some individuals who undergo reassignment procedures later regret their decision to do so. See id., at 83, 85–86, §§4.1.1–4.1.3, 4.4.2–4.4.3, 4.5.1. Among other things, the standards of care provided that hormonal and surgical sex reassignment treatments should be administered only to adults. See id., at 89, §4.14.4.

In 1998, WPATH revised its standards of care to permit healthcare professionals to administer puberty blockers (designed to delay the development of physical sex characteristics) and hormones to minors in “rar[e]” circumstances. S. Levine et al., The Standards of Care for Gender Identity Disorders (5th ed. 1998), reprinted in 11 J. Psychology & Human Sexuality 1, 20 (1999). Today, the standards discuss a range of factors regarding the provision of such treatments to minors. E. Coleman et al., Standards of Care for the Health of Transgender and Gender Diverse People, Version 8, 23 Int’l J. Transgender Health S1, S65–S66 (2022). The current standards recognize known risks associated with the provision of sex transition treatments to adolescents, including potential adverse effects on fertility and the possibility that an adolescent will later wish to detransition. See id., at S47, S57, S61–S62. They further state that there is “limited data on the optimal timing” of sex transition treatments or “the long-term physical, psychological, and neurodevelopmental outcomes in youth,” id., at S65, and note that “[o]ur understanding of gender identity development in adolescence is continuing to evolve,” id., at S44.

In recent years, the number of minors requesting sex transition treatments has increased. See 2 App. 644, 827–828. This increase has corresponded with rising debates regarding the relative risks and benefits of such treatments. Compare, e.g., Brief for State of California et al. as Amici Curiae 1–13, with Brief for Alabama as Amicus Curiae 1–9. In the last three years, more than 20 States have enacted laws banning the provision of sex transition treatments to minors, while two have enacted near total bans.

Meanwhile, health authorities in a number of European countries have raised significant concerns regarding the potential harms associated with using puberty blockers and hormones to treat transgender minors. In 2020, Finland’s Council for Choices in Health Care found that “gender reassignment of minors is an experimental practice” and that “the reliability of the existing studies” is “highly uncertain.” 2 App. 583–584 (alterations omitted); see id., at 715–722, 727–729. That same year, England’s National Institute for Health and Care Excellence published reports finding that the evidence for using puberty blockers to treat transgender adolescents is of “very low certainty” and that the long-term risks associated with using hormones to treat adolescents with gender dysphoria are “largely unknown.” Id., at 588–589. In 2022, Sweden’s National Board of Health and Welfare found that “the evidence on treatment efficacy and safety is still insufficient and inconclusive” and that the “risks” of puberty blockers and hormones “currently outweigh the possible benefits.” 1 id., at 339–340; see 2 id., at 584–587. And in 2023, the Norwegian Healthcare Investigation Board concluded that the “research-based knowledge” for hormonal sex transition treatments for minors is “insufficient,” while the “long-term effects are little known.” 1 id., at 341–342.

B

In March 2023, Tennessee joined the growing number of States restricting sex transition treatments for minors by enacting the Prohibition on Medical Procedures Performed on Minors Related to Sexual Identity, S. B. 1, 113th Gen. Assem., 1st Extra. Sess.; Tenn. Code Ann. §68–33–101 et seq. (SB1). While the State’s legislature acknowledged that discordance between a minor’s gender identity and biological sex can cause “discomfort or distress,” §68–33–101(c), it identified concerns regarding the use of puberty blockers and hormones to treat gender dysphoria in minors. In particular, the legislature found that such treatments “can lead to the minor becoming irreversibly sterile, having increased risk of disease and illness, or suffering from adverse and sometimes fatal psychological consequences,” §68–33–101(b), and that minors “lack the maturity to fully understand and appreciate” these consequences and may later regret undergoing the treatments, §68–33–101(h). The legislature further found that sex transition treatments were “being performed on and administered to minors in th[e] state with rapidly increasing frequency,” §68–33–101(g), notwithstanding the fact that the full range of harmful effects associated with the treatments were likely not yet known, see §68–33–101(b). The legislature also noted that guidelines regarding sex transition treatments for minors had “changed substantially in recent years,” §68–33–101(g), and that health authorities in Sweden, Finland, and the United Kingdom had “placed severe restrictions” on such treatments after determining that there was “no evidence” that their benefits outweigh their risks, §68–33–101(e); see supra, at 3. Finally, the legislature determined that there is evidence that gender dysphoria “can be resolved by less invasive approaches that are likely to result in better outcomes.” §68–33–101(c).

SB1 responds to these concerns by banning the use of certain medical procedures for treating transgender minors. In particular, the law prohibits a healthcare provider from “[s]urgically removing, modifying, altering, or entering into tissues, cavities, or organs of a human being,” or “[p]rescribing, administering, or dispensing any puberty blocker or hormone,” §68–33–102(5), for the purpose of (1) “[e]nabling a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex,” or (2) “[t]reating purported discomfort or distress from a discordance between the minor’s sex and asserted identity,” §68–33–103(a)(1). Among other things, these prohibitions are intended to “protec[t] minors from physical and emotional harm” by “encouraging minors to appreciate,” rather than “become disdainful of,” their sex. §68–33–101(m).

SB1 is limited in two relevant ways. First, SB1 does not restrict the administration of puberty blockers or hormones to individuals 18 and over. §68–33–102(6). Second, SB1 does not ban fully the administration of such drugs to minors. A healthcare provider may administer puberty blockers or hormones to treat a minor’s congenital defect, precocious (or early) puberty, disease, or physical injury. §68–33–103(b)(1)(A). The law defines the term “[c]ongenital defect” to include an “abnormality present in a minor that is inconsistent with the normal development of a human being of the minor’s sex,” §68–33–102(1), but excludes from the definitions of “[c]ongenital defect” and “disease” “gender dysphoria, gender identity disorder, [and] gender incongruence,” §§68–33–102(1), 68–33–103(b)(2).

SB1 contains three primary enforcement mechanisms. The law authorizes Tennessee’s attorney general to bring against any person who knowingly violates SB1 an action “to enjoin further violations, to disgorge any profits received due to the medical procedure, and to recover a civil penalty of [$25,000] per violation.” §68–33–106(b). SB1 further permits the relevant state regulatory authorities to discipline healthcare providers who violate the law’s prohibitions. §68–33–107. Finally, SB1 creates a private right of action that enables an injured minor or nonconsenting parent of an injured minor to sue a healthcare provider for violating the law. §68–33–105.

C

Three transgender minors, their parents, and a doctor (plaintiffs) brought a pre-enforcement challenge to SB1. Among other things, the plaintiffs asserted that SB1 violates the Equal Protection Clause of the Fourteenth Amendment. They moved for a preliminary injunction preventing the law’s bans on sex transition treatments for minors from going into effect. The United States intervened under 42 U. S. C. §2000h–2, which authorizes the Federal Government to intervene in a private equal protection suit “if the Attorney General certifies that the case is of general public importance.” See Memorandum Opinion and Order in No. 23–cv–00376 (MD Tenn., May 16, 2023), ECF Doc. 108.

The District Court partially enjoined enforcement of SB1’s prohibitions. See L. W. v. Skrmetti, 679 F. Supp. 3d 668, 677 (MD Tenn. 2023). The court concluded that the plaintiffs lacked standing to challenge the law’s ban on sex transition surgery for minors. Id., at 681–682. But the court held, as relevant, that the United States and plaintiffs were likely to succeed on their equal protection challenge to the law’s prohibitions on puberty blockers and hormones. Id., at 682–712. The court found that transgender individuals constitute a quasi-suspect class, that SB1 discriminates on the basis of sex and transgender status, and that SB1 was unlikely to survive intermediate scrutiny. Id., at 686–687, 698, 712. Having concluded that SB1 was likely unconstitutional on its face, the District Court issued a statewide injunction enjoining enforcement of all provisions of SB1 except for the private right of action and the law’s ban on sex transition surgery. See id., at 680–681, 716–718. Tennessee appealed, and the Sixth Circuit stayed the preliminary injunction pending appeal. L. W. v. Skrmetti, 83 F. 4th 460, 469 (CA6 2023).

The Sixth Circuit reversed. As relevant, the Sixth Circuit held that the United States and plaintiffs were unlikely to succeed on the merits of their equal protection claim. See id., at 479–489. The court first found that SB1 does not classify on the basis of sex because the law “regulate[s] sex-transition treatments for all minors, regardless of sex,” by prohibiting all minors from “receiv[ing] puberty blockers or hormones or surgery in order to transition from one sex to another.” Id., at 480. The court next declined to recognize transgender individuals as a suspect class, finding that transgender individuals are neither politically powerless nor a discrete group defined by obvious, immutable, or distinguishing characteristics. Id., at 486–487. Finally, the court concluded that the United States and plaintiffs had failed to establish that animus toward transgender individuals as a class was the operative force behind SB1. See id., at 487–488. The Sixth Circuit held that SB1 was subject to and survived rational basis review, finding that Tennessee had offered “considerable evidence” regarding the risks associated with the banned medical treatments and the flaws in existing research. Id., at 489.

Judge White dissented. Judge White would have held that the United States and plaintiffs were likely to succeed on the merits of their equal protection claim. Id., at 498. In her view, SB1 triggered heightened scrutiny because it “facially discriminate[s] based on a minor’s sex as assigned at birth and on a minor’s failure to conform with societal expectations concerning that sex.” Ibid. Judge White would have held that Tennessee had failed to “show an exceeding[ly] persuasive justification or close means-ends fit” for the law’s sex-based classifications. Ibid.

We granted certiorari to decide whether SB1 violates the Equal Protection Clause of the Fourteenth Amendment.[1] 602 U. S. ___ (2024).

II

The Fourteenth Amendment’s command that no State shall “deny to any person within its jurisdiction the equal protection of the laws,” U. S. Const., Amdt. 14, §1, “must coexist with the practical necessity that most legislation classifies for one purpose or another, with resulting disadvantage to various groups or persons,” Romer v. Evans, 517 U.S. 620, 631 (1996). We have reconciled the principle of equal protection with the reality of legislative classification by holding that, “if a law neither burdens a fundamental right nor targets a suspect class, we will uphold the legislative classification so long as it bears a rational relation to some legitimate end.” Ibid. We generally afford such laws “wide latitude” under this rational basis review, acknowledging that “the Constitution presumes that even improvident decisions will eventually be rectified by the democratic processes.” Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 440 (1985).

Certain legislative classifications, however, prompt heightened review. For example, laws that classify on the basis of race, alienage, or national origin trigger strict scrutiny and will pass constitutional muster “only if they are suitably tailored to serve a compelling state interest.” Ibid. We have similarly held that sex-based classifications warrant heightened scrutiny. See United States v. Virginia, 518 U.S. 515, 533 (1996). While our precedent does not make sex a “proscribed classification,” ibid., we have explained that sex “generally provides no sensible ground for differential treatment,” Cleburne, 473 U. S., at 440, and that sex-based lines too often reflect stereotypes or overbroad generalizations about the differences between men and women, see Sessions v. Morales-Santana, 582 U.S. 47, 62 (2017). We accordingly subject laws containing sex-based classifications to intermediate scrutiny, under which the State must show that the “classification serves important governmental objectives and that the discriminatory means employed are substantially related to the achievement of those objectives.” Virginia, 518 U. S., at 533 (internal quotation marks omitted).

A

We are asked to decide whether SB1 is subject to heightened scrutiny under the Equal Protection Clause. We hold it is not. SB1 does not classify on any bases that warrant heightened review.

1

On its face, SB1 incorporates two classifications. First, SB1 classifies on the basis of age. Healthcare providers may administer certain medical treatments to individuals ages 18 and older but not to minors. Second, SB1 classifies on the basis of medical use. Healthcare providers may administer puberty blockers or hormones to minors to treat certain conditions but not to treat gender dysphoria, gender identity disorder, or gender incongruence. Classifications that turn on age or medical use are subject to only rational basis review. See Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 312–314 (1976) (per curiam) (rational basis review applies to age-based classification); Vacco v. Quill, 521 U.S. 793, 799–808 (1997) (state laws outlawing assisted suicide “neither infringe fundamental rights nor involve suspect classifications”).

The plaintiffs argue that SB1 warrants heightened scrutiny because it relies on sex-based classifications. See Brief for Respondents in Support of Petitioner 20–37. We disagree.

Neither of the above classifications turns on sex. Rather, SB1 prohibits healthcare providers from administering puberty blockers and hormones to minors for certain medical uses, regardless of a minor’s sex. Cf. Vacco, 521 U. S., at 800 (“On their faces, neither New York’s ban on assisting suicide nor its statutes permitting patients to refuse medical treatment treat anyone differently from anyone else or draw any distinctions between persons. Everyone, regardless of physical condition, is entitled, if competent, to refuse unwanted lifesaving medical treatment; no one is permitted to assist a suicide.”).

The plaintiffs resist this conclusion, arguing that SB1 creates facial sex-based classifications by defining the prohibited medical care based on the patient’s sex. See Brief for Respondents in Support of Petitioner 22. This argument takes two forms. At times, the plaintiffs suggest that SB1 classifies on the basis of sex because its prohibitions reference sex. Alternatively, the plaintiffs contend that SB1 works a sex-based classification because application of the law turns on sex. Neither argument is persuasive.

This Court has never suggested that mere reference to sex is sufficient to trigger heightened scrutiny. See, e.g., Tuan Anh Nguyen v. INS, 533 U.S. 53, 64 (2001) (“The issue is not the use of gender specific terms instead of neutral ones. Just as neutral terms can mask discrimination that is unlawful, gender specific terms can mark a permissible distinction.”). Such an approach, moreover, would be especially inappropriate in the medical context. Some medical treatments and procedures are uniquely bound up in sex. The Food and Drug Administration itself recognizes that “[r]esearch has shown that biological differences between men and women (differences due to sex chromosome or sex hormones) may contribute to variations seen in the safety and efficacy of drugs, biologics, and medical devices.” FDA, Sex as a Biological Variable (Jan. 30, 2025) (online source archived at https://www.supremecourt.gov). Indeed, the agency frequently approves drugs for use by only one sex. See, e.g., FDA, FDA in Brief: FDA Encourages Inclusion of Male Patients in Breast Cancer Clinical Trials (Aug. 26, 2019) (online source archived at https://www.supremecourt .gov) (“many” breast cancer treatments approved for women only); FDA, FDA Approves Second Drug To Prevent HIV Infection as Part of Ongoing Efforts To End the HIV Epidemic (Oct. 3, 2019) (online source archived at https://www.supremecourt.gov) (drug to prevent HIV not approved for women). In the medical context, the mere use of sex-based language does not sweep a statute within the reach of heightened scrutiny.

We also reject the argument that the application of SB1 turns on sex. The plaintiffs and the dissent contend that an adolescent whose biological sex is female cannot receive puberty blockers or testosterone to live and present as a male, but an adolescent whose biological sex is male can, while an adolescent whose biological sex is male cannot receive puberty blockers or estrogen to live and present as a female, but an adolescent whose biological sex is female can. See Brief for Respondents in Support of Petitioner 22; post, at 10–15 (Sotomayor, J., dissenting). So conceived, they argue, SB1 prohibits certain treatments for minors of one sex while allowing those same treatments for minors of the opposite sex.

The plaintiffs and the dissent, however, contort the meaning of the term “medical treatment.” Notably absent from their framing is a key aspect of any medical treatment: the underlying medical concern the treatment is intended to address. The Food and Drug Administration approves drugs and requires that they be labeled for particular indications—the diseases or conditions that they treat, prevent, mitigate, diagnose, or cure. See 21 CFR §§201.57(c)(2), 314.50(a)(1) (2024). Different drugs can be used to treat the same thing (would you like Advil or Tylenol for your headache?), and the same drug can treat different things (take DayQuil to ease your cough, fever, sore throat, and/or minor aches and pains). For the term “medical treatment” to make sense of these various combinations, it must necessarily encompass both a given drug and the specific indication for which it is being administered. See Brief for Respondents in Support of Petitioner 5 (noting that “treatments” for adolescents with gender dysphoria include “puberty-delaying medication and hormone therapy” (emphasis added)).

When properly understood from the perspective of the indications that puberty blockers and hormones treat, SB1 clearly does not classify on the basis of sex. Both puberty blockers and hormones can be used to treat certain overlapping indications (such as gender dysphoria), and each can be used to treat a range of other conditions. Id., at 6–7. These combinations of drugs and indications give rise to various medical treatments. When, for example, a transgender boy (whose biological sex is female) takes puberty blockers to treat his gender incongruence, he receives a different medical treatment than a boy whose biological sex is male who takes puberty blockers to treat his precocious puberty.[2] SB1, in turn, restricts which of these medical treatments are available to minors: Under SB1, a healthcare provider may administer puberty blockers or hormones to any minor to treat a congenital defect, precocious puberty, disease, or physical injury, Tenn. Code Ann. §68–33–103(b)(1)(A); a healthcare provider may not administer puberty blockers or hormones to any minor to treat gender dysphoria, gender identity disorder, or gender incongruence, see §§68–33–102(1), 68–33–103(a)(1), (b)(2). The application of that prohibition does not turn on sex.

Of course, a State may not circumvent the Equal Protection Clause by writing in abstract terms. See Personnel Administrator of Mass. v. Feeney, 442 U.S. 256, 274 (1979) (explaining that both overt and covert sex-based classifications are subject to heightened review). The antimiscegenation law that this Court struck down in Loving v. Virginia, 388 U.S. 1 (1967), would not have shed its race-based classification had it, for example, prohibited “any person from marrying an individual of a different race.” Such a law would still have turned on a race-based classification: It would have prohibited Mildred Jeter (a black woman) from marrying Richard Loving (a white man), while permitting a white woman to do so. The law, in other words, would still “proscribe generally accepted conduct if engaged in by members of different races.” Id., at 11.

Here, however, SB1 does not mask sex-based classifications. For reasons we have explained, the law does not prohibit conduct for one sex that it permits for the other. Under SB1, no minor may be administered puberty blockers or hormones to treat gender dysphoria, gender identity disorder, or gender incongruence; minors of any sex may be administered puberty blockers or hormones for other purposes.

Nor are we persuaded that SB1’s prohibition on the prescription of puberty blockers and hormones to “[e]nabl[e] a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex” or to “[t]rea[t] purported discomfort or distress from a discordance between the minor’s sex and asserted identity,” Tenn. Code Ann. §68–33–103(a), reflects a sex-based classification, contra, post, at 10–15 (opinion of Sotomayor, J.). In the dissent’s view, this language “plainly classifies on the basis of sex” because it “turns on inconsistency with a protected characteristic.” Post, at 11. The dissent analogizes to a hypothetical law that “prohibit[s] minors from attending any services, rituals, or assemblies if done for the purpose of allowing the minor to identify with a purported identity inconsistent with the minor’s religion.” Ibid. (internal quotation marks omitted; emphasis deleted). Such a law, the dissent argues, would plainly classify on the basis of religion. “Whether the law prohibits a minor from attending any particular religious service turns on the minor’s religion: A Jewish child can visit a synagogue but not a church, while a Christian child can attend church but not the synagogue.” Ibid.

But a prohibition on the prescription of puberty blockers and hormones to “[e]nabl[e] a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex,” Tenn. Code Ann. §68–33–103(a)(1), is simply a prohibition on the prescription of puberty blockers and hormones to treat gender dysphoria, gender identity disorder, or gender incongruence. A law prohibiting attendance at a religious service “inconsistent with” the attendee’s religion may trigger heightened scrutiny. A law prohibiting the administration of specific drugs for particular medical uses does not. See Vacco, 521 U. S., at 799–808.

Finally, we reject the plaintiffs’ argument that, “by design, SB1 enforces a government preference that people conform to expectations about their sex.” Brief for Respondents in Support of Petitioner 23. The plaintiffs note that SB1’s statutory findings state that Tennessee has a compelling interest in “encouraging minors to appreciate their sex” and in prohibiting medical care “that might encourage minors to become disdainful of their sex.” Ibid. (quoting Tenn. Code Ann. §68–33–101(m)). They argue that these findings reveal that the law operates to force conformity with sex. See Brief for Respondents in Support of Petitioner 23; see also id., at 52 (“SB1’s purpose is . . . to force . . . boys and girls to look and live like boys and girls.” (internal quotation marks omitted)).

