by Lana Davidoff*
In making custody determinations, a court’s only goal is evaluating and effectuating the child’s best interest. This Contribution argues that courts making child custody decisions must treat a parent’s sexual conduct with greater consistency and analytical rigor. Although the “best interests of the child” standard requires courts to weigh factors like parental fitness, stability, and capacity to provide care, the analysis of sex addiction within this framework remains fragmented and often distorted by moral judgment. This Contribution contends that sex addiction should only be considered by courts making custody determinations through a structured inquiry that distinguishes between diagnosis and conduct, resists moralistic reasoning, and requires a causal nexus between the conduct and potential or actual harm. By reframing courts’ consideration of sex addiction in custody determinations with regard to parenting capacity, courts will be better equipped to protect children’s welfare without encroaching on parents’ liberty.
Child custody determinations sit at the fraught intersection of parental autonomy, judicial discretion, and child welfare. At their core, these decisions are meant to ensure that a child’s needs are met and that their growth and development are supported in a safe and stable environment.1 In principle, the “best interests of the child” standard that is typically applied by courts in custody decisions should be confined to the consideration of only factors that bear directly on a child’s health, safety, and overall well-being.2 In practice, however, moral judgments, social stigma, and misapplied notions of parental fitness too often intrude into this analysis. Judges may, consciously or unconsciously, penalize parents not for any demonstrated harm to their children, but for the choices they make in their personal lives—particularly regarding their sexual partners, identities, or practices. A parent’s private conduct should play a role in custody determinations only when supported by clear, demonstrable evidence showing a causal connection between that conduct and harm to the child’s health, safety, or welfare.
Nearly every jurisdiction employs the “best interests of the child” standard in custody cases, codified by state statutes.3 While formulations vary, the test generally focuses on a core set of considerations: the strength of the emotional bond between each parent and child; each parent’s demonstrated ability to provide for the child’s needs; the stability of each prospective home; and the mental and physical health of each parent.4 Notably, almost all of these statutes omit parental sexual conduct as an independent factor eligible to be considered by courts.5 The guiding objective of the best-interests test is to identify and effectuate the custody outcome which most fully promotes the child’s well-being.6 Accordingly, a parent’s private sexual conduct—when it poses no demonstrable risk of harm to the child—should not bear on custody determinations under the “best interests of the child” standard.
The Supreme Court has long recognized that the right to the care, custody, and control of one’s child is “perhaps the oldest of the fundamental liberty interests recognized by this Court.”7 This constitutional protection reflects more than a parent’s interest in decision-making authority; it embodies a core principle of family integrity and personal autonomy. Because custody determinations implicate this fundamental right, courts must exercise the utmost caution before curtailing it.
The Supreme Court has also made it clear that adults retain a protected sphere of privacy with respect to their intimate, consensual sexual conduct. In Lawrence v. Texas, the Court struck down criminal prohibitions on same-sex intimacy, grounding its decision in the recognition that liberty encompasses the right to make personal choices about one’s intimate life free from state intrusion.8 The logic of Lawrence has been extended beyond the criminal context; as the Court stated in Obergefell v. Hodges, “Lawrence confirmed a dimension of freedom that allows individuals to engage in intimate association without criminal liability, [but] it does not follow that freedom stops there.”9 In other words, the state cannot penalize adults simply for exercising their private, consensual sexual autonomy. As Judge Learned Hand observed: “We have answered in the negative the question whether an unmarried man must live completely celibate, or forfeit his claim to a ‘good moral character.’”10 The Court’s family law and constitutional law jurisprudence should thus be understood to establish the standard that private, consensual adult behavior should not factor into state restriction of parental rights absent demonstrable evidence of a causal nexus between the conduct and harm to the child.11
Courts must therefore act with precision and rigor when scrutinizing a parent’s intimate conduct, whether it involves non-traditional sexual practices, consensual relationships outside conventional norms, or the label of “sex addiction.” When such conduct has no causal negative effect on a child’s welfare, invoking it to deny or restrict parental custody amounts to weaponizing private behavior.12 Doing so not only distorts the best-interests standard by making use of a factor that is not statutorily prescribed, but also threatens the fundamental constitutional protections of privacy and equal protection. Courts are not empowered to impose their own moral preferences on parents; instead, their application of the “best interests of the child” standard should consider this private behavior only when it is connected to actual evidence of harm to the child. Under this causal nexus standard, harm to children occurs when sexual conduct creates an unsafe home environment (e.g., sexual activity in front of a child, sexual exploitation, or involvement of the child in sexualized settings), presenting a direct and recognizable risk to the child’s welfare.13
The propensity of judges to impute their own moralistic reasoning in custody disputes is well-known.14 Importing these subjective value judgments into the best-interests framework risks reinforcing outdated stereotypes about gender, sexuality, and family structure.15 It can also produce inconsistent and unpredictable results by leaving parents vulnerable to personal biases of individual judges rather than objective, standardized principles. Furthermore, it increases the risk of “erroneous factfinding” in a proceeding that necessarily raises important liberty interests.16 When judges rely on their personal moral judgments rather than concrete evidence, they may treat benign or constitutionally protected conduct as harmful, leading to findings that are unsupported by the record and distorted by subjective bias.
Moreover, reliance on contested psychiatric or medical labels to justify adverse findings of parental fitness compounds the problem by allowing speculative diagnoses to stand in for actual evidence of harm. For one, given that the reliability of “sex addiction” as a medical diagnosis remains contested, the courts should not be permitted to encroach upon a parent’s fundamental right to raise their children due to a label of ‘sex addict.’17 The American Psychiatric Association declined to include “hypersexual disorder” in the Diagnostic and Statistical Manual of Mental Disorders,18 and the World Health Organization declined to classify “Compulsive Sexual Behavior Disorder” in the 2019 International Classification of Diseases as an addiction.19 This divergence reflects a lack of scientific consensus about the validity and scope of a ‘sex addiction’ diagnosis. This disagreement clearly undermines the legitimacy of treating a sex addiction label as a proxy for parental fitness. Given that the validity of the diagnosis itself is uncertain, courts should not treat the label as per se disqualifying a parent’s right to custody, particularly considering that courts do not treat the arguably comparable and better established conditions of alcohol and substance abuse as disqualifying.20 So too, an alleged diagnosis of ‘sex addiction’ should be insufficient, standing alone, to justify restrictions on parental rights to custody.
The danger arises not from the mere existence of the ‘sex addict’ label, but from courts’ own application of that label—absent rigorous, admissible evidence that the conduct impairs caregiving—to the detriment of the parent, thereby ushering stigma rather than science into the courtroom. In doing so, courts risk collapsing complex medical debates into judicial shorthand for immorality. As such, courts should insist on a disciplined evidentiary approach requiring proof of a causal nexus between the parent’s private conduct and potential or actual harm to the child. Allegations of compulsive sexual behavior if raised should be supported by qualified expert testimony subject to transparent and rigorous standards of admissibility and proof. Even then, the critical inquiry for the admissibility of the evidence must remain tethered to the custody framework such that a parent’s sexual conduct only be considered if the conduct demonstrably endangers the child’s health, safety, or welfare. This will ensure that normative assumptions about sexuality—whether grounded in moral disapproval or in contested medical labels—are not substituted for concrete evidence of harm.
Several courts have improperly allowed parental sexual conduct or identity to overshadow the best-interests analysis, rather than requiring evidence of harm to the child. In Whaley v. Whaley, the Ohio Court of Appeals reversed a trial court’s decision to remove custody from a mother who had been involved in a relationship with a married man.21 The court observed that “[t]he judge’s decision demonstrates that the change in custody was ordered to punish Mrs. Whaley for conduct the court considered morally wrong,” rather than in response to evidence of harm to the child.22 Similarly, a Washington trial court awarded primary custody to a father based on the mother’s sexual orientation and her relationship with another woman.23 On review, the record revealed that “but for Rachelle’s sexual orientation and the challenge this presented to Charles’ religious beliefs, both Rachelle and Charles are capable parents.”24 These cases underscore that trial courts can reach outcomes so flawed and divorced from objective factfinding that appellate correction becomes inevitable, reinforcing why custody decisions must be grounded in evidence of actual or likely harm.