To start, the plaintiffs’ allegations of sex stereotyping are misplaced. True, a law that classifies on the basis of sex may fail heightened scrutiny if the classifications rest on impermissible stereotypes. See J. E. B. v. Alabama ex rel. T. B., 511 U.S. 127, 139, n. 11 (1994). But where a law’s classifications are neither covertly nor overtly based on sex, contrast, e.g., post, at 12–13, n. 8 (opinion of Sotomayor, J.) (referencing a hypothetical requirement that all children wear “sex-consistent clothing”), we do not subject the law to heightened review unless it was motivated by an invidious discriminatory purpose, see Personnel Administrator of Mass., 442 U. S., at 271–274; Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 264–266 (1977). No such argument has been raised here. See Tr. of Oral Arg. 57–59.

Regardless, the statutory findings to which the plaintiffs point do not themselves evince sex-based stereotyping. The plaintiffs fail to note that Tennessee also proclaimed a “legitimate, substantial, and compelling interest in protecting minors from physical and emotional harm.” Tenn. Code Ann. §68–33–101(m). And they similarly fail to acknowledge that Tennessee found that the prohibited medical treatments are experimental, can lead to later regret, and are associated with harmful—and sometimes irreversible—risks. §§68–33–101(b)–(e), (h). Tennessee’s stated interests in “encouraging minors to appreciate their sex” and in prohibiting medical care “that might encourage minors to become disdainful of their sex,” §68–33–101(m), simply reflect the State’s concerns regarding the use of puberty blockers and hormones to treat gender dysphoria, gender identity disorder, and gender incongruence, see Brief for Respondents 26–27 (“Given high desistance rates among youth and the tragic ‘regret’ of detransitioners, it was not improper to conclude that kids benefit from additional time to ‘appreciate their sex’ before embarking on body-altering paths. Nor is it improper for the State to protect minors from procedures that ‘encourage them to become disdainful of their sex’—and thus at risk for serious psychiatric conditions.” (citations and alterations omitted)); L. W., 83 F. 4th, at 485 (“A concern about potentially irreversible medical procedures for a child is not a form of stereotyping.”).

2

The plaintiffs separately argue that SB1 warrants heightened scrutiny because it discriminates against transgender individuals, who the plaintiffs assert constitute a quasi-suspect class. See Brief for Respondents in Support of Petitioner 37–38. This Court has not previously held that transgender individuals are a suspect or quasi-suspect class. And this case, in any event, does not raise that question because SB1 does not classify on the basis of transgender status. As we have explained, SB1 includes only two classifications: healthcare providers may not administer puberty blockers or hormones to minors (a classification based on age) to treat gender dysphoria, gender identity disorder, or gender incongruence (a classification based on medical use). The plaintiffs do not argue that the first classification turns on transgender status, and our case law forecloses any such argument as to the second.

We have explained that a State does not trigger heightened constitutional scrutiny by regulating a medical procedure that only one sex can undergo unless the regulation is a mere pretext for invidious sex discrimination. In Geduldig v. Aiello, 417 U.S. 484 (1974), for example, we held that a California insurance program that excluded from coverage certain disabilities resulting from pregnancy did not discriminate on the basis of sex. See id., at 486, 492–497. In reaching that holding, we explained that the program did not exclude any individual from benefit eligibility because of the individual’s sex but rather “remove[d] one physical condition—pregnancy—from the list of compensable disabilities.” Id., at 496, n. 20. We observed that the “lack of identity” between sex and the excluded pregnancy-related disabilities became “clear upon the most cursory analysis.” Id., at 497, n. 20. The California insurance program, we explained, divided potential recipients into two groups: “pregnant women and nonpregnant persons.” Ibid. Because women fell into both groups, the program did not discriminate against women as a class. See id., at 496, and n. 20. We thus concluded that, even though only biological women can become pregnant, not every legislative classification concerning pregnancy is a sex-based classification. Id., at 496, n. 20. As such, “[a]bsent a showing that distinctions involving pregnancy are mere pretexts designed to effect an invidious discrimination against the members of one sex or the other, lawmakers are constitutionally free to include or exclude pregnancy from the coverage of legislation . . . on any reasonable basis, just as with respect to any other physical condition.” Id., at 496–497, n. 20.

By the same token, SB1 does not exclude any individual from medical treatments on the basis of transgender status but rather removes one set of diagnoses—gender dysphoria, gender identity disorder, and gender incongruence—from the range of treatable conditions. SB1 divides minors into two groups: those who might seek puberty blockers or hormones to treat the excluded diagnoses, and those who might seek puberty blockers or hormones to treat other conditions. See Tenn. Code Ann. §68–33–103. Because only transgender individuals seek puberty blockers and hormones for the excluded diagnoses, the first group includes only transgender individuals; the second group, in contrast, encompasses both transgender and nontransgender individuals. Thus, although only transgender individuals seek treatment for gender dysphoria, gender identity disorder, and gender incongruence—just as only biological women can become pregnant—there is a “lack of identity” between transgender status and the excluded medical diagnoses. The plaintiffs, moreover, have not argued that SB1’s prohibitions are mere pretexts designed to effect an invidious discrimination against transgender individuals. Under these circumstances, we decline to find that SB1’s prohibitions on the use of puberty blockers and hormones exclude any individuals on the basis of transgender status.[3]

3

Finally, Bostock v. Clayton County, 590 U.S. 644 (2020), does not alter our analysis. In Bostock, we held that an employer who fires an employee for being gay or transgender violates Title VII’s prohibition on discharging an individual “because of ” their sex. See id., at 650–652, 654–659. We reasoned that Title VII’s “because of ” test incorporates the traditional but-for causation standard, which “directs us to change one thing at a time and see if the outcome changes. If it does, we have found a but-for cause.” Id., at 656. Applying that test, we held that, “[f]or an employer to discriminate against employees for being homosexual or transgender, the employer must intentionally discriminate against individual men and women in part because of sex.” Id., at 662. In such a case, the employer has penalized a member of one sex for a trait or action that it tolerates in members of the other. Ibid.

The plaintiffs urge us to apply Bostock’s reasoning to this case. In their view, SB1 violates the Equal Protection Clause because it prohibits a minor whose biological sex is female from receiving testosterone to live as a male but allows a minor whose biological sex is male to receive testosterone for the same purposes (and vice versa). Applying Bostock’s reasoning, they argue that SB1 discriminates on the basis of sex because it intentionally penalizes members of one sex for traits and actions that it tolerates in another. See Brief for Respondents in Support of Petitioner 24–25.

We have not yet considered whether Bostock’s reasoning reaches beyond the Title VII context, and we need not do so here. For reasons we have already explained, changing a minor’s sex or transgender status does not alter the application of SB1. If a transgender boy seeks testosterone to treat his gender dysphoria, SB1 prevents a healthcare provider from administering it to him. See Tenn. Code Ann. §68–33–103(a). If you change his biological sex from female to male, SB1 would still not permit him the hormones he seeks because he would lack a qualifying diagnosis for the testosterone—such as a congenital defect, precocious puberty, disease, or physical injury. The transgender boy could receive testosterone only if he had one of those permissible diagnoses. And, if he had such a diagnosis, he could obtain the testosterone regardless of his sex or transgender status. Under the reasoning of Bostock, neither his sex nor his transgender status is the but-for cause of his inability to obtain testosterone.

The dissent counters that, whatever causal factors are at play, sex is at least one but-for cause of SB1’s operation. See post, at 19–20 (opinion of Sotomayor, J.). To illustrate this argument, the dissent posits a minor girl with facial hair inconsistent with her sex. Under SB1, the dissent notes, a healthcare provider can prescribe puberty blockers or hormones to the minor to suppress her hair growth. Ibid. Change the minor’s sex to male, the dissent reasons, and SB1 prevents the minor from obtaining the same drugs for the same purpose. Ibid. Any corresponding change in diagnosis, the dissent concludes, simply reveals that both sex and diagnosis are causal factors at “ ‘play.’ ” Post, at 20 (quoting Bostock, 590 U. S., at 661).

The dissent’s reasoning overlooks a key distinction between the operation of SB1 and the logic of Bostock. Under Bostock’s reasoning, an employer who fires a homosexual male employee for being attracted to men while retaining the employee’s straight female colleague has discriminated on the basis of sex because it has penalized the male employee for a trait (attraction to men) that it tolerates in the female employee. See id., at 660. Bostock held that, in such a circumstance, sex is the but-for cause of the employer’s decision—change the homosexual male employee’s sex and he becomes a straight female whose attraction to men the employer tolerates.

Not so with SB1. Consider again the minor girl with unwanted facial hair inconsistent with her sex. If she has a diagnosis of hirsutism (male-pattern hair growth), a healthcare provider may, consistent with SB1, prescribe her puberty blockers or hormones. But changing the minor’s sex to male does not automatically change the operation of SB1. If hirsutism is replaced with gender dysphoria, the now-male minor may not receive puberty blockers or hormones; but if hirsutism is replaced with precocious puberty, SB1 does not bar either treatment. Unlike the homosexual male employee whose sexuality automatically switches to straight when his sex is changed from male to female, there is no reason why a female minor’s diagnosis of hirsutism automatically changes to gender dysphoria when her sex is changed from female to male. Under the logic of Bostock, then, sex is simply not a but-for cause of SB1’s operation.

B

The rational basis inquiry “employs a relatively relaxed standard reflecting the Court’s awareness that the drawing of lines that create distinctions is peculiarly a legislative task and an unavoidable one.” Massachusetts Bd. of Retirement, 427 U. S., at 314. Under this standard, we will uphold a statutory classification so long as there is “any reasonably conceivable state of facts that could provide a rational basis for the classification.” FCC v. Beach Communications, Inc., 508 U.S. 307, 313 (1993). Where there exist “plausible reasons” for the relevant government action, “our inquiry is at an end.” Id., at 313–314 (internal quotation marks omitted).

SB1 clearly meets this standard. Tennessee determined that administering puberty blockers or hormones to a minor to treat gender dysphoria, gender identity disorder, or gender incongruence “can lead to the minor becoming irreversibly sterile, having increased risk of disease and illness, or suffering from adverse and sometimes fatal psychological consequences.” Tenn. Code Ann. §68–33–101(b). It further found that it was “likely that not all harmful effects associated with these types of medical procedures when performed on a minor are yet fully known, as many of these procedures, when performed on a minor for such purposes, are experimental in nature and not supported by high- quality, long-term medical studies.” Ibid. Tennessee determined that “minors lack the maturity to fully understand and appreciate the life-altering consequences of such procedures and that many individuals have expressed regret for medical procedures that were performed on or administered to them for such purposes when they were minors.” §68–33–101(h). At the same time, Tennessee noted evidence that discordance between sex and gender “can be resolved by less invasive approaches that are likely to result in better outcomes for the minor.” §68–33–101(c). SB1’s age- and diagnosis-based classifications are plainly rationally related to these findings and the State’s objective of protecting minors’ health and welfare. §68–33–101(a).

The plaintiffs argue that SB1 fails even rational basis review because the law’s classifications are “so far removed from [Tennessee’s] asserted justifications that it is impossible to credit those interests.” Brief for Respondents in Support of Petitioner 51 (internal quotation marks and alterations omitted). In their view, Tennessee has failed to explain why it has banned access to puberty blockers and hormones “only where they would allow a transgender minor to ‘identify’ or ‘live’ in a way ‘inconsistent’ with their ‘sex.’ ” Id., at 52.

This argument fails. As we have explained, there is a rational basis for SB1’s classifications. Tennessee concluded that there is an ongoing debate among medical experts regarding the risks and benefits associated with administering puberty blockers and hormones to treat gender dysphoria, gender identity disorder, and gender incongruence. SB1’s ban on such treatments responds directly to that uncertainty. Contrast Cleburne, 473 U. S., at 448 (record did not reveal “any rational basis” for city zoning ordinance); Romer, 517 U. S., at 632 (“sheer breadth” of law was “so discontinuous with the reasons offered for it that the [law] seem[ed] inexplicable by anything but animus toward the class it affect[ed]”).

We also decline the plaintiffs’ invitation to second-guess the lines that SB1 draws. It may be true, as the plaintiffs contend, that puberty blockers and hormones carry comparable risks for minors no matter the purposes for which they are administered. But it may also be true, as Tennessee determined, that those drugs carry greater risks when administered to treat gender dysphoria, gender identity disorder, and gender incongruence. We afford States “wide discretion to pass legislation in areas where there is medical and scientific uncertainty.” Gonzales v. Carhart, 550 U.S. 124, 163 (2007). “[T]he fact the line might have been drawn differently at some points is a matter for legislative, rather than judicial, consideration.” Railroad Retirement Bd. v. Fritz, 449 U.S. 166, 179 (1980); see Dandridge v. Williams, 397 U.S. 471, 485 (1970) (“In the area of economics and social welfare, a State does not violate the Equal Protection Clause merely because the classifications made by its laws are imperfect.”); Lindsley v. Natural Carbonic Gas Co., 220 U.S. 61, 78 (1911) (“A classification having some reasonable basis does not offend against [the Equal Protection Clause] merely because it is not made with mathematical nicety or because in practice it results in some inequality.”).

Recent developments only underscore the need for legislative flexibility in this area. After Tennessee enacted SB1, a report commissioned by England’s National Health Service (NHS England) characterized the evidence concerning the use of puberty blockers and hormones to treat transgender minors as “remarkably weak,” concluding that there is “no good evidence on the long-term outcomes of interventions to manage gender-related distress.” H. Cass, Independent Review of Gender Identity Services for Children and Young People: Final Report 13 (Apr. 2024). The report cautioned that “results of studies are exaggerated or misrepresented by people on all sides of the debate to support their viewpoint,” ibid., and concluded that the “current understanding of the long-term health impacts of hormone interventions is limited and needs to be better understood,” id., at 22. In response to the report, NHS England enacted prohibitions on the administration of puberty blockers to new patients under the age of 18 outside of research settings and instituted a process for reviewing referrals for hormones for adolescents under the age of 16. See NHS England, Children and Young People’s Gender Services: Implementing the Cass Review Recommendations 6–7 (Aug. 2024); Tr. of Oral Arg. 14–18.

We cite this report and NHS England’s response not for guidance they might provide on the ultimate question of United States law, see Schriro v. Summerlin, 542 U.S. 348, 356 (2004) (contemporary foreign practice is “irrelevant” to constitutional interpretation), but to demonstrate the open questions regarding basic factual issues before medical authorities and other regulatory bodies. Such uncertainty “afford[s] little basis for judicial responses in absolute terms.” Marshall v. United States, 414 U.S. 417, 427 (1974). And “[t]he calculus of effects, the manner in which a particular law reverberates in a society, is a legislative and not a judicial responsibility.” Personnel Administrator of Mass., 442 U. S., at 272.

*  *  *

This case carries with it the weight of fierce scientific and policy debates about the safety, efficacy, and propriety of medical treatments in an evolving field. The voices in these debates raise sincere concerns; the implications for all are profound. The Equal Protection Clause does not resolve these disagreements. Nor does it afford us license to decide them as we see best. Our role is not “to judge the wisdom, fairness, or logic” of the law before us, Beach Communications, 508 U. S., at 313, but only to ensure that it does not violate the equal protection guarantee of the Fourteenth Amendment. Having concluded it does not, we leave questions regarding its policy to the people, their elected representatives, and the democratic process.

The judgment of the United States Court of Appeals for the Sixth Circuit is affirmed.

It is so ordered.

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Justice Alito, concurring in part and concurring in the judgment.

I concur in the judgment and join Parts I and II–B of the opinion of the Court. I agree with much of the discussion in Part II–A–1, which holds that Tennessee’s Senate Bill 1 (SB1) does not classify on the basis of “sex,” but I set out my own analysis of this issue in Part I of this opinion. I do not join Part II–A–2 of the opinion of the Court, which concludes that SB1 does not classify on the basis of “transgender status.” There is a strong argument that SB1 does classify on that ground, but I find it unnecessary to decide that question. I would assume for the sake of argument that the law classifies based on transgender status, but I would nevertheless sustain the law because such a classification does not warrant heightened scrutiny. I also do not join Part II–A–3 of the Court’s opinion because I do not believe that the reasoning employed in Bostock v. Clayton County, 590 U.S. 644 (2020), is applicable when determining whether a law classifies based on sex for Equal Protection Clause purposes.

I

A

To begin, I agree with the Court that SB1 does not classify on the basis of “sex” within the meaning of our equal protection precedents. What those cases have always meant by “sex” is the status of having the genes of a male or female. That was the common understanding of the term in 1971 when the Court, in Reed v. Reed, 404 U.S. 71, 74, first held that a law that discriminated against women violated the Equal Protection Clause. See, e.g., Random House Dictionary of the English Language 1307 (1966) (defining “sex” as “the fact or character of being either male or female”); Webster’s Third New International Dictionary 2081 (1966) (defining “sex” as “one of the two divisions of . . . human beings respectively designated male or female”). And all the Court’s subsequent cases in this line have shared that understanding.

In Frontiero v. Richardson, 411 U.S. 677 (1973), which was handed down in the next Term after Reed, a plurality referred to “sex” as “an immutable characteristic determined solely by the accident of birth.” 411 U. S., at 686. Twenty-five years later, Justice Ginsburg’s landmark opinion for the Court in United States v. Virginia, 518 U.S. 515 (1996) (VMI), exhibited the same understanding. The opinion observed that the “[p]hysical differences between men and women . . . are enduring” and that the “ ‘[i]nherent differences’ between men and women” are “cause for celebration.” Id., at 533.

While the earliest cases in this line referred solely to discrimination on the basis of “sex,” see, e.g., Reed, 404 U. S., at 75–77; Frontiero, 411 U. S., at 682–688 (plurality opinion), later equal protection cases referred to classifications based on “gender,” see Craig v. Boren, 429 U.S. 190, 192 (1976). But it is clear that these cases used “gender” as a synonym for “sex.” See, e.g., id., at 199 (using “sex” and “gender” interchangeably). In employing the term “gender” in this way, our opinions tracked a change in usage in ordinary speech. As the Oxford English Dictionary explains, “as sex came increasingly to mean sexual intercourse . . . , gender began to replace it . . . as the usual word for the biological grouping of males and females.” Oxford English Dictionary (3d ed., June 2011), https://doi.org/10.1093/ OED/8610510183. Thus, our use of the term “gender” had no substantive significance. None of our equal protection decisions has used “gender” in the sense in which it is now sometimes used, i.e., to denote “a group of people in a society who share particular qualities or ways of behaving which that society associates with being male, female, or another identity.”[1]

For these reasons a party claiming that a law violates the Equal Protection Clause because it classifies on the basis of sex cannot prevail simply by showing that the law draws a distinction on the basis of “gender identity.” See, e.g., Merriam-Webster’s Collegiate Dictionary 520 (11th ed. 2020) (defining “gender identity”). Rather, such a plaintiff must show that the challenged law differentiates between the two biological sexes: male and female.

B

1

What, then, does it mean for a law to “classify” based on sex? The succinct answer is that a law classifies based on sex for equal protection purposes when it “[p]rescrib[es] one rule for [women], [and] another for [men].” Sessions v. Morales-Santana, 582 U.S. 47, 58 (2017). And as we have explained, the general rule is that a law meets this test if it employs an “overt gender criterion.” Craig, 429 U. S., at 198.

A few examples illustrate the point. A law setting one drinking age for women and another for men is a sex classification. Id., at 191–192, 197–199. A college policy granting admission to women but not to men (or vice versa) is a sex classification. Mississippi Univ. for Women v. Hogan, 458 U.S. 718, 720–723 (1982); VMI, 518 U. S., at 530–531. A law imposing different citizenship requirements for children with citizen fathers compared to children with citizen mothers is a sex classification. Tuan Anh Nguyen v. INS, 533 U.S. 53, 59–62 (2001).