In yet another case, a Mississippi trial court openly acknowledged that while “a lesbian relationship is more acceptable today, it is not the norm,” and that although sexual orientation could not be the sole basis for its custody decision, it was nonetheless treated as a factor—a determination that was affirmed on appeal.25 The Utah Supreme Court’s decision to affirm the trial court’s determination in Shioji v. Shioji further illustrates this improper judicial use of normative assessments.26 There, the trial court removed custody from a mother based on her extramarital relationship, finding that her “conduct and attitude with respect to the relationship had a material and adverse effect on her parenting ability,” and that she and her partner were “either unwilling or unable to appreciate the adverse impact of their conduct on the children.”27 These decisions demonstrate how reliance on moralistic judgments, rather than evidence, can lead courts seriously astray, producing outcomes incompatible with a principled best-interests analysis. They reveal how easily personal biases—often shaped by views on sexual orientation or unconventional relationships—can seep into custody rulings. Even when appellate courts sometimes correct such errors,28 many decisions go uncorrected, as these cases illustrate, exposing the fragility of constitutional protections of liberty and due process. Parents should not have to gamble on appellate review; custody determinations at every level must be grounded in proof of actual or likely harm to the child, not in normative disapproval of private conduct.
Conversely, a strictly evidence-based approach aligns with constitutional commitments to both family sanctity and personal privacy, respects the statutory design of the best-interests standard, and offers a principled way to separate relevant parental behavior from irrelevant moral judgments. Ultimately, custody decisions should reflect an evidence-based inquiry into what truly impacts a child’s well-being, and they should not constitute a moral referendum on parents’ private choices. Only by insisting on a clear causal link between parental sexual conduct and child harm can courts ensure that custody law fulfills its purpose: protecting children without infringing on the rights and dignity of their parents.29
Several courts have applied this causal nexus approach in determining whether the private sexual behavior of a parent should bear on custody determinations, particularly in recent decades. In Kraft v. Kraft, the Delaware Superior Court refused to penalize a father based on allegations of “sex addiction,” affirming the trial court’s conclusion that the alleged addiction did not independently justify a custody limitation.30 Crucially, the court emphasized that the petitioner mother failed to identify any “specific facts that causally link that alleged condition to any demonstrated harm or endangerment” to the child.31 A similar approach guided the New York Family Court in Matter of RPF v. FG, a custody dispute in which both parents leveled allegations of sexual indiscretion against one another.32 Rather than allowing these accusations to drive the proceeding, the court deliberately reframed the inquiry around the correct statutory standard, focusing on what arrangement would best protect the children’s welfare.33 Importantly, even though the court credited allegations that the father had pursued sexual encounters with strangers online, the court found that such behavior did not, in itself, evince a risk to the children’s welfare.34 The father maintained stable, functioning relationships, and “there [was] no evidence that the children were ever exposed to any of these alleged sexual encounters.”35 By insisting on a demonstrated connection between parental conduct and actual harm, these courts safeguarded both the integrity of the best-interests standard and the constitutional limits on state intrusion into private adult behavior and the right to raise one’s child.
Family courts operate within a uniquely challenging terrain—balancing the imperative to protect children with the constitutional duty to respect parental autonomy. The proper path is narrow but clear: Custody determinations must be anchored in evidence of actual harm or a substantial risk of harm to the child. Absent such a demonstrated causal nexus, neither a contested diagnosis such as “sex addiction” nor a non-traditional but private and consensual sexual lifestyle can lawfully justify an adverse custody ruling. This rule not only safeguards fundamental liberty interests, but also aligns with medical expertise, reduces the risk of judicial bias, and preserves the evidentiary integrity of family law. Most importantly, it ensures that custody decisions remain where they belong: focused squarely on the needs, safety, and welfare of children, rather than on the moral judgments imposed upon adults.