What is apparent in each of these cases is that sex serves as an explicit “criterion,” dictating that a particular legal standard applies to one sex but not the other. See also Weinberger v. Wiesenfeld, 420 U.S. 636 (1975) (different rules for husbands and wives); Stanton v. Stanton, 421 U.S. 7 (1975) (different rule for men and women); Califano v. Goldfarb, 430 U.S. 199 (1977) (different rules for widows and widowers); Califano v. Webster, 430 U.S. 313 (1977) (per curiam) (different rule for men and women); Orr v. Orr, 440 U.S. 268 (1979) (different rule for husbands and wives); Caban v. Mohammed, 441 U.S. 380 (1979) (different rule for unwed mothers and unwed fathers); Wengler v. Druggists Mut. Ins. Co., 446 U.S. 142 (1980) (different rules for widows and widowers); Kirchberg v. Feenstra, 450 U.S. 455 (1981) (different rule for husbands and wives); Morales-Santana, 582 U.S. 47 (different rules for unwed mothers and unwed fathers).

In contrast to what our cases have demanded, we have “never suggested that mere reference to sex is sufficient to trigger heightened scrutiny.” Ante, at 10 (citing Nguyen, 533 U. S., at 64); see also Dobbs v. Jackson Women’s Health Organization, 597 U.S. 215, 236–237 (2022) (holding that rational basis review applied to a prohibition on abortion, despite the fact that the law in question mentioned “the physical health of the mother”).

We have also explicitly rejected the proposition that a law classifies based on sex when it employs a non-sex classification that correlates with differential treatment of men and women. In Geduldig v. Aiello, 417 U.S. 484 (1974), for example, we considered a California insurance program that “exclude[d] from coverage certain disabilities resulting from pregnancy.” Id., at 486. Although we recognized that “only women can become pregnant,” we explained that “it does not follow that every legislative classification concerning pregnancy is a sex-based classification.” Id., at 496, n. 20. In the absence of a showing that the pregnancy classification at issue was being used as a “mere pretex[t] designed to effect an invidious discrimination against the members of one sex or the other,” we were unwilling to conclude that it was a proxy for a sex classification. Id., at 496–497, n. 20.

We applied a similar principle in Personnel Administrator of Mass. v. Feeney, 442 U.S. 256 (1979). There, we considered a Massachusetts policy that conferred an “absolute advantage” on veterans who applied for state civil service positions. Id., at 264. At the time of the lawsuit, “over 98% of the veterans in Massachusetts were male,” and we acknowledged that “[t]he impact of the veterans’ preference law upon the public employment opportunities of women has thus been severe.” Id., at 270–271. Even so, such “severe” disparate impact did not make the law a sex classification. The distinction made by the law was “quite simply between veterans and nonveterans, not between men and women.” Id., at 275. And such a classification was not a sex classification unless it could be “shown that a gender-based discriminatory purpose has, at least in some measure, shaped the Massachusetts veterans’ preference legislation.” Id., at 276.

The upshot of all these prior equal protection cases is that we will generally not find that a law classifies on the basis of sex unless it does so overtly, but that a challenger may escape this general rule by showing that a purportedly sex-neutral classification has been used as a “mere pretex[t] designed to effect an invidious discrimination against the members of one sex or the other.” Geduldig, 417 U. S., at 496–497, n. 20.[2]

2

When these principles are applied to Tennessee’s SB1, it is clear that the law is not a sex classification. As the Court notes, SB1 classifies based on the purpose for which a minor seeks the covered medical treatments. Specifically, it restricts those treatments if they are sought either for the purpose of “[e]nabling a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex” or for the purpose of “[t]reating purported discomfort or distress from a discordance between the minor’s sex and asserted identity.” Tenn. Code Ann. §§68–33–103(a)(1)(A), (a)(1)(B) (2023). This scheme certainly refers to sex and may be seen as indirectly related to sex, but it is clearly not the sort of discrimination between males and females that our cases have treated as sex discrimination. It does not lay down one rule for males and another for females. Instead, it classifies based on something quite different: a minor’s reason for seeking particular treatment.

This classification scheme is also not a “mere pretex[t] designed to effect an invidious discrimination against the members of one sex or the other.” Geduldig, 417 U. S., at 496–497, n. 20. The law begins with a panoply of legislative findings that make clear that the legislature’s purpose was to “protect the health and welfare of minors.” §68–33–101(a). The legislature concluded that the prohibited medical procedures were “experimental in nature and not supported by high-quality, long-term medical studies,” and that often “a minor’s discordance can be resolved by less invasive approaches that are likely to result in better outcomes.” §§68–33–101(b), (c).

These findings are consistent with those made by other respected bodies that cannot be charged with hostility to minors experiencing gender dysphoria or to transgender people in general. See ante, at 3–4. And the limited scope of SB1 strongly supports the conclusion that the legislature’s true purpose was exactly the one set out in the statutory findings. SB1 targets only the experimental medical procedures that the legislature found to be unsupported and dangerous. It does not regulate any other behavior in which minors might engage for the purpose of expressing their gender identity. It says nothing at all about names, pronouns, hair styles, attire, recreational activities or hobbies, or career interests. And the law’s restrictions apply only to the treatment available to minors. Once individuals reach the age at which they are able to make informed decisions about medical care, the law imposes no restrictions.

3

In an effort to show that SB1 classifies based on sex, the plaintiffs, the dissent, and some of the plaintiffs’ amici rely on what they understand to be the Court’s reasoning in Bostock, 590 U.S. 644. See Brief for Respondents in Support of Petitioner 24–32; post, at 14–15 (Sotomayor, J., dissenting); Brief for State of California et al. as Amici Curiae 14–16; Brief for Kentucky Plaintiffs et al. as Amici Curiae 10–16. This argument is misguided. The decision in Bostock was based on the conclusion that the specific language employed in Title VII of the Civil Rights Act of 1964 prohibits an adverse employment action if sex is a “but-for cause” of that action. 590 U. S., at 656–660. And in fleshing out what this means, the Court engaged in a controversial form of counterfactual reasoning.[3] I dissented in Bostock, but I accept the decision as a precedent that is entitled to the staunch protection we give statutory interpretation decisions. See Kimble v. Marvel Entertainment, LLC, 576 U.S. 446, 456 (2015) (citing Patterson v. McLean Credit Union, 491 U.S. 164, 172–173 (1989)). But there is no reason to apply Bostock’s methodology here.

The Equal Protection Clause does not contain the same wording as Title VII, and our cases have never held that Bostock’s methodology applies in cases in which a law is challenged as an unconstitutional sex classification. On the contrary, as I have explained, our cases have adopted an entirely different methodology. I would follow those precedents.

II

My main point of disagreement with the Court concerns its analysis of the plaintiffs’ argument that SB1 unconstitutionally discriminates on the basis of transgender status. See Brief for Respondents in Support of Petitioner 37–38. The Court holds that the law does not classify on this ground, and the Court therefore applies rational basis review. Ante, at 16–18. I am uneasy with that analysis and would reject the plaintiffs’ argument for a different reason: because neither transgender status nor gender identity should be treated as a suspect or “quasi-suspect” class.

A

I will not dwell on the question whether SB1 classifies on the basis of transgender status or gender identity because, in the end, I do not think that the answer to that question has any effect on the outcome of this case. But the argument that SB1 classifies on those grounds cannot easily be dismissed. As noted, the law prohibits medical procedures that are intended either to “[e]nabl[e] a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex,” or to “[t]rea[t] purported discomfort or distress from a discordance between the minor’s sex and asserted identity.” Tenn. Code Ann. §§68–33–103(a)(1)(A), (a)(1)(B). Therefore, the underlying basis for the classification is a minor’s intent to express a gender identity different from the minor’s biological sex. If being “transgender” is defined as “hav[ing] a gender identity that differs from . . . sex,” see Brief for Respondents in Support of Petitioner 4, then the intent to “identify with, or live as, a purported identity inconsistent with” one’s sex would appear to be the natural result or consequence of being transgender.

The Court nonetheless concludes that SB1 does not classify based on transgender status, and in doing so, it relies chiefly on our decision in Geduldig, 417 U.S. 484. Ante, at 16–17. The dissent responds by denigrating Geduldig and contending that the decision should be discarded.[4] Post, at 23–24 (opinion of Sotomayor, J.).

I would not enter into this debate about SB1’s classification scheme. I would assume for the sake of argument that SB1 classifies on the basis of transgender status and move on to the question whether such a classification is either suspect or “quasi-suspect” and thus warrants some form of heightened scrutiny. That important question has divided the Courts of Appeals,[5] and if we do not confront it now, we will almost certainly be required to do so very soon.

B

In my view, transgender status does not qualify under our precedents as a suspect or “quasi-suspect” class.[6] We have never set out a hard-and-fast test that can be used to identify such classes, but, as I explain in more detail below, our decisions have identified certain key factors that transgender individuals do not share with members of suspect and “quasi-suspect” classes. Transgender status is not “immutable,” and as a result, persons can and do move into and out of the class. Members of the class differ widely among themselves, and it is often difficult for others to determine whether a person is a member of the class. And transgender individuals have not been subjected to a history of discrimination that is comparable to past discrimination against the groups we have classified as suspect or “quasi-suspect.”

1

In order to understand why transgender status should not be treated as either a suspect or “quasi-suspect” class, it is helpful to recall the path that led the Court to identify those groups and afford their members heightened protection. As the Court notes, ante, at 8, laws routinely confer benefits or impose burdens on particular classes of individuals, and we have long held that equal protection principles permit such classifications so long as they “bea[r] some fair relationship to a legitimate public purpose,” Plyler v. Doe, 457 U.S. 202, 216 (1982).

We first developed that standard during the New Deal era, when the Court was frequently called upon to decide whether economic legislation was consistent with the Constitution. In response to those challenges, the Court adopted the principle that “regulatory legislation affecting ordinary commercial transactions is not to be pronounced unconstitutional unless . . . it is of such a character as to preclude the assumption that it rests upon some rational basis within the knowledge and experience of the legislators.” United States v. Carolene Products Co., 304 U.S. 144, 152 (1938).

At the same time that the Court developed this “rational basis” standard, however, it suggested that some laws should be afforded a “narrower” presumption of constitutionality and should therefore receive “more exacting judicial scrutiny.” Ibid., n. 4. The Court opined that a different standard of review might apply to legislation “directed at particular religious, or national, or racial minorities.” Id., at 153, n. 4 (citations omitted). It reasoned that a “more searching judicial inquiry” might be required for such legislation because “prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities.” Ibid.

Consistent with that discussion, the Court soon held that “[c]lassifications based solely upon race must be scrutinized with particular care, since they are contrary to our traditions and hence constitutionally suspect.” Bolling v. Sharpe, 347 U.S. 497, 499 (1954). Such classifications, the Court later noted, “must be viewed in light of the historical fact that the central purpose of the Fourteenth Amendment was to eliminate racial discrimination emanating from official sources in the States.” McLaughlin v. Florida, 379 U.S. 184, 192 (1964).

The discrimination that the Court had chiefly in mind was discrimination against blacks, who undoubtedly constituted a “discrete and insular minorit[y]” that was denied equal participation in the political process. Carolene Products, 304 U. S., at 153, n. 4. As our cases from the period plainly illustrate, blacks faced widespread discrimination not only in fact but also in law. State and local authorities enforced a regime of official segregation in transportation, see Plessy v. Ferguson, 163 U.S. 537, 540 (1896), schools, see Brown v. Board of Education, 347 U.S. 483, 487–488 (1954), and all manner of public accommodations, see Watson v. Memphis, 373 U.S. 526, 528 (1963) (concerning the segregation of “municipal parks and other city owned or operated recreational facilities”).

Blacks were also widely impeded from participation in the political process. For example, several States enacted “literacy tests for voter registration” that were “designed to prevent African-Americans from voting.” Shelby County v. Holder, 570 U.S. 529, 536 (2013) (citing South Carolina v. Katzenbach, 383 U.S. 301, 310 (1966)). States also devised methods for excluding or impeding black citizens from serving in public office. See, e.g., Nixon v. Herndon, 273 U.S. 536, 541 (1927) (holding unconstitutional a law that excluded black citizens from “tak[ing] part in a primary election”); Anderson v. Martin, 375 U.S. 399, 400 (1964) (holding unconstitutional a law that required ballots to “designate the race of candidates for elective office”).

Given this history of pervasive discrimination and the fact that “the central purpose of the Fourteenth Amendment was to eliminate racial discrimination,” the Court concluded that racial classifications are “constitutionally suspect, and subject to the most rigid scrutiny.” McLaughlin, 379 U. S., at 192 (citation and internal quotation marks omitted). And at around the same time, the Court also treated national origin and ancestry as suspect classes, largely because of their proximal relationship to race. See, e.g., Oyama v. California, 332 U.S. 633, 646 (1948); Korematsu v. United States, 323 U.S. 214, 216 (1944), overruled by Trump v. Hawaii, 585 U.S. 677 (2018).[7]

The Court has also suggested that religion is a suspect class. See Carolene Products, 304 U. S., at 152, n. 4. That determination follows from the First Amendment, which prohibits any impairment of the “free exercise” of “religion.” But because this right is expressly protected by that provision, questions of religious discrimination have generally been decided on First Amendment grounds. See, e.g., Fulton v. Philadelphia, 593 U.S. 522, 532 (2021); Espinoza v. Montana Dept. of Revenue, 591 U.S. 464, 473–474 (2020); Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Comm’n, 584 U.S. 617, 638 (2018).

With this history in mind, it is apparent that the circumstances that led to the identification of race and national origin as suspect classes were truly extraordinary. As the Court subsequently explained, the designation of a suspect class is reserved for those classes “saddled with such disabilities, or subjected to such a history of purposeful unequal treatment, or relegated to such a position of political powerlessness as to command extraordinary protection from the majoritarian political process.” San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 28 (1973). And entitlement to “suspect class” status is largely reserved for those groups whose members tend to “carry an obvious badge” of their membership in the suspect class, which in part explains “the severity or pervasiveness of the historic legal and political discrimination against” the group. Mathews v. Lucas, 427 U.S. 495, 506 (1976). Suspect class status is therefore generally inappropriate for “large, diverse, and amorphous” groups, Rodriguez, 411 U. S., at 28, that do not share “obvious, immutable, or distinguishing characteristics that define them as a discrete group,” Lyng v. Castillo, 477 U.S. 635, 638 (1986). See also Mathews, 427 U. S., at 506; Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 442–443, 445 (1985).

No one can doubt that race satisfies all these criteria. Racial minorities experienced a long history of invidious discrimination and lack of political power. Race, as that concept was long understood in this society, is an immutable characteristic that often coincides with a visible and distinguishable “badge” of membership in the group. Mathews, 427 U. S., at 506. And both our Constitution and our “traditions” provide that discrimination based on race is proscribed in all but the narrowest circumstances. Bolling, 347 U. S., at 499. We have therefore viewed, and continue to view, racial classifications as “inherently suspect.” Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, 600 U.S. 181, 209 (2023) (internal quotation marks omitted). And since Brown v. Board of Education, 347 U.S. 483, we have struck down nearly every race- or national-origin-based classification that has come before us; our now-overruled affirmative action decisions were the exception to the rule. Students for Fair Admissions, 600 U. S., at 211–214, 224–225 (overruling Regents of Univ. of Cal. v. Bakke, 438 U.S. 265 (1978), and Grutter v. Bollinger, 539 U.S. 306 (2003)); see also 600 U. S., at 287 (Thomas, J., concurring) (“The Court’s opinion rightly makes clear that Grutter is, for all intents and purposes, overruled”).

2

This Court has never “equat[ed]” classifications based on sex with classifications based on race or national origin for Equal Protection Clause purposes, VMI, 518 U. S., at 532, and thus has never held that sex-based classifications are “suspect.” But since the 1970s, the Court has recognized that such classifications warrant more careful inspection than is provided by ordinary “rational basis” review. See ibid.; Craig, 429 U. S., at 198. We often refer to this as “heightened scrutiny” (or “intermediate scrutiny”), and we have used the term “quasi-suspect” to describe groups that qualify for this form of heightened review. See, e.g., Cleburne, 473 U. S., at 442. Under heightened or intermediate scrutiny, it must be shown that a sex-based classification “serves important governmental objectives and that the discriminatory means employed are substantially related to the achievement of those objectives.” Morales-Santana, 582 U. S., at 59 (internal quotation marks omitted).

This “heightened scrutiny” standard was developed in recognition of the fact that classifications based on sex share many features with classifications based on race. Early on, the lead opinion in Frontiero v. Richardson observed that “our Nation has had a long and unfortunate history of sex discrimination” that resulted in “statute books . . . laden with gross, stereotyped distinctions between the sexes.” 411 U. S., at 684–685. Although the opinion acknowledged “that the position of women in America ha[d] improved markedly,” it noted that “women still face[d] pervasive, although at times more subtle, discrimination.” Id., at 685–686. That pervasive discrimination against women could be explained “in part because of the high visibility of the sex characteristic.” Id., at 686. And such sex-discrimination was particularly unfair, the opinion reasoned, because “sex, like race and national origin, is an immutable characteristic determined solely by the accident of birth.” Ibid.

On these bases, the Frontiero plurality opined that classifications based on sex should be treated as “inherently suspect,” just like classifications based on race. Id., at 688. Although the full Court never adopted that position, it has justified the imposition of “heightened scrutiny” on largely the same grounds. As the Court later noted in Craig, a whole range of laws still on the books reflected “archaic and overbroad generalizations” and “increasingly outdated misconceptions concerning the role of females.” 429 U. S., at 198–199 (internal quotation marks omitted). The Court has further observed that women, like blacks and other racial minorities, tend to “carry an obvious badge” of their membership in the disadvantaged class, and the Court saw this as a partial explanation for “the severity or pervasiveness” of the discrimination experienced by both groups. Mathews, 427 U. S., at 506. And women, like blacks, had long been excluded, either by law or prejudice, from equal participation in the political process. See VMI, 518 U. S., at 531.

Thus, the application of “heightened scrutiny” to sex classifications can be explained in large part by the fact that sex discrimination shares many characteristics with racial discrimination: it was historically entrenched and pervasive; it was based on identifiable and immutable characteristics; and it included barriers to full participation in the political process.

Despite all this, however, the Court has not perfectly equated these two forms of discrimination. See id., at 532. We have acknowledged that the “[p]hysical differences between men and women . . . are enduring” and “remain cause for celebration.” Id., at 533. For this reason, sex is not a categorically “proscribed classification.” Ibid. “Principles of equal protection do not require” legislators to “ignore th[e] reality” that there are real differences between men and women that may sometimes justify legislation that classifies based on sex. Nguyen, 533 U. S., at 66. And classifications based on sex have occasionally been upheld. See, e.g., Michael M. v. Superior Court, Sonoma Cty., 450 U.S. 464, 475–476 (1981) (plurality opinion); Nguyen, 533 U. S., at 73.

3

Although the Court has held that classifications based on race, national origin, and sex call for a higher level of scrutiny, it has frequently refused to apply such scrutiny to other classifications. And it has done so even when those classifications share some characteristics with race, national origin, and sex. A few examples are sufficient to illustrate the Court’s general approach. Despite the fact that poor people have often been subjected to harsh and disrespectful treatment, a class defined by poverty is too “large, diverse, and amorphous” to qualify as suspect or “quasi-suspect.” Rodriguez, 411 U. S., at 28. Although age is an immutable characteristic, “the aged . . . have not experienced” the “ ‘history of purposeful unequal treatment’ ” that is needed to justify a higher level of scrutiny. Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 313 (1976) (per curiam) (quoting Rodriguez, 411 U. S., at 28). Presence in this country in violation of the immigration laws, although sometimes associated with social stigma, cannot define membership in a protected class because that status is not “an absolutely immutable characteristic” and may be relevant to “proper legislative goal[s].” Plyler, 457 U. S., at 220. Family relational status is likewise not entitled to elevated scrutiny because “[c]lose relatives . . . have not been subjected to discrimination” and “do not exhibit obvious, immutable, or distinguishing characteristics that define them as a discrete group.” Lyng, 477 U. S., at 638.

Even in close cases, the Court has been notably reluctant to apply an elevated level of scrutiny. This is particularly striking in the case of persons with disabilities. In Cleburne, the Court considered whether it should apply “[h]eightened scrutiny” to laws that classify based on intellectual disability. 473 U. S., at 442–443. The Court acknowledged that the intellectually disabled are “immutably” different and that “there have been and there will continue to be instances of discrimination against [them] that are in fact invidious.” Id., at 442, 446. Nonetheless, the Court found that “the States’ interest in dealing with and providing for [these individuals] is plainly a legitimate one,” id., at 442, and that “lawmakers have been addressing their difficulties in a manner that belies a continuing antipathy or prejudice,” id., at 443. The Court further recognized that the intellectually disabled are a “large and diversified group” and are not “all cut from the same pattern.” Id., at 442. In light of all these facts, the Court was reluctant to identify a new suspect or “quasi-suspect” class based on the existence of “immutable disabilities” and “some degree of prejudice from at least part of the public at large.” Id., at 445.

Overall, our decisions refusing to identify new suspect and “quasi-suspect” classes exhibit two salient features. First, the identification of a suspect or “quasi-suspect” class has been exceedingly rare. Such status has been denied to groups, like persons with disabilities and the aged, who were found by Congress to need special legislation to protect them from widespread discrimination. See, e.g., Rehabilitation Act of 1973, 29 U. S. C. §701 et seq.; Americans with Disabilities Act of 1990, 42 U. S. C. §12101 et seq.; Individuals with Disabilities Education Act, 20 U. S. C. §1400 et seq.; Age Discrimination in Employment Act of 1967, 29 U. S. C. §621 et seq. Accordingly, the Court’s reluctance to apply a special level of scrutiny to a proposed class should not be taken as a denial of the fact that the class has suffered from harmful discrimination or a lack of political power.

Second, no single characteristic is independently sufficient to qualify a proposed class as suspect or “quasi-suspect”; instead, in the rare instances in which the Court has identified a suspect or “quasi-suspect” class, it has done so based on a strong showing of multiple relevant criteria: a history of widespread and conspicuous discrimination, de facto or de jure exclusion from equal participation in the political process, and an immutable characteristic that tends to serve as an obvious badge of membership in a clearly defined and readily identifiable group.

4

With this background in mind, I do not think that transgender status is sufficiently similar to race, national origin, or sex to warrant a higher level of scrutiny.

Although transgender persons have undoubtedly experienced discrimination, the plaintiffs and their many amici have not been able to show a history of widespread and conspicuous discrimination that is similar to that experienced by racial minorities or women. Instead, they provide little more than conclusory statements. See, e.g., Brief for United States 29; Brief for Respondents in Support of Petitioner 37. But as we explained in Cleburne, heightened scrutiny cannot be justified on the ground that a proposed class has suffered from “some degree of prejudice from at least part of the public at large.” 472 U. S., at 445. Rather, a higher level of scrutiny is reserved for those groups, like racial minorities and women, who have suffered from a long history of discrimination that is both severe and pervasive. See Frontiero, 411 U. S., at 684 (plurality opinion) (“[O]ur Nation has had a long and unfortunate history of sex discrimination”); Mathews, 427 U. S., at 506 (characterizing the historic discrimination faced by women and blacks as “sever[e] and pervasiv[e]”).

Furthermore, there is no evidence that transgender individuals, like racial minorities and women, have been excluded from participation in the political process. It is certainly true that the very small size of the transgender population means that the members of this group cannot wield much political clout simply by casting their votes. But that is true of “a variety of other groups . . . who cannot themselves mandate the desired legislative responses.” Cleburne, 473 U. S., at 445. And despite the small size of the transgender population, the members of this group have had notable success in convincing many lawmakers to address their problems. See Brief for Respondents 47 (citing Cal. Educ. Code Ann. §221.5(f ) (West 2021); Va. Code Ann. §38.2–3449.1 (2020); Wash. Rev. Code Ann. §28 A. 642.080 (2024)); see also Cleburne, 473 U. S., at 443 (arguing that the “distinctive legislative response” to the problems of the intellectually disabled “belies a continuing antipathy or prejudice and a corresponding need for more intrusive oversight by the judiciary”).

The parties in this case also admit that transgender status is not an immutable characteristic. See Tr. of Oral Arg. 97–98. Instead, a person’s gender identity may “shif[t],” and a person who is transgender now may not be transgender later. Id., at 98; see also Brief for Society for Evidence-Based Gender Medicine as Amicus Curiae 19–25 (discussing the rates of desistance among transgender youth). Moreover, transgender status, unlike race and sex, is often not accompanied by visibly identifiable characteristics. A person’s “gender identity” is an “internal sense,” Merriam-Webster’s Collegiate Dictionary, at 520, and transgender persons as a class do not uniformly “exhibit obvious, immutable, or distinguishing characteristics that define them as a discrete group,” Lyng, 477 U. S., at 638. Nor do they necessarily tend to “carry an obvious badge” of their membership in the class that might serve to exacerbate discrimination. Mathews, 427 U. S., at 506.

Finally, the definition of transgender status that we have been given reveals that transgender people make up a “diverse” and “amorphous class.” Rodriguez, 411 U. S., at 28. Individuals are regarded as transgender whenever “they have a gender identity that differs from the sex they were assigned at birth.” Brief for Respondents in Support of Petitioner 4. That definition encompasses not just biological men who permanently identify as women and biological women who permanently identify as men, but also individuals who might identify with a particular gender at a particular point in time and individuals who identify permanently or temporarily with both sexes, neither sex, or some other identity. See Brief for American Psychological Association et al. as Amici Curiae 6, and n. 7 (describing “transgender youth” as an “umbrella term” that can refer to minors who are “gender diverse” or “nonbinary”). We have previously refused to apply a higher level of scrutiny to such “amorphous” classes for good practical reasons. See, e.g., Rodriguez, 411 U. S., at 28; Cleburne, 473 U. S., at 442–443. Since such classes are not rigidly defined, it is hard to pin down whether they share the relevant characteristics that make closer scrutiny warranted. And it is difficult for both courts and legislatures to identify the outer bounds of such groups.

In light of all the above, I am unwilling to conclude that transgender status, like race, national origin, and sex, is entitled to a higher level of scrutiny than ordinary rational basis review. That conclusion, however, should not be taken as a denial of the discrimination that transgender people have faced. Nor should it be taken as an evaluation of any specific legislative action concerning transgender persons. It simply means that transgender persons, like members of other disadvantaged groups—the poor, the aged, the disabled, etc.—have not made the extraordinary showing that they are entitled to a higher level of constitutional scrutiny.

III

Because transgender status is not a suspect or “quasi-suspect” class, even if Tennessee’s SB1 classifies on that ground, it must be sustained so long as it “bears some fair relationship to a legitimate public purpose.” Plyler, 457 U. S., at 216. As the Court notes, SB1 easily satisfies that standard. Ante, at 21–24.

I therefore agree with the Court that the judgment of the United States Court of Appeals for the Sixth Circuit should be affirmed.

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Justice Thomas, concurring.

A Tennessee law prevents children from receiving certain medical interventions if administered to treat gender dysphoria. See Prohibition on Medical Procedures Performed on Minors Related to Sexual Identity, S. B. 1, 113th Gen. Assem., 1st Extra. Sess.; Tenn. Code Ann. §68–33–101 et seq. (2023) (SB1). The United States and private plaintiffs challenged the law on Equal Protection Clause grounds, arguing that it discriminates based on sex and fails heightened scrutiny. Today, the Court correctly concludes that SB1 does not classify on the basis of sex and thus is subject only to rational-basis review. I join the Court’s opinion in full. I write separately to address some additional arguments made in defense of Tennessee’s law.

I

Before this Court, the United States and the private plaintiffs asserted that, under the reasoning of Bostock v. Clayton County, 590 U.S. 644 (2020), SB1 discriminates on the basis of sex. See Brief for United States 22, 27–28;[1] Brief for Respondents in Support of Petitioner 18, 24–25. In Bostock, the Court held that, in the context of Title VII of the Civil Rights Act of 1964, “homosexuality and transgender status are inextricably bound up with sex,” such that discriminating on the basis of either characteristic amounts to discrimination “because of ” sex under that statute. 590 U. S., at 660–661, 665. The United States and the private plaintiffs have argued that Bostock’s “fundamental insight about the nature of sex discrimination applies in the equal-protection context” too. Brief for United States 27. I would reject that argument for several reasons.

While I continue to think that the Bostock majority’s logic “fails on its own terms,” see 590 U. S., at 689–699 (Alito, J., dissenting), I see in any event no reason to import Bostock’s Title VII analysis into the Equal Protection Clause. The Bostock Court recognized that “other federal . . . laws that prohibit sex discrimination” were not before it, id., at 681 (majority opinion), and thus rested its analysis on what it took to be the ordinary meaning of the relevant statutory terms—“ ‘because of,’ ” “ ‘otherwise . . . discriminate against,’ ” and “individual”—within the context of Title VII, id., at 656–659; see 42 U. S. C. §2000e–2(a)(1).

The Equal Protection Clause includes none of this language. See Amdt. 14, §1 (“No State shall . . . deny to any person within its jurisdiction the equal protection of the laws”). “That such differently worded provisions should mean the same thing is implausible on its face.” Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, 600 U.S. 181, 308 (2023) (Gorsuch, J., concurring); cf. Department of Ed. v. Louisiana, 603 U.S. 866, 867 (2024) (per curiam) (unanimously holding that “preliminary injunctive relief ” was warranted to enjoin a rule extending Bostock’s reasoning to Title IX of the Education Amendments of 1972).[2]

Extending the Bostock framework here would depart dramatically from this Court’s Equal Protection Clause jurisprudence. We have faced sexual-orientation claims in the equal protection context for decades. See, e.g., Obergefell v. Hodges, 576 U.S. 644 (2015); United States v. Windsor, 570 U.S. 744 (2013); Romer v. Evans, 517 U.S. 620 (1996). “But in those cases, the Court never suggested that sexual orientation discrimination is just a form of sex discrimination” warranting heightened constitutional scrutiny. Bostock, 590 U. S., at 797 (Kavanaugh, J., dissenting). For example, while pregnancy is undeniably “bound up with sex,” id., at 661 (majority opinion), the Court has rejected the contention that the exclusion of pregnancy-related conditions from disability benefits violates the Equal Protection Clause, see Geduldig v. Aiello, 417 U.S. 484, 494 (1974); see also Dobbs v. Jackson Women’s Health Organization, 597 U.S. 215, 236 (2022) (“[T]he regulation of a medical procedure that only one sex can undergo does not trigger heightened constitutional scrutiny”).

Applying Bostock’s reasoning to the Equal Protection Clause would also invite sweeping consequences. Many statutes “regulate medical procedures defined by sex.” L. W. v. Skrmetti, 83 F. 4th 460, 482 (CA6 2023) (collecting examples, including laws referencing testicular and prostate cancer). If heightened scrutiny applied to such laws, then “[a]ny person with standing to challenge” such a decision could “haul the State into federal court and compel it to establish by evidence (presumably in the form of expert testimony) that there is an ‘exceedingly persuasive justification’ for the classification.” United States v. Virginia, 518 U.S. 515, 597 (1996) (Scalia, J., dissenting). Given the ensuing potential for “high-cost, high-risk lawsuit[s],” ibid., States might simply decline to adopt or enforce sex-based medical laws or regulations, even where such rules would be best medical practice. The burden of skeptical judicial review is therefore far from the “modest step” of requiring a State to “show its work” that the dissent posits. Post, at 31 (opinion of Sotomayor, J.).[3]

And, if Bostock’s reasoning applies to sex, it is difficult to see why it would not apply to other protected characteristics. Race presumably would be a but-for cause of—or, at least, “inextricably bound up with,” 590 U. S., at 660–661—a university’s decision to credit “an applicant’s discussion of how race affected his or her life,” Students for Fair Admissions, Inc., 600 U. S., at 230. Under Bostock’s reasoning, such an essay is permissible only if it can survive our “daunting” strict-scrutiny standard. 600 U. S., at 206; but see, e.g., Washington v. Davis, 426 U.S. 229, 239 (1976) (noting that the Court has “never held that the constitutional standard for adjudicating claims of invidious racial discrimination is identical to the standards applicable under Title VII”).

The Constitution compels none of this. While the majority concludes that SB1 does not discriminate based on sex even under Bostock’s incorrect reasoning, see ante, at 18–19, I would make clear that, in constitutional challenges, courts need not engage Bostock at all.

II

The Court rightly rejects efforts by the United States and the private plaintiffs to accord outsized credit to claims about medical consensus and expertise. The United States asserted that “the medical community and the nation’s leading hospitals overwhelmingly agree” with the Government’s position that the treatments outlawed by SB1 can be medically necessary. Brief for United States 35; see also Brief for Respondents in Support of Petitioner 5 (asserting that “[e]very major medical association in the United States” supports this position). The implication of these arguments is that courts should defer to so-called expert consensus.

There are several problems with appealing and deferring to the authority of the expert class. First, so-called experts have no license to countermand the “wisdom, fairness, or logic of legislative choices.” FCC v. Beach Communications, Inc., 508 U.S. 307, 313 (1993). Second, contrary to the representations of the United States and the private plaintiffs, there is no medical consensus on how best to treat gender dysphoria in children. Third, notwithstanding the alleged experts’ view that young children can provide informed consent to irreversible sex-transition treatments, whether such consent is possible is a question of medical ethics that States must decide for themselves. Fourth, there are particularly good reasons to question the expert class here, as recent revelations suggest that leading voices in this area have relied on questionable evidence, and have allowed ideology to influence their medical guidance.

Taken together, this case serves as a useful reminder that the American people and their representatives are entitled to disagree with those who hold themselves out as experts, and that courts may not “sit as a super-legislature to weigh the wisdom of legislation.” Day-Brite Lighting, Inc. v. Missouri, 342 U.S. 421, 423 (1952). By correctly concluding that SB1 warrants the “paradigm of judicial restraint,” Beach Communications, 508 U. S., at 314, the Court reserves to the people of Tennessee the right to decide for themselves.

A

The views of self-proclaimed experts do not “shed light on the meaning of the Constitution.” Dobbs, 597 U. S., at 272–273. Thus, whether “major medical organizations” agree with the result of Tennessee’s democratic process is irrelevant. Post, at 5, n. 5 (opinion of Sotomayor, J.). To hold otherwise would permit elite sentiment to distort and stifle democratic debate under the guise of scientific judgment, and would reduce judges to mere “spectators . . . in construing our Constitution.” 83 F. 4th, at 479.

Just a few Terms ago, this Court acknowledged the importance of reserving to the democratic process the right to decide controversial medical questions. In Dobbs, the respondents sought to invoke the authority of “overwhelming medical consensus” and “numerous major medical organizations” to dispatch with Mississippi’s asserted interest in minimizing pain for the unborn. Brief for Respondents, O. T. 2021, No. 19–1932, pp. 31–32. The Court pointedly rejected the notion that a consensus among popular expert groups could remove “the mitigation of fetal pain” from the “legitimate interests” of the people. 597 U. S., at 301.

Rational-basis review is critical to safeguarding these legitimate interests. Under this level of review, courts ask only whether a law is “rationally related to a legitimate governmental interest.” Department of Agriculture v. Moreno, 413 U.S. 528, 533 (1973). That deferential standard is not only legally compelled in this case, but is practically essential for preserving “the original constitutional proposition that courts do not substitute their social and economic beliefs for the judgment of legislative bodies.” Ferguson v. Skrupa, 372 U.S. 726, 730 (1963). When legislation does not cross constitutional lines, States must have leeway to effect the judgment of their citizens—no matter whether experts disagree. And, when this Court has nonetheless given exalted status to expert opinion, it has been to our detriment: Past deference to expertise provided the theory of eugenics “added legitimacy and considerable momentum,” with “[t]his Court thr[owing] its prestige behind the eugenics movement in its 1927 decision upholding the constitutionality of Virginia’s forced-sterilization law.” Box v. Planned Parenthood of Ind. and Ky., Inc., 587 U.S. 490, 499–500 (2019) (Thomas, J., concurring) (citing Buck v. Bell, 274 U.S. 200 (1927)). Fortunately, we do not repeat that mistake today.

B

Before this Court, the United States asserted that “overwhelming evidence” supports the use of puberty blockers and cross-sex hormones for treating pediatric gender dysphoria, and that this view represents “the overwhelming consensus of the medical community.” Pet. for Cert. 2, 7. These claims are untenable. “[T]he concept of gender dysphoria as a medical condition is relatively new and the use of drug treatments that change or modify a child’s sex characteristics is even more recent.” 83 F. 4th, at 472. The treatments at issue are subject to a rapidly evolving debate that demonstrates a lack of medical consensus over their risks and benefits. Under these conditions, it is imperative that courts treat state legislation with “a strong presumption of validity,” Beach Communications, 508 U. S., at 314, and in turn protect States’ ability to enact “high-stakes medical policies, in which compassion for the child points in both directions,” 83 F. 4th, at 472.

1

SB1 prohibits puberty blockers, cross-sex hormones, and surgery for the purpose of treating gender dysphoria in children. See Tenn. Code Ann. §§68–33–102(5)(A)–(B), 68–33–103(a). The United States and the dissent have described these medications and procedures as “gender-affirming care.” Brief for United States 2; post, at 4 (opinion of Sotomayor, J.). But, that “sanitized description” obscures the nature of the medical interventions at issue. Stenberg v. Carhart, 530 U.S. 914, 983 (2000) (Thomas, J., dissenting). I therefore begin with an overview of the treatments regulated under SB1.

Puberty Blockers. Puberty blockers are powerful synthetic drugs “designed to slow the development of male and female physical features.” 83 F. 4th, at 467. The Food and Drug Administration (FDA) initially approved these drugs “to treat prostate cancer; endometriosis, a painful disease that causes uterine tissue to grow elsewhere in the body; and the unusually early onset of puberty,” also known as “precocious puberty.” M. Twohey & C. Jewett, Pressing Pause on Puberty, N. Y. Times, Nov. 14, 2022, pp. A14–A15 (Twohey 2022).

For purposes of treating gender dysphoria, however, puberty blockers generally are administered “off-label,” meaning without FDA authorization for the specific use. See 2 App. 838–839; 83 F. 4th, at 478. Although it is neither unusual nor unlawful for drugs to be used off-label, the FDA has recognized that “just because a drug has been approved for one class of patients doesn’t mean it’s safe for another.” Twohey 2022, at A15. That admonition is important here: To treat precocious puberty, puberty blockers are administered until the age appropriate for puberty; to treat gender dysphoria, however, puberty blockers are administered to stop puberty throughout the years it would normally occur. See 2 App. 677. The “use of drugs to suppress normal puberty has multiple organ system effects whose long-term consequences have not been investigated.” Ibid.

This absence of evidence is a “major drawback” in assessing the effects of puberty blockers on children with gender dysphoria. G. Betsi, P. Goulia, S. Sandhu, & P. Xekouki, Puberty Suppression in Adolescents With Gender Dysphoria: An Emerging Issue With Multiple Implications, Frontiers in Endocrinology 16 (2024). “The existing studies are limited in number, of small sample size, uncontrolled, observational, usually short-term, [and] potentially subject to bias.” Ibid.; see also, e.g., C. Terhune, R. Respaut, & M. Conlin, As More Transgender Children Seek Medical Care, Families Confront Many Unknowns, Reuters (Oct. 6, 2022), https://www.reuters.com/investigates/special-report/usa-transyouth-care (“No clinical trials have established [puberty blockers’] safety for such off-label use”).

It is undisputed, however, that these treatments carry risks. Research suggests that, aside from interrupting a child’s normal pubertal development, puberty blockers may lead to decreased bone density and impacts on brain development. See, e.g., 2 App. 678–680; M. Cretella, Gender Dysphoria in Children, 32 Issues in L. & Med. 287, 297 (2017). And, “[d]espite widespread assertions that puberty blockers are ‘fully reversible,’ ” it is unclear whether “patients ever develop normal levels of fertility if puberty blockers are terminated after a ‘prolonged delay of puberty.’ ” 2 App. 678. At bottom, “[t]here remains considerable uncertainty regarding the effects of puberty blockers in individuals experiencing” gender dysphoria. A. Miroshnychenko et al., Puberty Blockers for Gender Dysphoria in Youth: A Systematic Review and Meta-Analysis, Online First, Archives of Disease in Childhood (Jan. 24, 2025) (draft, at 1), https://adc.bmj.com/content/110/6/429.[4]

Cross-sex hormones. Following puberty blockers, the next stage of sex-transition treatments for children involves cross-sex hormones. This treatment is also typically “off-label,” 2 App. 780, and requires “very high doses” of hormones of the opposite sex, id., at 769. For example, one of the organizations that sets standards for pediatric sex- transition treatment recommends raising transitioning females’ levels of testosterone “6 to 100 times higher than native female testosterone levels.” Id., at 774. For males seeking to transition into females, the organization recommends raising levels of estradiol, a type of estrogen, to “2 to 43 times above the normal range.” Id., at 780.

Prescribing such high doses of testosterone to girls induces “hyperandrogenism,” which can cause increased cardiovascular risk, “irreversible changes to the vocal cords,” “clitoromegaly and atrophy of the lining of the uterus and vagina,” as well as “ovarian and breast cancer.” Id., at 772–779. Giving high doses of estrogen to boys induces “hyperestrogenemia,” which can produce similarly severe side effects including, among other things, increased cardiovascular risk, breast cancer, and sexual dysfunction. Id., at 779–781. And, for girls and boys alike, “it is generally accepted, even by advocates of transgender hormone therapy, that hormonal treatment impairs fertility, which may be irreversible.” Id., at 520–521; accord, W. Hembree et al., Endocrine Treatment of Gender-Dysphoric/Gender-Incongruent Persons: An Endocrine Society Clinical Practice Guideline, 102 J. Clinical Endocrinology & Metabolism 3869, 3882 (2017) (ES Guidelines).

Surgery. SB1 also bans “[s]urgically removing, modifying, altering, or entering into tissues, cavities, or organs” as a treatment for gender dysphoria. Tenn. Code Ann. §68–33–102(5)(A). The District Court concluded that the plaintiffs lacked standing to challenge SB1’s ban on sex-transition surgery for minors, see ante, at 6, and the parties do not address this provision’s constitutionality here. But, the United States has taken the position that “surgery is essential and medically necessary to alleviate gender dysphoria.” Amended Complaint in Intervention in Boe v. Marshall, No. 2:22–cv–00184 (MD Ala., May 4, 2022), ECF Doc. 92, p. 9, ¶39. The practice therefore warrants brief discussion.

Sex-transitioning surgeries for girls include “the surgical removal of the breasts” and “phalloplasty,” that is, an “attemp[t] to create a pseudo-penis” by transplanting “a roll of skin and subcutaneous tissue” from another area of the body “to the pelvis.” 2 App. 784–785; see also Lange v. Houston Cty., 101 F. 4th 793, 802 (CA11) (Brasher, J., dissenting) (“[A] natal woman’s phalloplasty ‘involves removal of the uterus, ovaries, and vagina, and creation of a neophallu[s] and scrotum with scrotal prostheses,’ which ‘is a multistage reconstructive procedure’ ”), vacated and reh’g en banc granted, 110 F. 4th 1254 (CA11 2024). For boys, surgical interventions include “removal of the testicles alone to permanently lower testosterone levels,” as well as an “attempt to create a pseudo-vagina” by “surgically open[ing]” the boy’s penis, removing “erectile tissue,” and then “clos[ing] and invert[ing the penis] into a newly created cavity in order to simulate a vagina.” 2 App. 784. These surgical interventions are irreversible, entail significant complications, and, in some cases, result in permanent infertility. Id., at 782–786; see also ES Guidelines 3893.

2

The ongoing debate over the efficacy of sex-transition treatments for children confirms that medical and regulatory authorities are not of one mind about the treatments’ risks and benefits. These conditions illustrate why States may rightly be skeptical of groups or advocates claiming that expert consensus supports their position, and why courts must exercise restraint in reviewing state legislatures’ decisions in this area. Accord, e.g., Beach Communications, 508 U. S., at 314.

The treatments now referred to as “gender-affirming care” were “not available for minors until just before the millennium.” 83 F. 4th, at 467. These treatments originated with Dutch healthcare workers in the 1990s, who first “began using puberty blockers . . . to treat gender dysphoria in minors.” Ibid. The so-called “Dutch Protocol” “permitted puberty blockers for minors during the early stages of puberty, allowed hormone therapy at 16, and allowed genital surgery at 18.” Ibid. (internal quotation marks omitted).

In 1998, the World Professional Association for Transgender Health (WPATH)—which is regarded by some as “the leading association of medical professionals treating transgender individuals,” Brief for United States 3—revised its treatment standards to “endorse the Dutch Protocol.” 83 F. 4th, at 467. Originally, WPATH’s guidelines permitted puberty blockers at the onset of puberty, cross-sex hormones for those 16 or older, and sex-change surgery only for adults. Ibid. WPATH relaxed its recommendations in 2012, and began permitting cross-sex hormones for children under the age of 16. Ibid. WPATH further relaxed its recommendations when it published the eighth (and current) version of its standards of care in 2022. See E. Coleman et al., Standards of Care for the Health of Transgender and Gender Diverse People, Version 8, 23 Int’l J. Transgender Health (2022) (WPATH 2022 Guidelines or Guidelines). These Guidelines endorse using puberty blockers and cross-sex hormones at the onset of puberty and allowing children to receive many surgical treatments previously reserved for adults. See id., at S64–S66. “On the whole, the standards of care for minors ‘have become less restrictive over the course of time.’ ” 83 F. 4th, at 468.

At the same time, the number of children identifying as transgender has surged, and medical professionals have increasingly expressed doubts over the quality of evidence supporting the use of puberty blockers, cross-sex hormones, and surgery to treat them. See ante, at 3. Over the past several years, public health authorities in different countries have concluded that these sex-transition treatments are experimental in practice, and that the evidence supporting their use is of “ ‘very low certainty,’ ” “ ‘insufficient,’ ” and “ ‘inconclusive.’ ” Ibid. “In countries like Sweden, Norway, France, the Netherlands and Britain—long considered exemplars of gender progress—medical professionals have recognized that early research on medical interventions for childhood gender dysphoria was either faulty or incomplete.” P. Paul, Gender Dysphoric Kids Deserve Better Care, N. Y. Times, Feb. 4, 2024, p. 9 (Paul 2024); accord, Tenn. Code Ann. §68–33–101(e) (“The legislature finds that health authorities in Sweden, Finland, and the United Kingdom . . . have found no evidence that the benefits of these procedures outweigh the risks”); 1 App. 332–342 (describing countries’ skepticism over the use of puberty blockers and cross-sex hormones as treatments).[5]

The Cass Review, published in April 2024, offers an influential example of the degree to which the debate over pediatric sex-transition treatments remains unsettled. See H. Cass, Independent Review of Gender Identity Services for Children and Young People: Final Report (Cass Review). After witnessing a 40-fold increase in the number of referrals to its centralized clinic for sex-transitioning services, the United Kingdom’s National Health Service (NHS) commissioned this report to conduct a “thorough independent review of the use of puberty blockers and cross-sex hormones” to treat children with gender dysphoria. 1 App. 333–334. The report concludes that “we have no good evidence on the long-term outcomes of interventions to manage gender-related distress,” and highlights the lack of reliable evidence to support the use of puberty blockers and cross-sex hormones in treating transgender kids. Cass Review 13, 32–33 (observing “insufficient/inconsistent evidence about the effects of puberty suppression,” and “ ‘a lack of high-quality research assessing the outcomes of hormone interventions in adolescents with gender dysphoria/incongruence’ ”); see also ante, at 23. Among other things, the Cass Review determined that the “evidence [the researchers] found did not support th[e] conclusion” that “hormone treatment reduces the elevated risk of death by suicide” among children suffering from gender dysphoria. Cass Review 33; see also id., at 187 (“[T]he evidence does not adequately support the claim that gender-affirming treatment reduces suicide risk”).

This shifting scientific landscape has forced governments to act quickly under conditions of uncertainty. In the months following the Cass Review’s publication, for example, NHS imposed new restrictions on the use of puberty blockers and cross-sex hormones for sex-transition treatments. See ante, at 23. And, just a week after oral argument in this case, the United Kingdom indefinitely banned new prescriptions of puberty blockers to treat children with gender dysphoria, except in clinical trials. See S. Castle, Ban on Puberty Blockers for U. K. Teens Is Settled, N. Y. Times Int’l, Dec. 13, 2024, p. A11. In areas with this much “medical and scientific uncertainty,” courts must afford States “wide discretion.” Gonzales v. Carhart, 550 U.S. 124, 163 (2007).

C

Setting aside whether sex-transition treatments for children are effective, States may legitimately question whether they are ethical. States have a legitimate interest “in protecting the integrity and ethics of the medical profession.” Washington v. Glucksberg, 521 U.S. 702, 731 (1997). And, as the United States has acknowledged, “the ‘general ethical principles’ governing pediatric care” require the patient’s informed consent. Brief for United States 5. Mounting evidence gives States reason to question whether children are capable of providing informed consent to irreversible sex-transition treatments, and thus whether these treatments can be ethically administered.

1

States could reasonably conclude that the level of young children’s cognitive and emotional development inhibits their ability to consent to sex-transition treatments. Consistent with WPATH’s recommendation that puberty blockers be available from the onset of puberty, see WPATH 2022 Guidelines S111, S256, “[m]any physicians in the United States and elsewhere” now “prescrib[e] blockers to patients at the first stage of puberty—as early as age 8.” Twohey 2022, at A14.

There is no dispute, however, that the “decision-making capacity” of adolescents “is developing, but not yet complete.” 2 App. 895. This Court has recognized as much in other contexts, explaining that children’s “lack of maturity” and “underdeveloped sense of responsibility” often lead to “impetuous and ill-considered actions and decisions.” Roper v. Simmons, 543 U.S. 551, 569 (2005) (internal quotation marks omitted). It is therefore unsurprising that “[t]he risks associated with puberty blockers and cross-sex hormones are difficult for adolescents to comprehend and appreciate,” as the “near certainty of infertility . . . is likely to not be appreciated until the age during which most individuals consider having children.” 2 App. 894.

But, these are precisely the risks to which children who receive these treatments are required to consent. Consider the contents of a consent form obtained from a gender clinic in Alabama. After providing a long list of potential risks and side effects, many of which are discussed above, see supra, at 7–10, the form requires both the child and parent to initial their consent to various statements. Among these are acknowledgments that “the side effects and safety of these medicines are not completely known,” that the proposed treatment “may affect my sex life in different ways and future ability to cause a pregnancy,” and that the treatments may lead to permanent infertility. See Boe, ECF Doc. 78–41, pp. 3–4, 10. The capacity to knowingly consent to these medical interventions requires a level of comprehension about science, sex, and fertility that state legislatures could determine a child is unlikely to possess. See 2 App. 893–895; Tenn. Code Ann. §68–33–101(h) (finding that “minors lack the maturity to fully understand and appreciate the life-altering consequences” of the treatments at issue).[6]

2

The voices of “detransitioners”—individuals who have undergone sex-transition treatments but no longer view themselves as transgender—provide States with an additional reason to question whether children are providing informed consent to the medical interventions described above. See, e.g., Brief for Larger Detransitioners Community as Amici Curiae 24–28; Brief for Partners for Ethical Care et al. as Amici Curiae 17–38.

A recurring theme in discussions of detransitioners is that doctors have responded to the “skyrocketing” “number of adolescents requesting [sex-transitioning] medical care” by “hastily dispensing medicine or recommending medical doctors prescribe it.” L. Edwards-Leeper & E. Anderson, The Mental Health Establishment Is Failing Trans Kids, Washington Post, Nov. 24, 2021, pp. B1–B2. In many cases, evidence suggests that children “are being rushed toward” medical treatment “[w]ithout proper assessment,” and “the rising number of detransitioners that clinicians report seeing . . . indicates that this approach can backfire.” Id., at B2; accord, e.g., Eknes-Tucker v. Governor of Ala., 114 F. 4th 1241, 1267 (CA11 2024) (opinion of Lagoa, J.) (“Alabama presented evidence from many detransitioners who uniformly testified that they were not aware of the long-term impacts of the treatments they underwent”); Brief for Respondents 12–13 (explaining that, before enacting SB1, the Tennessee Legislature heard testimony “from a detransitioner who explained that she was not ‘capable of making informed lifelong decisions’ as a teenager” but nevertheless received transition treatments).[7]

States have an interest in ensuring that minor patients have the time and capacity to fully understand the irreversible treatments they may undergo. Cf. Gonzales, 550 U. S., at 159 (identifying State’s “legitimate concern” regarding “lack of information” provided by abortionists). And, despite the supposed expert consensus that young children can consent to irreversible sex-transition treatments, States have good reasons to disagree; as “any parent knows,” children’s comprehension is limited, Roper, 543 U. S., at 569, and the growing number of detransitioners illustrates the risks of assuming otherwise.

D

Recent revelations suggest that WPATH, long considered a standard bearer in treating pediatric gender dysphoria, see Brief for United States 3, bases its guidance on insufficient evidence and allows politics to influence its medical conclusions. Beyond the lack of consensus over the efficacy and ethics of pediatric sex-transition treatments, these developments provide States even stronger bases for treating supposed authorities in this area with skepticism.

WPATH itself recognizes that evidence supporting the efficacy of puberty blockers, cross-sex hormones, and surgical intervention for treating gender dysphoria in children is lacking. In its most recent Guidelines, for example, the group notes that “[a] key challenge in adolescent transgender care is the quality of evidence evaluating the effectiveness of medically necessary gender-affirming medical and surgical treatments . . . over time.” WPATH 2022 Guidelines S45–S46 (emphasis added). A contributor to the Guidelines underscored this challenge, explaining that, “ ‘[o]ur concerns, echoed by the social justice lawyers we spoke with, is that evidence-based review reveals little or no evidence and puts us in an untenable position in terms of affecting policy or winning lawsuits.’ ” Eknes-Tucker, 114 F. 4th, at 1261 (opinion of Lagoa, J.).

Nevertheless, WPATH publicly represents that “[g]ender-affirming interventions are based on decades of clinical experience and research,” and are “safe and effective” treatments. Guidelines S18. WPATH appears to rest this conclusion on self-referencing consensus rather than evidence-based research, which may help explain the group’s confidence in the face of concededly inadequate evidence. See Cass Review 130. In its analysis of several “guidelines” for transgender medicine—including not only the WPATH 2022 Guidelines, but also those from groups like the Endocrine Society—the Cass Review notes that “most of the guidelines described insufficient evidence about the risks and benefits of medical treatment in adolescents,” but nevertheless “went on to cite this same evidence to recommend medical treatments,” or to base their recommendations on “other guidelines” prescribing the same course of action. Ibid. (emphasis added). This approach was particularly pronounced in the WPATH 2022 Guidelines, which “cited many of the other national and regional guidelines to support some of its recommendations, despite these guidelines having been considerably influenced by WPATH 7,” the prior version of WPATH’s Standards of Care. Cass Review 130.

States would also have good reason to question whether WPATH has a basis for believing that children can provide informed consent to sex-transition treatments. “[I]n a leaked recording of a WPATH Panel,” for example, an endocrinologist acknowledged the difficulty of explaining cross-sex hormones and puberty blockers to children, noting that “ ‘the thing you have to remember about kids is that we’re often explaining these sorts of things to people who haven’t even had biology in high school yet.’ ” Eknes-Tucker, 114 F. 4th, at 1268–1269 (opinion of Lagoa, J.). “ ‘[I]t’s always a good theory that you talk about fertility preservation with a 14 year old,’ ” the endocrinologist continued, “ ‘but I know I’m talking to a blank wall.’ ” Id., at 1269. Analogizing a teenage patient’s comprehension to that of a blank wall should raise serious concerns regarding the patient’s ability to provide informed consent. Given WPATH’s recognition that “[c]onsent requires the cognitive capacity to understand the risks and benefits of a treatment,” Guidelines S38, States thus might reasonably question whether WPATH could be “genuine in its claim that these treatments are safe, effective, and well understood, particularly for minors,” Eknes-Tucker, 114 F. 4th, at 1268 (opinion of Lagoa, J.).

Other “recent revelations” might reinforce the conclusion that “WPATH’s lodestar is ideology, not science.” Id., at 1261. For example, newly released documents suggest that WPATH tailored its Standards of Care in part to achieve legal and political objectives. In one instance, the chair of WPATH’s guidelines committee testified that it was “ethically justifiable” for the authors of the WPATH 2022 Guidelines to “advocate for language changes [in these Guidelines] to strengthen [their] position in court.” Boe, ECF Doc. 700–3, p. 42. One of the Guidelines’ contributors was more direct: “My hope with these [Guidelines] is that they land in such a way as to have serious effect in . . . law and policy settings.” ECF Doc. 700–13, p. 25; see also Brief for State of Alabama as Amicus Curiae 11–15 (Alabama Brief ) (describing similar statements from other WPATH contributors).

Worse, recent reporting has exposed that WPATH changed its medical guidance to accommodate external political pressure. See Brief for Respondents 9–11; Alabama Brief 15–23. Unsealed documents reveal that a senior official in the Biden administration “pressed [WPATH] to remove age limits for adolescent surgeries from guidelines for care of transgender minors” on the theory that “ ‘specific listings of ages, under 18, will result in devastating legislation for trans care.” A. Ghorayshi, Biden Officials Pushed To Remove Age Limits for Trans Surgery, N. Y. Times, June 27, 2024, p. A17. Despite some internal disagreement, WPATH acceded and “removed the age minimums in its eighth edition of the standards of care.” Ibid.; see Alabama Brief 17–20.[8]

Over a decade ago, one of WPATH’s contributors explained that “ ‘WPATH aspires to be both a scientific organization and an advocacy group for the transgendered,’ ” and admitted that WPATH’s Standards of Care “ ‘is not a politically neutral document.’ ” Kosilek v. Spencer, 774 F.3d 63, 78 (CA1 2014). WPATH’s apparent willingness to let political interests influence its medical conclusions highlights this reality. States are never required to substitute expert opinion for their legislative judgment, and, when the experts appear to have compromised their credibility, it makes good sense to chart a different course.[9]

*  *  *

This case carries a simple lesson: In politically contentious debates over matters shrouded in scientific uncertainty, courts should not assume that self-described experts are correct.

Deference to legislatures, not experts, is particularly critical here. Many prominent medical professionals have declared a consensus around the efficacy of treating children’s gender dysphoria with puberty blockers, cross-sex hormones, and surgical interventions, despite mounting evidence to the contrary. They have dismissed grave problems undercutting the assumption that young children can consent to irreversible treatments that may deprive them of their ability to eventually produce children of their own. They have built their medical determinations on concededly weak evidence. And, they have surreptitiously compromised their medical recommendations to achieve political ends.

The Court today reserves “to the people, their elected representatives, and the democratic process” the power to decide how best to address an area of medical uncertainty and extraordinary importance. Ante, at 24. That sovereign prerogative does not bow to “major medical organizations.” Post, at 5, n. 5 (opinion of Sotomayor, J.). “[E]xperts and elites have been wrong before—and they may prove to be wrong again.” Students for Fair Admissions, Inc., 600 U. S., at 268 (Thomas, J., concurring).

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Justice Barrett, with whom Justice Thomas joins, concurring.

Because the Court concludes that Tennessee’s Senate Bill 1 does not classify on the basis of transgender status, it does not resolve whether transgender status constitutes a suspect class. Ante, at 16–18; see Geduldig v. Aiello, 417 U.S. 484, 496 (1974). I write separately to explain why, in my view, it does not.

I

As a “practical necessity,” “most legislation classifies for one purpose or another.” Romer v. Evans, 517 U.S. 620, 631 (1996). Laws distribute benefits that advantage particular groups (like in-state tuition for residents), draw lines that might seem arbitrary (like income thresholds for means-tested benefits), and set rules for specific categories of people (like a particular profession or age group). Such classifications do not usually render a law unconstitutional. Instead, as a general matter, laws are presumed to be constitutionally valid, and a legislative classification will be upheld “so long as it bears a rational relation to some legitimate end.” Ibid.

There are only a few exceptions to this rule: classifications based on race, sex, and alienage. Racial and ethnic classifications receive strict scrutiny; to survive a constitutional challenge, they must be “ ‘narrowly tailored’ ” to serve “ ‘compelling governmental interests.’ ” Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, 600 U.S. 181, 206–207 (2023); see also Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 292 (1978) (opinion of Powell, J.) (observing that the Equal Protection Clause applies “to all ethnic groups seeking protection from official discrimination”). Classifications based on alienage are subject to similarly close scrutiny.[1] Nyquist v. Mauclet, 432 U.S. 1, 7 (1977). And laws distinguishing between men and women receive intermediate scrutiny; to survive a constitutional challenge, they must be “ ‘ “substantially related” ’ ” to achieving an “ ‘ “important governmental objectiv[e].” ’ ” United States v. Virginia, 518 U.S. 515, 533 (1996).

Beyond these categories, the set has remained virtually closed. Indeed, this Court “has not recognized any new constitutionally protected classes in over four decades, and instead has repeatedly declined to do so.” Ondo v. Cleveland, 795 F.3d 597, 609 (CA6 2015). So in urging us to recognize transgender status as a suspect classification, the plaintiffs face a high bar.[2]

To determine whether a group constitutes a “suspect class” akin to the canonical examples of race and sex, we apply a test derived from the famous footnote 4 in United States v. Carolene Products Co. See 304 U.S. 144, 152–153, n. 4 (1938) (suggesting that “prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities, and which may call for a correspondingly more searching judicial inquiry”). We consider whether members of the group in question “exhibit obvious, immutable or distinguishing characteristics that define them as a discrete group,” whether the group has, “[a]s a historical matter, . . . been subjected to discrimination,” and whether the group is “a minority or politically powerless.” Lyng v. Castillo, 477 U.S. 635, 638 (1986). The test is strict, as evidenced by the failure of even vulnerable groups to satisfy it: We have held that the mentally disabled, the elderly, and the poor are not suspect classes. See Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 442 (1985) (mental disability); Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 313–314 (1976) (per curiam) (age); San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 28 (1973) (poverty). In fact, as far as I can tell, we have never embraced a new suspect class under this test. Our restraint reflects the principle that “[w]hen social or economic legislation is at issue, the Equal Protection Clause allows the States wide latitude, and the Constitution presumes that even improvident decisions will eventually be rectified by the democratic processes.” Cleburne, 473 U. S., at 440 (citation omitted).

II

The Sixth Circuit held that transgender individuals do not constitute a suspect class, and it was right to do so.[3] To begin, transgender status is not marked by the same sort of “ ‘obvious, immutable, or distinguishing characteristics’ ” as race or sex. L. W. v. Skrmetti, 83 F. 4th 460, 487 (2023) (quoting Bowen v. Gilliard, 483 U.S. 587, 602 (1987)); see Lyng, 477 U. S., at 638. In particular, it is not defined by a trait that is “ ‘definitively ascertainable at the moment of birth.’ ” 83 F. 4th, at 487 (quoting Ondo, 795 F. 3d, at 609). The plaintiffs here, for instance, began to experience gender dysphoria at varying ages—some from a young age, others not until the onset of puberty. See Brief for Respondents in Support of Petitioner 8–12. Meanwhile, the plaintiffs acknowledge that some transgender individuals “detransition” later in life—in other words, they begin to identify again with the gender that corresponds to their biological sex. See Tr. of Oral Arg. 49, 108. Accordingly, transgender status does not turn on an “immutable . . . characteristi[c].” Lyng, 477 U. S., at 638.

Nor is the transgender population a “discrete group,” as our cases require. Ibid. Instead, like classes we have declined to recognize as suspect, the category of transgender individuals is “large, diverse, and amorphous.” Rodriguez, 411 U. S., at 28. The World Professional Association for Transgender Health states that the term “ ‘transgender’ can describe ‘a huge variety of gender identities and expressions.’ ” 83 F. 4th, at 487 (quoting Standards of Care for the Health of Transgender and Gender Diverse People S15 (8th ed. 2022)). The American Psychological Association similarly uses the phrase “ ‘transgender youth’ ” as an “umbrella term” “to describe . . . varied groups” with “many diverse gender experiences.” Brief for American Psychological Association et al. as Amici Curiae 6, n. 7. Underscoring the point, plaintiffs’ counsel acknowledged at oral argument that “there are people who fall within a transgender identity who may not fit into a binary identity.” Tr. of Oral Arg. 100. The boundaries of the group, in other words, are not defined by an easily ascertainable characteristic that is fixed and consistent across the group.

Finally, holding that transgender people constitute a suspect class would require courts to oversee all manner of policy choices normally committed to legislative discretion. The parties agree that the States have a legitimate interest in regulating health care. They also agree that transgender status implicates physical and mental health—indeed, this case is about the medical treatment of children with gender dysphoria, which is “clinically significant distress resulting from the incongruence between . . . gender identity and . . . sex assigned at birth,” and which “can result in severe anxiety, depression, self-harm, and even suicide.” Brief for Respondents in Support of Petitioner 4–5. The question of how to regulate a medical condition such as gender dysphoria involves a host of policy judgments that legislatures, not courts, are best equipped to make. See Cleburne, 473 U. S., at 441–442 (declining to recognize a suspect class when the “distinguishing characteristics” of the proposed class are “relevant to interests the State has the authority to implement”).

Consider just a few: What are the relevant risks and benefits to children of puberty blockers and hormone treatments? What is the age at which these treatments become appropriate? 15? 16? 18? What about surgeries? Expert disagreements highlight the difficulty of such choices. As the Court recounts, England, Finland, Norway, and Sweden have raised concerns about using puberty blockers or hormone treatments on juveniles with gender dysphoria and have limited such treatments, in some cases by allowing them to go forward only in a research setting. See 1 App. 332–342, 409–411; 2 id., at 726–727; ante, at 3–4. By contrast, the guidelines promulgated by the Endocrine Society, upon which the plaintiffs rely, broadly recommend treatment for adolescents with sustained gender dysphoria and the capacity to give informed consent. App. to Pet. for Cert. 256a–259a. As we have emphasized before, “state and federal legislatures [have] wide discretion to pass legislation in areas where there is medical and scientific uncertainty.” Gonzales v. Carhart, 550 U.S. 124, 163 (2007). The prospect of courts second-guessing legislative choices in this area should set off alarm bells. Cf. Lochner v. New York, 198 U.S. 45, 72 (1905) (Harlan, J., dissenting) (“What the precise facts are it may be difficult to say. It is enough for . . . this court to know . . . that the question is one about which there is room for debate and for an honest difference of opinion”).

Beyond the treatment of gender dysphoria, transgender status implicates several other areas of legitimate regulatory policy—ranging from access to restrooms to eligibility for boys’ and girls’ sports teams. If laws that classify based on transgender status necessarily trigger heightened scrutiny, then the courts will inevitably be in the business of “closely scrutiniz[ing] legislative choices” in all these domains. Cleburne, 473 U. S., at 441–442. To be sure, an individual law “ ‘inexplicable by anything but animus’ ” is unconstitutional. Trump v. Hawaii, 585 U.S. 667, 706 (2018). But legislatures have many valid reasons to make policy in these areas, and so long as a statute is a rational means of pursuing a legitimate end, the Equal Protection Clause is satisfied.

III

The conclusion that transgender individuals do not share the “obvious, immutable, or distinguishing characteristics” of “a discrete group” is enough to demonstrate that transgender status does not define a suspect class. Lyng, 477 U. S., at 638. But the second factor—whether the group has, “[a]s a historical matter, . . . been subjected to discrimination,” ibid.—also poses a problem for the plaintiffs’ argument.

In addressing this factor, the plaintiffs assume that a history of private discrimination may satisfy this condition. For instance, the plaintiffs argue that “it is undeniable that transgender individuals, as a class, have ‘historically been subject to discrimination including in education, employment, housing, and access to healthcare.’ ” Brief for United States 29; Brief for Respondents in Support of Petitioner 37 (adopting the arguments made by the United States).[4] The Solicitor General confirmed at oral argument that this argument did not turn on “discrimination . . . reflected in the laws.” Tr. of Oral Arg. 60. The District Court also assumed that a history of private discrimination could suffice to establish that a group comprises a suspect class. See L. W. v. Skrmetti, 679 F. Supp. 3d 668, 690 (MD Tenn. 2023).

This assumption is mistaken. For purposes of the Fourteenth Amendment, the relevant question is whether the group has been subject to a longstanding pattern of discrimination in the law. In other words, we ask whether the group has suffered a history of de jure discrimination.

Existing suspect classes had such a history. Most obviously, “[t]he clear and central purpose of the Fourteenth Amendment was to eliminate all official state sources of invidious racial discrimination in the States.” Loving v. Virginia, 388 U.S. 1, 10 (1967). We have made that point “repeatedly.” Students for Fair Admissions, Inc., 600 U. S., at 206 (gathering cases). In recognizing sex as a suspect class, we similarly emphasized that women faced more than a century’s worth of discrimination in the law: “[N]ot until 1920 did women gain a constitutional right to the franchise. And for a half century thereafter, it remained the prevailing doctrine that government, both federal and state, could withhold from women opportunities accorded men so long as any ‘basis in reason’ could be conceived for the discrimination.” Virginia, 518 U. S., at 531 (citation omitted); see also Frontiero v. Richardson, 411 U.S. 677, 684–685 (1973) (plurality opinion) (“As a result of notions such as these, our statute books gradually became laden with gross, stereotyped distinctions between the sexes”). And in protecting alienage, we underscored the many state laws that discriminated on that ground, typically by targeting individuals of a particular national origin. See, e.g., Takahashi v. Fish and Game Comm’n, 334 U.S. 410, 427 (1948) (Murphy, J., concurring) (discussing a state law “directed in spirit and in effect solely against aliens of Japanese birth”); Yick Wo v. Hopkins, 118 U.S. 356, 373–374 (1886) (identifying ordinances that discriminated against Chinese nationals). Indeed, Congress criminalized discrimination on the basis of alienage by state actors in 1870, “in response to California legislation restricting the rights of Chinese immigrants.” Rajaram v. Meta Platforms, Inc., 105 F. 4th 1179, 1183–1184 (CA9 2024); see 16Stat. 144 (codified, as amended, 18 U. S. C. §242).

The distinction between de jure discrimination and private animus is consistent with the Fourteenth Amendment’s text and purpose. Most fundamentally, the Fourteenth Amendment constrains state action, not private conduct. See National Collegiate Athletic Assn. v. Tarkanian, 488 U.S. 179, 191 (1988). And state actors are entitled to a presumption that their actions turn on constitutionally legitimate motivations rather than impermissible animus. Schilb v. Kuebel, 404 U.S. 357, 364 (1971). Of course, this presumption can be defeated, and a widespread history of state action that reflects animus or stereotyping gives courts good reason to be suspicious of the government’s motives. But because we presume that state actors abide by the Constitution, the fact of private discrimination—which is not itself unconstitutional, even if morally blameworthy—does not provide a basis for inferring that state actors are also likely to discriminate and thereby violate the Constitution.

This focus on de jure discrimination is not only theoretically sound—it is also judicially manageable. Courts are ill suited to conduct an open-ended inquiry into whether the volume of private discrimination exceeds some indeterminate threshold. By contrast, they are well equipped to analyze whether there is a history of legislation that has discriminated against the group in question.

Focusing the inquiry on de jure state action would also clarify the test for political powerlessness, which is another factor we have used to determine whether a classification is suspect. Carolene Products, the source of the “discrete and insular minority” test, equates political powerlessness with laws burdening those who lacked a vote. See 304 U. S., at 152–153, n. 4 (citing McCulloch v. Maryland, 4 Wheat. 316, 428 (1819) (a State regulating the Federal Government); South Carolina Highway Dept. v. Barnwell Brothers, Inc., 303 U.S. 177, 184, n. 2 (1938) (a State regulating out-of-state corporations)). This kind of “political powerlessness,” which leaves the affected persons altogether unable to protect themselves in the political process, tracks the experience of the existing suspect classes.

We have said little, however, about what “political powerlessness” means for our recognition of new suspect classes. See Lyng, 477 U. S., at 638 (stating without elaboration that close relatives are not “politically powerless”); Murgia, 427 U. S., at 313 (same for the elderly). And in the absence of clear guidance from us, lower courts have resorted to considering evidence like whether the group has drawn the support of powerful interest groups, achieved equal representation in government, or obtained affirmative statutory protection from discrimination in the private sector. See, e.g., 83 F. 4th, at 487 (evaluating whether transgender litigants are supported by “major medical organizations” and “large law firms”); 679 F. Supp. 3d, at 691 (suggesting that the analysis turns on whether the group has “achiev[ed] relatively equal representation in political bodies”); Grimm v. Gloucester Cty. School Bd., 972 F.3d 586, 613 (CA4 2020) (concluding that transgender people are politically powerless because of a “dearth of openly transgender persons serving in the executive and legislative branches” or in the judiciary). These markers reflect sociological intuitions about a group’s relative political power; they do not constitute an objective, legally grounded standard that courts can apply consistently. A legacy of de jure discrimination, by contrast, more precisely (and objectively) captures the interests that lie at the heart of the Equal Protection Clause.

Because the litigants assumed that evidence of private discrimination could suffice for the suspect-class inquiry, they did not thoroughly discuss whether transgender individuals have suffered a history of de jure discrimination as a class. And because the group of transgender individuals is an insufficiently discrete and insular minority, the question is largely academic.[5] In future cases, however, I would not recognize a new suspect class absent a demonstrated history of de jure discrimination.

*  *  *

The Equal Protection Clause does not demand heightened judicial scrutiny of laws that classify based on transgender status. Rational-basis review applies, which means that courts must give legislatures flexibility to make policy in this area.

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Justice Sotomayor, with whom Justice Jackson joins, and with whom Justice Kagan joins as to all but Part V, dissenting.

To give meaning to our Constitution’s bedrock equal protection guarantee, this Court has long subjected to heightened judicial scrutiny any law that treats people differently based on sex. See United States v. Virginia, 518 U.S. 515, 533 (1996). If a State seeks to differentiate on that basis, it must show that the sex classification “serves important governmental objectives and that the discriminatory means employed are substantially related to the achievement of those objectives.” Ibid. (internal quotation marks omitted). Such review (known as intermediate scrutiny) allows courts to ascertain whether the State has a sound, evidence-based reason to distinguish on the basis of sex or whether it does so in reliance on impermissible stereotypes about the sexes.

Today, the Court considers a Tennessee law that categorically prohibits doctors from prescribing certain medications to adolescents if (and only if) they will help a patient “identify with, or live as, a purported identity inconsistent with the minor’s sex.” Tenn. Code Ann. §68–33–103(a)(1)(A) (2023). In addition to discriminating against transgender adolescents, who by definition “identify with” an identity “inconsistent” with their sex, that law conditions the availability of medications on a patient’s sex. Male (but not female) adolescents can receive medicines that help them look like boys, and female (but not male) adolescents can receive medicines that help them look like girls.

Tennessee’s law expressly classifies on the basis of sex and transgender status, so the Constitution and settled precedent require the Court to subject it to intermediate scrutiny. The majority contorts logic and precedent to say otherwise, inexplicably declaring it must uphold Tennessee’s categorical ban on lifesaving medical treatment so long as “ ‘any reasonably conceivable state of facts’ ” might justify it. Ante, at 21. Thus, the majority subjects a law that plainly discriminates on the basis of sex to mere rational-basis review. By retreating from meaningful judicial review exactly where it matters most, the Court abandons transgender children and their families to political whims. In sadness, I dissent.

I

A

Begin with the medical context in which Tennessee’s law operates. See Tenn. Code Ann. §68–33–101 et seq.; see also S. B. 1, 113th Gen. Assem., 1st Extra. Sess. (2023) (SB1). Doctors in the United States prescribe hormones and puberty inhibitors to treat a range of medical conditions. Often, they are administered to help minors conform to the typical appearance associated with their sex identified at birth. Children who start experiencing puberty at a premature age (precocious puberty), for example, have long received puberty-delaying medications to stave off puberty until adolescence. See App. 22. Adolescent boys might also receive the hormone testosterone to initiate puberty delayed beyond its typical start. App. to Pet. for Cert. 266a. Without testosterone, puberty would “eventually initiate naturally” in most patients, but medication “is often prescribed to avoid some of the social stigma that comes from undergoing puberty later than one’s peers.” Ibid. Adolescent females with delayed puberty may receive the hormone estrogen for the same reason. Ibid.

After puberty begins, doctors may prescribe these same medicines to adolescents whose physical appearance does not align with what one might expect from their sex identified at birth. An adolescent female, for example, might receive testosterone suppressors and hormonal birth control to reduce the growth of unwanted hair on her face or body (sometimes called male-pattern hair growth or hirsutism). See ibid.; see also App. 100 (“[M]edications that are used to suppress testosterone can be used to address symptoms of polycystic ovarian syndrome, which can include unwanted facial hair and body hair, excessive sweating, and body odor”); Brief for Experts on Gender Affirming Care as Amici Curiae 12 (describing the prevalence of hirsutism in people identified as female at birth).[1] An adolescent male may also receive hormones to address a benign but atypical increase in breast gland tissue (known as gynecomastia), sometimes resulting from below-average testosterone levels. See, e.g., G. Kanakis et al., EAA Clinical Practice Guidelines—Gynecomastia Evaluation and Management, 7 Andrology 778, 779–780 (2019). Like any medical treatment, hormones and puberty blockers come with the potential for side effects. See, e.g., App. to Pet. for Cert. 266a–267a; App. 970–974; Brief for United States 45–46. Yet patients and their parents may decide to proceed with treatment on the advice of a physician, despite the accompanying medical risks.

Physicians prescribe these same medications to transgender adolescents, whose gender identity is inconsistent with their sex identified at birth. Hormones and puberty blockers help align transgender adolescents’ physical appearance with their gender identity, as they do when prescribed to adolescents who want to align their appearances with their sex identified at birth. The same puberty suppressants prescribed to pause the onset of precocious puberty can pause puberty for transgender adolescents, giving them “time to further understand their gender identity.” App. to Pet. for Cert. 256a.

Hormone therapy later allows transgender teens to initiate puberty consistent with their gender identity. That typically involves testosterone for adolescent transgender boys (who were identified as female at birth) and testosterone suppression and estrogen for adolescent transgender girls (who were identified as male at birth). Such treatments help adolescents identified as female at birth look more masculine and those identified as male at birth look more feminine. As is true for most medical treatment for minors, puberty blockers and hormones should be administered only after a comprehensive and individualized risk-benefit assessment, and with parental consent. See American Medical Association, Code of Medical Ethics, 2.2.1 Pediatric Decision Making (2022); E. Coleman et al., Standards of Care for the Health of Transgender and Gender Diverse People, Version 8, 23 Int’l J. Transgender Health S1, S58 (2022).[2]

Transgender adolescents’ access to hormones and puberty blockers (known as gender-affirming care) is not a matter of mere cosmetic preference. To the contrary, access to care can be a question of life or death. Some transgender adolescents suffer from gender dysphoria, a medical condition characterized by clinically significant and persistent distress resulting from incongruence between a person’s gender identity and sex identified at birth. App. to Pet. for Cert. 251a–252a. If left untreated, gender dysphoria can lead to severe anxiety, depression, eating disorders, substance abuse, self-harm, and suicidality. See, e.g., Coleman, 23 Int’l J. Transgender Health, at S62. Suicide, in particular, is a major concern for parents of transgender teenagers, as the lifetime prevalence of suicide attempts among transgender individuals may be as high as 40%. App. to Pet. for Cert. 264a. Tragically, studies suggest that as many as one-third of transgender high school students attempt suicide in any given year.[3]

When provided in appropriate cases, gender-affirming medical care can meaningfully improve the health and wellbeing of transgender adolescents, reducing anxiety, depression, suicidal ideation, and (for some patients) the need for more invasive surgical treatments later in life.[4] That is why the American Academy of Pediatrics, American Medical Association, American Psychiatric Association, American Psychological Association, and American Academy of Child Adolescent Psychiatry all agree that hormones and puberty blockers are “appropriate and medically necessary” to treat gender dysphoria when clinically indicated. Id., at 285a.[5]

B

Tennessee has taken a different tack. The State enacted SB1 to categorically prohibit physicians from prescribing puberty blockers and hormone therapy for the purpose of treating gender dysphoria in minors. Tennessee’s blanket ban applies only when hormones and puberty blockers are prescribed to “[e]nabl[e] a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex” or to alleviate “discomfort or distress from a discordance between the minor’s sex and asserted identity.” Tenn. Code Ann. §68–33–103(a)(1). SB1 leaves untouched the use of the same drugs to treat any other medical condition, including delayed (or early) puberty and any other “physical or chemical abnormality present in a minor that is inconsistent with the normal development of a human being of the minor’s sex.” §68–33–102(1). In other words, SB1 allows physicians to help align adolescents’ physical appearance with their gender identity (despite associated risks) if it is consistent with their sex identified at birth, but not if inconsistent. Indeed, Tennessee’s stated interests in SB1 include “encouraging minors to appreciate their sex.” §68–33–101(m).

C

Tennessee’s ban applies no matter what the minor’s parents and doctors think, with no regard for the severity of the minor’s mental health conditions or the extent to which treatment is medically necessary for an individual child. The stories of the plaintiffs in this case reflect the stakes of that harsh reality.

Ryan Roe, now 16, felt as early as elementary school that he “was a boy.” App. to Pet. for Cert. 234a. Before puberty, Ryan thought “there wasn’t that much of a difference between boys and girls” and that he “could manage existing in the middle.” Ibid. As puberty approached, however, Ryan grew increasingly anxious about the impending changes to his body. He started throwing up every morning before school. As his voice changed, Ryan contemplated going mute. Id., at 235a. Eventually, after two years of psychotherapy and extensive consultations with his parents and doctors, Ryan’s physicians prescribed him testosterone. Ryan began to find his voice again. He started raising his hand in class, participating in school, and looking at himself in the mirror. Ryan attests that “[g]ender-affirming health care saved [his] life.” Id., at 234a. For Ryan’s parents, “[i]t is simply not an option to cut [him] off from this care.” Id., at 246a. “I worry about his ability to survive,” Ryan’s mother attests. “[L]osing him would break me.” Ibid.

L. W., too, began to question her gender as early as fourth grade. At the time, she felt like she was “drowning” and “trapped in the wrong body,” often sick at school because she “did not feel comfortable using the boy’s bathroom.” Id., at 223a. At age 13, L. W. and her parents sought out medical treatment. Puberty blockers and estrogen, prescribed to L. W. after consultation with her parents and doctors, changed her life. “We have a confident, happy daughter now, who is free to be herself,” her mom explains. App. 85. “As a mother, I could not bear watching my child go through physical changes that would destroy her well-being and cause her life-long pain.” Id., at 86.

Echoing a similar refrain, John Doe and his family attest that John felt from an early age he was a boy. He chose a male name for himself around the age of three. As puberty approached, John grew terrified of undergoing what he saw as “the wrong puberty,” recognizing that “some of those changes could be permanent.” App. to Pet. for Cert. 232a. After years of psychotherapy, he began taking puberty-delaying medication. His mother, who “shed many tears during the first year” of this process, acknowledges that “John’s gender transition has not been easy.” App. 95. Yet she attests that John’s access to medical treatment is “the one thing” that gives her hope that he can “have a fulfilling life.” Id., at 94.

D

Faced with the choice between leaving Tennessee in search of treatment and risking their children’s lives, Ryan, John, L. W., and their parents sued to enjoin SB1. The United States intervened in support.[6] Together, they argued that SB1 unconstitutionally discriminates on the basis of sex and transgender status. After review of the factual record, the District Court agreed, holding that the law would likely fail intermediate scrutiny because its targeted ban on promoting inconsistency with sex was not substantially related to Tennessee’s asserted interest in protecting minors from dangerous medical procedures. L. W. v. Skrmetti, 679 F. Supp. 3d 668, 710 (MD Tenn. 2023).

A divided panel of the Sixth Circuit reversed. All three judges appeared to “accept the premise” that “the statut[e] treat[s] minors differently based on sex.” L. W. v. Skrmetti, 83 F. 4th 460, 481 (2023); see also id., at 484 (“[T]he necessity of heightened review . . . will not be present every time that sex factors into a government decision”). Yet the majority refused to apply intermediate scrutiny because it believed that the law did not necessarily “disadvantage ‘persons’ based on their sex.” Id., at 483. Because the Sixth Circuit never applied intermediate scrutiny to SB1, the only question this Court must decide is whether the Constitution required it to do so.

II

A

The level of constitutional scrutiny courts apply in reviewing state action is enormously consequential. Where a state law neither “proceeds along suspect lines nor infringes fundamental constitutional rights,” reviewing courts generally uphold a challenged law under the Equal Protection Clause so long as “any reasonably conceivable state of facts . . . could provide a rational basis for the classification.” FCC v. Beach Communications, Inc., 508 U.S. 307, 313 (1993). That lenient standard, which the majority erroneously applies today, demands hardly more than a cursory glance at the State’s reasons for legislating.

This Court has long recognized, however, that a more “searching” judicial review is warranted when the rights of “discrete and insular minorities” are at stake. United States v. Carolene Products Co., 304 U.S. 144, 153, n. 4 (1938). Because such minorities often face systemic barriers to vindicating their interests through the political process, courts have a comparative advantage over the elected branches in safeguarding their rights. Ibid. Such judicial scrutiny is at its apex in reviewing laws that classify on the basis of race and national origin. States may not enact laws that classify on those bases unless they can pass through the “daunting two-step examination known in our cases as ‘strict scrutiny.’ ” Students for Fair Admissions, Inc. v. President and Fellows of Harvard College, 600 U.S. 181, 206 (2023); see id., at 206–207 (“Under that standard we ask . . . whether the racial classification is used to ‘further compelling governmental interests’ ” and then “whether the government’s use of race is ‘narrowly tailored—meaning ‘necessary’—to achieve that interest”).

For nearly half a century, the Court has applied a different standard, known as intermediate scrutiny, to all “statutory classifications that distinguish between males and females.” Nevada Dept. of Human Resources v. Hibbs, 538 U.S. 721, 728 (2003); see Craig v. Boren, 429 U.S. 190, 197–199 (1976). States can differentiate on the basis of sex only to “ ‘serv[e] important governmental objectives’ ” and only if the sex classification is “ ‘substantially related to the achievement of those objectives.’ ” Hibbs, 538 U. S., at 728. The standard is an intermediate one because it strikes an important balance. On the one hand, there are some genuine “[p]hysical differences between men and women,” so not all sex-based legislation is discriminatory or constitutionally proscribed. Virginia, 518 U. S., at 533. On the other hand, sex-based legislation always presents a serious risk of invidious discrimination that relies on “overbroad generalizations about the different talents, capacities, or preferences of males or females.” Ibid. Intermediate scrutiny is the core judicial tool to differentiate innocuous sex-based laws from discriminatory ones.

B

SB1 plainly classifies on the basis of sex, so the Constitution demands intermediate scrutiny. Recall that SB1 prohibits the prescription of hormone therapy and puberty blockers only if done to “enable a minor to identify with, or live as, a purported identity inconsistent with the minor’s sex” or to alleviate “discomfort or distress from a discordance between the minor’s sex and asserted identity.” Tenn. Code Ann. §68–33–103(a)(1). Use of the same drugs to treat any other “ ‘disease’ ” is unaffected. §68–33–103(b)(1)(A). Physicians may continue, for example, to prescribe hormones and puberty blockers to treat any “physical or chemical abnormality present in a minor that is inconsistent with the normal development of a human being of the minor’s sex.” §68–33–102(1).

What does that mean in practice? Simply that sex determines access to the covered medication. Physicians in Tennessee can prescribe hormones and puberty blockers to help a male child, but not a female child, look more like a boy; and to help a female child, but not a male child, look more like a girl. Put in the statute’s own terms, doctors can facilitate consistency between an adolescent’s physical appearance and the “normal development” of her sex identified at birth, but they may not use the same medications to facilitate “inconsisten[cy]” with sex. All this, the State openly admits, in service of “encouraging minors to appreciate their sex.” §68–33–101(m).

Like any other statute that turns on inconsistency with a protected characteristic, SB1 plainly classifies on the basis of sex. A simple analogy illustrates the point. Suppose Tennessee prohibited minors from attending “ ‘any services, rituals, or assemblies if done for the purpose of allowing the minor to identify with a purported identity inconsistent with the minor’s religion.’ ” Brief for Yale Philosophers as Amici Curiae 10. No one would seriously dispute that such a rule classifies on the basis of religion. Whether the law prohibits a minor from attending any particular religious service turns on the minor’s religion: A Jewish child can visit a synagogue but not a church, while a Christian child can attend church but not the synagogue.

SB1 operates in the same way. Consider the mother who contacts a Tennessee doctor, concerned that her adolescent child has begun growing unwanted facial hair. This hair growth, the mother reports, has spurred significant distress because it makes her child look unduly masculine. The doctor’s next step depends on the adolescent’s sex. If the patient was identified as female at birth, SB1 allows the physician to alleviate her distress with testosterone suppressants. See App. to Pet. for Cert. 266a (describing such treatments); App. 100 (same). What if the adolescent was identified male at birth, however? SB1 precludes the patient from receiving the same medicine.

Now consider the parents who tell a Tennessee pediatrician that their teenage child has been experiencing an unwanted (but medically benign) buildup of breast gland tissue. See supra, at 3. Again, the pediatrician’s next move depends on the patient’s sex. Identified male at birth? SB1 allows the physician to prescribe hormones to reduce the buildup of such tissue. Yet a child identified as female at birth experiencing the same (or more) distress must be denied the same prescription. In both scenarios, SB1 “provides that different treatment be accorded to [persons] on the basis of their sex,” and therefore necessarily “establishes a classification subject to scrutiny under the Equal Protection Clause.” Reed v. Reed, 404 U.S. 71, 75 (1971).[7] The Sixth Circuit apparently agreed. 83 F. 4th, at 481 (accepting the premise that “the statut[e] treat[s] minors differently based on sex”).

Tennessee, too, essentially concedes the point. It admits that a prohibition on wearing clothing “ ‘inconsistent with’ ” the wearer’s sex would trigger intermediate scrutiny, as would a law prohibiting professionals from working in jobs “ ‘inconsistent with’ ” their sex. Brief for Respondents 25. That is because for some jobs and some outfits, “a male can have the job” or wear the outfit, “and a female cannot.” Ibid. SB1 draws exactly the same kind of sex-based line: For some treatments that help adolescents look and feel more masculine, a male minor can have the treatment, and a female minor cannot.[8]

That SB1 conditions a patient’s access to treatment even in part on her sex is enough to trigger intermediate scrutiny. This Court’s equal protection precedents ask only whether a law “differentiates on the basis of gender.” Sessions v. Morales-Santana, 582 U.S. 47, 58 (2017). If so, the law “attract[s] heightened review under the Constitution’s equal protection guarantee.” Ibid. A long line of this Court’s equal protection precedents confirms that much. See Hibbs, 538 U. S., at 728 (“[S]tatutory classifications that distinguish between males and females are subject to heightened scrutiny”); Virginia, 518 U. S., at 531 (“Parties who seek to defend gender-based government action must demonstrate an ‘exceedingly persuasive justification’ for that action”); J. E. B. v. Alabama ex rel. T. B., 511 U.S. 127, 136 (1994) (“[A]ll gender-based classifications today” “warran[t] . . . heightened scrutiny”). That is why an Alabama statute that “authoriz[es] the imposition of alimony obligations on husbands, but not on wives,” “ ‘establishes a classification subject to scrutiny under the Equal Protection Clause’ ”: The plaintiff, “Mr. Orr[,] bears a burden he would not bear were he female.” Orr v. Orr, 440 U.S. 268, 273, 278 (1979).

This Court’s decision in Bostock v. Clayton County, 590 U.S. 644 (2020), confirms the classification on SB1’s face. As Bostock explained in the context of Title VII’s prohibition on employment discrimination, “it is impossible to discriminate against a person for being homosexual or transgender without discriminating against that individual based on sex.” Id., at 660. In deciding that discrimination based on incongruence between sex and gender identity was discrimination “because of sex,” Bostock asked the very same question our equal protection precedents do: whether “changing the employee’s sex would have yielded a different choice by the employer.” Id., at 659–660; cf. Students for Fair Admissions, Inc., 600 U. S., at 231 (applying strict scrutiny to government actions that treat people differently “on the basis of race”).[9] The answer was clearly yes, for the simple reason that discrimination against transgender employees necessarily “penalizes a person identified as male at birth for traits or actions that it tolerates in an employee identified as female at birth.” Bostock, 590 U. S., at 660. Nor was it a defense to liability that the discrimination might apply equally to both sexes: “[A]n employer who fires a woman, Hannah, because she is insufficiently feminine and also fires a man, Bob, for being insufficiently masculine” in both cases “fires an individual in part because of sex.” Id., at 659. The same is true of SB1. By depriving adolescents of hormones and puberty blockers only when such treatment is “inconsistent with” a minor’s sex, the law necessarily deprives minors identified as male at birth of the same treatment it tolerates for an adolescent identified as female at birth (and vice versa).

III

Notwithstanding that SB1 distinguishes between males and females in the medical treatments it authorizes, the Sixth Circuit declined to apply intermediate scrutiny. It believed SB1’s treatment of both sexes to be “evenhande[d],” 83 F. 4th, at 479, meaning (in the panel’s judgment) the classifications were not “invidious” or “unfai[r].” Id., at 483–484. Intermediate scrutiny, of course, is how this Court determines whether a particular sex-based classification is invidious or unfair. See, e.g., Virginia, 518 U. S., at 531. The Sixth Circuit thus effectively held that intermediate scrutiny did not apply to SB1 because it thought SB1 might well pass such scrutiny. Even the majority today does not endorse this circular approach.[10]

Though it skirts the Sixth Circuit’s error, the majority rests its conclusion on an equally implausible ground: that SB1’s prohibition on treatments “inconsistent with [a] minor’s sex” contains no sex classification at all. Tenn. Code Ann. §68–33–103(a)(1). As the statute’s text itself makes clear, that conclusion is indefensible.

A

How does the majority wriggle itself (and the Sixth Circuit) free of any obligation to take a closer look? It abstracts away the sex classification on SB1’s face, asserting that the law classifies based only on “age” and “medical purpose.” The theory, apparently, is that SB1 is sex neutral because it simply allows doctors to “administer puberty blockers or hormones to minors to treat certain conditions but not to treat gender dysphoria.” Ante, at 9. Unlike a law that prohibits attendance at a religious service “inconsistent with” the attendee’s religion, the majority says, “[a] law prohibiting the administration of specific drugs for particular medical uses” simply does not trigger heightened scrutiny. Ante, at 14.

The problem with the majority’s argument is that the very “medical purpose” SB1 prohibits is defined by reference to the patient’s sex. Key to whether a minor may receive puberty blockers or hormones is whether the treatment facilitates the “medical purpose” of helping the minor live or appear “inconsistent with” the minor’s sex. That is why changing a patient’s sex yields different outcomes under SB1. Again, take the adolescent distressed by newly developing facial hair. Was the patient identified female at birth? SB1 authorizes the prescription of medication. Male at birth? SB1 prohibits it.

For truly sex-neutral laws, it is impossible to imagine a single scenario where changing a patient’s sex yields a different result. To borrow from the majority’s catalog of apparently benign medical-use distinctions, imagine Tennessee allowed consumption of DayQuil to ease coughs, but not minor aches and pains. See ante, at 12. The regulated medical purposes (treatment of coughs, aches, and pains) are unrelated to sex, so a patient’s sex will never determine whether she can consume DayQuil. All that matters is whether the patient has a cough.

So too for New York’s ban on assisted suicide, which the majority equates to SB1. Ante, at 10. In Vacco v. Quill, 521 U.S. 793 (1997), this Court subjected the assisted-suicide ban to rational-basis review because it neither “treat[ed] anyone differently from anyone else” nor “dr[ew] any distinctions between persons.” Id., at 800. In New York, the Court explained, “[e]veryone” can “refuse unwanted lifesaving medical treatment” and “no one is permitted to assist a suicide.” Ibid. Yet unlike for SB1, neither sex nor any other protected characteristic distinguished the terminally ill patient who could permissibly “ ‘hasten death’ ” from another prohibited from doing so. Id., at 800–801. All that mattered was the patient’s existing connection to life-support systems: Those connected could lawfully hasten death by discontinuing treatment, while others (who required a prescription for lethal medication to do so) could not. The patient’s sex (or race, or national origin) would never decide the outcome. SB1, by contrast, renders every treatment decision it regulates dependent on two things: a minor’s sex identified at birth, and the consistency of the requested treatment with that sex.

That the majority finds a way to recast SB1 in sex- neutral terms is no evidence that SB1 is sex neutral in the way hypothetical prohibitions on DayQuil or assisted suicide would be. Contra, ante, at 14. The majority emphasizes that, in Tennessee, “no minor may be administered puberty blockers or hormones to treat gender dysphoria,” while “minors of any sex may be administered puberty blockers or hormones for other purposes.” Ante, at 13. But nearly every discriminatory law is susceptible to a similarly race- or sex-neutral characterization. A prohibition on interracial marriage, for example, allows no person to marry someone outside of her race, while allowing persons of any race to marry within their races. See Loving v. Virginia, 388 U.S. 1, 9 (1967).[11] The same is true of a hypothetical law prohibiting attendance at services “inconsistent with” a child’s religion, while allowing all children to attend religion-consistent services. See supra, at 11. Indeed, the majority itself seems to recognize that laws prohibiting professions “inconsistent” with a person’s sex, marriages “inconsistent” with a person’s race, or religious services “inconsistent” with a person’s faith must be subject to heightened review, even if rewritten as ostensibly neutral prohibitions on sex-, race-, and faith-inconsistent behavior. See ante, at 13–14. And although the majority insists that its logic would not apply to the hypothetical religion-consistent services law, ante, at 14, it offers no principled reason to differentiate that law from SB1’s prohibition on promoting “inconsisten[cy] with” the patient’s sex.

B

Recognizing, perhaps, that this Court already decided in Bostock that discrimination based on incongruence between sex and gender identity was itself discrimination “because of sex,” the majority seeks to distinguish Bostock away. Unlike in Bostock, the majority urges, “changing a minor’s sex or transgender status does not alter the application of SB1.” Ante, at 19. Again, it emphasizes that no “medical treatment” under SB1 is actually doled out on the basis of sex, because (it says) medical “treatment” necessarily encompasses “both a given drug and the specific indication for which it is being administered.” Ante, at 12–13, 18–19. The majority’s logic is as follows: “If a transgender boy [who was identified as female at birth] seeks testosterone to treat his gender dysphoria, SB1 prevents a healthcare provider from administering it to him.” Ante, at 19. “If you change his biological sex from female to male,” the majority says, “SB1 would still not permit him the hormones he seeks because he would lack a qualifying diagnosis for the testosterone—such as a congenital defect, precocious puberty, disease, or physical injury.” Ibid.

As should be abundantly clear by this point, the majority’s recharacterization of SB1 is impossible to reconcile with the statute’s plain terms. SB1 allows physicians to prescribe hormones and puberty blockers to treat not just some defined category of cancers and rashes, but any “physical or chemical abnormality present in a minor that is inconsistent with the normal development of a human being of the minor’s sex.” §68–33–102(1). If a minor has some physical “abnormality” (say, medically benign facial hair) typically perceived as “inconsistent” with her sex identified at birth (female), SB1 deems it a “congenital defect” that physicians can treat. Change the patient’s sex from female to male, and the law now forbids providing the same drugs to rid the minor of the same facial hair. In other words, SB1 makes explicit that the very reason why a doctor can treat an adolescent female for “hirsutism (male-pattern hair growth),” but not gender dysphoria is that the former will promote consistency with sex, while the latter does the opposite. Cf. ante, at 20 (majority opinion). As was true in Bostock, then, the law deprives minors of medical treatment based, in part, on sex.

To be sure, when the hypothetical minor is male, not female, the patient’s diagnosis may well change too: The female adolescent distressed by facial hair might receive a diagnosis of hirsutism while the male adolescent may be diagnosed with gender dysphoria. See supra, at 3, 11; see also ante, at 20 (majority opinion). The same, however, was true in Bostock. When an employer fires an employee because she is transgender, the Court explained, “two causal factors may be in play”: the individual’s sex and the sex “with which the individual identifies.” 590 U. S., at 661. Yet so long as the plaintiff ’s sex is “one but-for cause of that decision,” the employer discriminates on the basis of sex. Id., at 656. So too with SB1. Sex and diagnosis may both “be in play.” Id., at 661. As long as sex is one of the law’s distinguishing features, however, the law classifies on the basis of sex, and the Equal Protection Clause requires application of intermediate scrutiny.

C

In a final bid to avoid applying our equal protection precedents, the majority asserts that “mere reference to sex” is insufficient to trigger intermediate scrutiny, especially in the “medical context.” Ante, at 10. Of course, not every legislative mention of sex triggers intermediate scrutiny. A law mandating that no person, “regardless of sex,” can consume a dangerous drug, for example, would be subject to rational-basis review. Yet SB1 does not just mention sex. It defines an entire category of prohibited conduct based on inconsistency with sex. And it is hard to imagine a law that prohibits conduct “inconsistent with” sex that could avoid intermediate scrutiny.

Nor does the fact that SB1 concerns the “medical context” change the relevant analysis. Ibid. No one disputes that “[s]ome medical treatments and procedures are uniquely bound up in sex” or that there are “biological differences between men and women.” Ibid. That there are such physical differences is, after all, one of the reasons why sex is not altogether a proscribed classification. See Virginia, 518 U. S., at 533. A law that allowed only women to receive certain breast cancer treatments, for example, might well be consistent with the Constitution’s equal protection mandate if the State establishes that the relevant treatments are suited to women’s (and not men’s) bodies. Cf. ante, at 11 (noting “ ‘many’ breast cancer treatments [are] approved for women only”). Laws that differentiate based on biological distinctions between men and women are precisely the sort that States might successfully defend under intermediate scrutiny. Biological differences between the sexes, however, are no reason to skirt such scrutiny altogether.

Fashioning a medical-context-only exception also runs counter to decades of equal protection precedents. This Court has clarified that, although not every sex-based distinction is “marked by misconception and prejudice,” Tuan Anh Nguyen v. INS, 533 U.S. 53, 73 (2001), every sex-based distinction does warrant intermediate scrutiny. See J. E. B., 511 U. S., at 136 (“[A]ll gender-based classifications today” “warran[t] . . . heightened scrutiny” (emphasis added)).

Take, for example, Tuan Anh Nguyen, where this Court assessed the constitutionality of a law imposing one set of citizenship-acquisition requirements on children born abroad out of wedlock to U. S. citizen mothers and another on those born of U. S. citizen fathers. 533 U. S., at 60. The Court ultimately decided that the “different set of rules” for fathers and mothers was “neither surprising nor troublesome from a constitutional perspective” because “[f]athers and mothers are not similarly situated with regard to the proof of biological parenthood.” Id., at 63. We reached that conclusion, however, only after demanding of the Government an explanation for why that sex classification “ ‘serve[d] “important governmental objectives” ’ ” and how “ ‘ “the discriminatory means employed” [were] “substantially related to the achievement of those objectives.” ’ ” Id., at 60 (quoting Virginia, 518 U. S., at 533). In no sense did the biological differences between the sexes relieve courts of the obligation to examine the sex classification with a careful constitutional eye. Nor is any medical-context exception necessary because intermediate scrutiny itself allows the State to maintain classifications where justified by biology.

IV

Having blithely dispensed with the notion that SB1 classifies on the basis of sex, the majority next asserts that “SB1 does not classify on the basis of transgender status.” Ante, at 16. That too is contrary to the statute’s text and plainly wrong.

SB1 prohibits Tennessee physicians from offering hormones and puberty blockers to allow a minor to “identify with” a gender identity inconsistent with her sex. Tenn. Code Ann. §68–33–103(a)(1)(A). Desiring to “identify with” a gender identity inconsistent with sex is, of course, exactly what it means to be transgender. The two are wholly coextensive. See Oxford English Dictionary (3d ed., Dec. 2023), https://www.oed.com/dictionary/transgender_adj (Trans-gender, when used as an adjective, means “a person whose sense of personal identity and gender does not correspond to that person’s sex at birth . . . ”). That is why it would defy common sense to suggest an employer’s policy of firing all persons identifying with or living as an identity inconsistent with their sex does not discriminate on the basis of transgender status.

Left with nowhere else to turn, the Court hinges its conclusion to the contrary on the by-now infamous footnote 20 of Geduldig v. Aiello, 417 U.S. 484 (1974), which declared that discrimination on the basis of pregnancy is not discrimination on the basis of sex. See id., at 496–497, n. 20. The footnote reasoned that, although “only women can become pregnant,” “[n]ormal pregnancy is an objectively identifiable physical condition with unique characteristics” and “lawmakers are constitutionally free to include or exclude pregnancy from the coverage of legislation . . . on any reasonable basis, just as with respect to any other physical condition.” Ibid. The takeaway, according to the majority, is that “not . . . every legislative classification concerning pregnancy is a sex-based classification,” and so (apparently) not every legislative classification concerning “gender incongruence” (at least in the context of medical treatments) classifies on the basis of transgender status. Id., at 496, n. 20.

Geduldig was “egregiously wrong” when it was decided, both “[b]ecause pregnancy discrimination is inevitably sex discrimination” and because discrimination against women is so “tightly interwoven with society’s beliefs about pregnancy and motherhood.” Coleman v. Court of Appeals of Md., 566 U.S. 30, 56–57 (2012) (Ginsburg, J., dissenting). That the majority must resuscitate so unpersuasive a source, widely rejected as indefensible even 40 years ago, is itself a telling sign of the weakness of its position. See S. Law, Rethinking Sex and the Constitution, 132 U. Pa. L. Rev. 955, 983 (1984) (“Criticizing Geduldig has . . . become a cottage industry”). That the Court today extends Geduldig’s logic for the first time beyond pregnancy and abortion is more troubling still. Divorced from its fact-specific context, Geduldig’s reasoning may well suggest that a law depriving all individuals who “have ever, or may someday, menstruate” of access to health insurance would be sex neutral merely because not all women menstruate.

In any event, even Geduldig’s faulty reasoning cannot save the majority’s conclusion that SB1 is innocent of transgender discrimination. Unlike pregnancy, a desire to “identify with, or live as, a purported identity inconsistent with [one’s] sex,” Tenn. Code Ann. §68–33–103(a)(1)(A), is not some “objectively identifiable physical condition” that legislatures can target without reference to sex or transgender status, Geduldig, 417 U. S., at 496, n. 20. And while not all women are pregnant, ibid., all transgender people, by definition, “identify with, or live as, a purported identity inconsistent with [their] sex,” Tenn. Code Ann. §68–33–103(a)(1)(A). So, unlike the classes of pregnant persons and women, the class of minors potentially affected by SB1 and transgender minors are one and the same.

That SB1 discriminates on the basis of transgender status is yet another reason it must be subject to heightened scrutiny. For one, this Court already decided in Bostock that “it is impossible to discriminate against a person for being . . . transgender without discriminating against that individual based on sex,” 590 U. S., at 660, and sex discrimination is of course subject to heightened scrutiny. Nor should there be serious dispute that transgender persons bear the hallmarks of a quasi-suspect class.[12] See Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 441 (1985) (describing the standard).

Transgender people have long been subject to discrimination in healthcare, employment, and housing, and to rampant harassment and physical violence. See Grimm v. Gloucester Cty. School Bd., 972 F.3d 586, 611 (CA4 2020) (detailing that history); see also K. Barry, B. Farrell, J. Levi, & N. Vanguri, A Bare Desire To Harm: Transgender People and the Equal Protection Clause, 57 B. C. L. Rev. 507, 556–557 (2016) (describing Congress’s exclusion of transgender people from the Fair Housing Act, Americans with Disabilities Act, and Rehabilitation Act). Individuals whose gender identity diverges from their sex identified at birth (whether labeled as “transgender” at the time or not), moreover, have been subject to a lengthy history of de jure discrimination in the form of cross-dressing bans, police brutality, and anti-sodomy laws. See, e.g., K. Redburn, Before Equal Protection: The Fall of Cross-Dressing Bans and the Transgender Legal Movement, 1963–86, 40 L. and Hist. Rev. 679, 685, 687 (2022); A. Lvovsky, Vice Patrol 29, 108 (2021); W. Eskridge, GayLaw: Challenging the Apartheid of the Closet 328–337 (1999) (cataloging state consensual sodomy laws, 1610–1988). Beginning in 1843, cities ranging from “major metropolitan centers such as Chicago and Los Angeles to small cities and towns including Cheyenne, Wyoming and Vermillion, South Dakota” enacted ordinances that (most commonly) criminalized any person “ ‘appear[ing] upon any public street or other public place . . . in a dress not belonging to his or her sex.’ ” Redburn, 40 L. and Hist. Rev., at 687. In any event, those searching for more evidence of de jure discrimination against transgender individuals, see ante, at 7–9 (Barrett, J., concurring), need look no further than the present. The Federal Government, for example, has started expelling transgender servicemembers from the military and threatening to withdraw funding from schools and nonprofits that espouse support for transgender individuals.[13]

Transgender persons, moreover, have a defining characteristic (incongruence between sex and gender identity) that plainly “ ‘bears no relation to [the individual’s] ability to perform or contribute to society.’ ” Cleburne, 473 U. S., at 441. As a group, the class is no more “ ‘large, diverse, and amorphous,’ ” ante, at 4 (opinion of Barrett, J.); ante, at 14 (Alito, J., concurring in part and concurring in judgment), than most races or ethnic groups, many of which similarly include individuals with “ ‘a huge variety’ ” of identities and experiences, ante, at 5 (opinion of Barrett, J.). (Not all racial, ethnic, or religious minorities, for example, “ ‘carry an obvious badge’ of their membership in the disadvantaged class.” Cf. ante, at 16 (opinion of Alito, J.).)[14] As evidenced by the recent rise in discriminatory state and federal policies and the fact that transgender people “are underrepresented in every branch of government,” Grimm, 972 F. 3d, at 611–613, moreover, the class lacks the political power to vindicate its interests before the very legislatures and executive agents actively singling them out for discriminatory treatment. See Lyng v. Castillo, 477 U.S. 635, 638 (1986). In refusing to say as much, the Court today renders transgender Americans doubly vulnerable to state-sanctioned discrimination.[15]

V

SB1’s classifications by sex and transgender status clearly require the application of intermediate scrutiny. The majority’s choice instead to subject SB1 to rational-basis review, the most cursory form of constitutional review, is not only indefensible as a matter of precedent but also extraordinarily consequential. Instead of scrutinizing the legislature’s classifications with an eye towards ferreting out unconstitutional discrimination, the majority declares it will uphold Tennessee’s ban as long as there is “ ‘any reasonably conceivable state of facts that could provide a rational basis for the classification.’ ” Ante, at 21 (quoting Beach Communications, Inc., 508 U. S., at 313; emphasis added). That marks the first time in 50 years that this Court has applied such deferential review, normally employed to assess run-of-the-mill economic regulations, to legislation that explicitly differentiates on the basis of sex. As a result, the Court never even asks whether Tennessee’s sex-based classification imposes the sort of invidious discrimination that the Equal Protection Clause prohibits.

The majority says that it does not want to “second-guess the lines that SB1 draws,” ante, at 22, or to “resolve” disagreements about the safety and efficacy of “medical treatments in an evolving field,” ante, at 24. The concurrences, too, warn that applying intermediate scrutiny in this case may “require courts to oversee all manner of policy choices normally committed to legislative discretion,” including in “areas of legitimate regulatory policy . . . ranging from access to restrooms to eligibility for boys’ and girls’ sports teams.” Ante, at 5, 6 (opinion of Barrett, J.); see also ante, at 4 (Thomas, J., concurring) (highlighting the potential for “ ‘high-cost, high-risk lawsuit[s]’ ”). Looking carefully at a legislature’s proffered reasons for acting, as our equal protection precedents demand, is neither needless “second-guess[ing],” ante, at 22 (majority opinion), nor judicial encroachment on “areas of legitimate regulatory policy,” ante, at 6 (opinion of Barrett, J.). After all, “ ‘closely scrutiniz[ing] legislative choices’ ” is exactly how courts distinguish “legitimate regulatory polic[ies]” from discriminatory ones. Ibid.

Indeed, judicial scrutiny has long played an essential role in guarding against legislative efforts to impose upon individuals the State’s views about how people of a particular sex (or race) should live or look or act. Women, it was once thought, were not suited to attend military schools with men. Virginia, 518 U. S., at 520–523, 540–541. Men and women, others said, should not marry those of a different race. Loving, 388 U. S., at 4. Those laws, too, posed politically fraught and contested questions about race, sex, and biology. In a passage that sounds hauntingly familiar to readers of Tennessee’s brief, Virginia argued in Loving that, should this Court intervene, it would find itself in a “bog of conflicting scientific opinion upon the effects of interracial marriage, and the desirability of preventing such alliances, from the physical, biological, genetic, anthropological, cultural, psychological, and sociological point of view.” Brief for Appellee in Loving v. Virginia, O. T. 1966, No. 395, p. 7. “In such a situation,” Virginia continued, “it is the exclusive province of the Legislature of each State to make the determination for its citizens as to the desirability of a policy of permitting or preventing such [interracial] alliances—a province which the judiciary may not constitutionally invade.” Id., at 7–8.

This Court, famously, rejected the States’ invitation in Loving to “defer to the wisdom of the state legislature” based on assertions that “the scientific evidence is substantially in doubt.” 388 U. S., at 8. In considering the constitutionality of Virginia’s male-only military academy, too, the Court itself assessed the “opinions of Virginia’s expert witnesses” that “ ‘[m]ales tend to need an atmosphere of adversativeness,’ ” while “ ‘[f]emales tend to thrive in a cooperative atmosphere.’ ” 518 U. S., at 541. What the Court once recognized as an imperative check against discrimination, it today abandons.

Yet the task of ascertaining SB1’s constitutionality is a familiar one. Tennessee has proffered an undoubtedly important interest in “protect[ing] the health and welfare of minors” by prohibiting medical procedures that carry “risks and harms.” Tenn. Code Ann. §§68–33–101(a), (b)–(e); see New York v. Ferber, 458 U.S. 747, 756–757 (1982) (States’ “interest in ‘safeguarding the physical and psychological well-being of a minor’ ” is “ ‘compelling’ ”). All, including the Solicitor General, agree that the State may strictly regulate access to cross-sex hormones and puberty blockers to achieve that purpose. See Tr. of Oral Arg. 39–40, 152–153 (agreeing that West Virginia’s more tailored limitations on gender-affirming care would likely survive intermediate scrutiny). It may well be, too, that “[d]eference to legislatures” is “particularly critical” in this context, where the provision of medical care to minors is at issue. Ante, at 22 (opinion of Thomas, J.). But that does not change the Court’s obligation, as mandated by our precedents, to determine whether the challenged sex classification in SB1’s categorical ban is tailored to protecting minors’ health and welfare, or instead rests on unlawful stereotypes about how boys and girls should look and act. See Virginia, 518 U. S., at 533. Infusing that antecedent legal question with a host of evidence relevant only to the subsequent application of judicial scrutiny, as Justice Thomas would have us do, see ante, at 7–22, simply puts the cart before the horse.

The present record offers reason to question (as the District Court did) whether Tennessee’s categorical ban on treating gender dysphoria bears the “requisite direct, substantial relationship” to its interest in protecting minors’ health. Mississippi Univ. for Women v. Hogan, 458 U.S. 718, 725 (1982). Tennessee has offered little evidence, for example, that it is more dangerous to receive puberty blockers to “identify with, or live as, a purported identity inconsistent with the minor’s sex” than to treat other conditions like precocious puberty.[16] Why, then, does SB1 proscribe the regulated medications to treat gender dysphoria, while leaving them available for myriad other purposes? So too is it difficult to ignore that Tennessee professes concern with protecting the health of minors while categorically banning gender-affirming care for even those minors exhibiting the most severe mental-health conditions, including suicidality.

The majority’s choice to avoid applying intermediate scrutiny is all the more puzzling, however, because this Court need not itself resolve these questions or wade into what it dubs the “fierce scientific and policy debates about the safety, efficacy, and propriety of medical treatments in an evolving field.” Ante, at 24. The Sixth Circuit never even asked whether the challenged sex classification in SB1 “serves ‘important governmental objectives’ ” or is “ ‘substantially related to the achievement of those objectives.’ ” Virginia, 518 U. S., at 533. All the United States requested of this Court was confirmation that intermediate scrutiny applied. Brief for United States 32. On remand, the courts could have taken due account of the “[r]ecent developments” that (according to the majority) “underscore the need for legislative flexibility in this area,” including a recent report from England’s National Health Service on the use of puberty blockers and hormones to treat transgender minors. Ante, at 23. Yet the majority inexplicably refuses to take even the modest step of requiring Tennessee to show its work before the lower courts.

*  *  *

This case presents an easy question: whether SB1’s ban on certain medications, applicable only if used in a manner “inconsistent with . . . sex,” contains a sex classification. Because sex determines access to the covered medications, it clearly does. Yet the majority refuses to call a spade a spade. Instead, it obfuscates a sex classification that is plain on the face of this statute, all to avoid the mere possibility that a different court could strike down SB1, or categorical healthcare bans like it. The Court’s willingness to do so here does irrevocable damage to the Equal Protection Clause and invites legislatures to engage in discrimination by hiding blatant sex classifications in plain sight. It also authorizes, without second thought, untold harm to transgender children and the parents and families who love them. Because there is no constitutional justification for that result, I dissent.

SUPREME COURT OF THE UNITED STATES

_________________

No. 23–477

_________________

UNITED STATES, PETITIONER v. JONATHAN SKRMETTI, ATTORNEY GENERAL AND REPORTER FOR TENNESSEE, et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 18, 2025]

Justice Kagan, dissenting.

For all the reasons Justice Sotomayor gives, Tennessee’s SB1 warrants heightened judicial scrutiny. See ante, at 9–27 (dissenting opinion). That means the law survives if, but only if, its sex-based classifications are “substantially related to the achievement” of “important governmental objectives.” United States v. Virginia, 518 U.S. 515, 533 (1996). As Justice Sotomayor notes, the point of applying that test is to smoke out “invidious” or otherwise unfounded discrimination. Ante, at 10; Michael M. v. Superior Court, Sonoma Cty., 450 U.S. 464, 469 (1981) (plurality opinion). More concretely put, heightened scrutiny reveals whether a law is based on “overbroad generalizations,” stereotypes, or prejudices, or is instead based on legitimate state interests, such as the one here asserted in protecting minors’ health. Virginia, 518 U. S., at 533. Because the Court is wrong in not subjecting SB1 to that kind of examination, I join Parts I through IV of Justice Sotomayor’s dissent.

I take no view on how SB1 would fare under heightened scrutiny, and therefore do not join Part V. The record evidence here is extensive, complex, and disputed, and the Court of Appeals (because it applied only rational-basis review) never addressed the relevant issues. Still more, both the plaintiffs and the Government asked this Court not to itself apply heightened scrutiny, but only to remand that inquiry to the lower courts. So I would both start and stop at the question of what test SB1 must satisfy. As Justice Sotomayor shows, it is heightened scrutiny. I respectfully dissent.