* Lana Davidoff is a J.D. Candidate (2026) at New York University School of Law. This Contribution is a commentary on the problem at the 2025 Dominick L. Gabrielli National Family Law Moot Court Competition hosted by Albany Law School. One of the questions presented was whether a parent should be deemed unfit for physical custody when they have a sex addiction. The views expressed herein do not necessarily represent the author’s views.
1. 1 Child Custody and Visitation § 1.05 (2025).
2. See, e.g., N.Y. Dom. Rel. L. § 240 (McKinney 2023); Cal. Fam. Code § 3011 (West 2023).
3. See, e.g., id.
4. See, e.g., Matter of Supangkat v. Torres, 954 N.Y.S.2d 915, 916 (App. Div. 2012).
5. 1 Jeff Atkinson, Mod. Child Custody Prac. § 402 (2d ed. 2025).
6. See, e.g., Gloria S. v. Richard B., 437 N.Y.S.2d 411, 415 (App. Div. 1981) (“The courts are commanded by law and by sound considerations of policy to find not merely a solution, but the best solution in terms of the welfare of the child.”).
7. Troxel v. Granville, 530 U.S. 57, 65 (2000).
8. Lawrence v. Texas, 539 U.S. 558, 579 (2003).
9. Obergefell v. Hodges, 576 U.S. 644, 667 (2015).
10. Schmidt v. United States, 177 F.2d 450, 452 (2d Cir. 1949).
11. See, e.g., State ex rel. Tina K. v. Adam B., 307 Neb. 1, 14 (2020) (“When a fit parent has not forfeited his or her superior right to custody, the best interests of a child will negate the parental preference principle only in an exceptional case. . . . [which] requires proof of serious physical or psychological harm or a substantial likelihood of such harm.”); In re S.F., 308 Cal. Rptr. 3d 573, 585–90 (Ct. App. 2023) (reversing adverse custody decision absent evidence that the child had actually witnessed incidents of domestic violence or threats against mother, or evidence that subject father was still using drugs or that past or current drug use presented substantial risk of harm to child); In re J.N., 276 Cal. Rptr. 3d 885, 892 (Ct. App. 2021) (refusing to accept that “a parent’s violent criminal record, without more, necessarily establishes that a parent has a violent disposition sufficient to establish the requisite risk of physical harm to a particular child.”).
12. See, e.g., McCloud v. McCloud, No. 364877, 2023 Mich. App. LEXIS 5113, at *11 (Ct. App. July 20, 2023) (“A trial court cannot use infidelity to measure a parent’s moral fitness . . . unless the infidelity interferes with the parent’s ability to parent his or her child.”); Fletcher v. Fletcher, 447 Mich. 871, 887 (1994) (“Because of its limited probative value and the significant potential for prejudicially ascribing disproportionate weight to that fact, extramarital conduct, in and of itself, may not be relevant to [parental fitness].”).
13. See, e.g., In re Smith, 853 P.2d 282, 286 (Or. 1993) (affirming trial court’s decision to restrict father’s custody based on his sexual conduct towards another child in the home); In re G.M., 649 S.W.3d 801, 809 (Tex. App. 2022) (finding that “a mother’s act of allowing her boyfriend back into her home after her daughters reported that he sexually abused them may also be considered evidence of endangerment”); In re E.A.G., 373 S.W.3d 129, 148 (Tex. App. 2012) (finding that sexual abuse of one child in the home posed a risk to the other children in the household).
14. 1 Jeff Atkinson, Mod. Child Custody Prac. § 1–11 (2d ed. 2025) (“[C]ustody cases may bring out a judge’s personal background and emotions more than most other types of cases with which the judge deals. A judge’s personal predilections may be particularly important on issues of giving a preference to mothers over fathers, the weight given to a parent’s nonmarital sexual involvements, the weight given a child’s preference, a parent’s request to move out of state with the child, and other issues within the judge’s personal experience.”).
15. Id.
16. See, e.g., Santosky v. Kramer, 455 U.S. 745, 762 (1982) (writing that “imprecise substantive standards that leave determinations unusually open to the subjective values of the judge” present a unique and heightened risk of erroneous factfinding); see also Taylor v. Taylor, 849 S.W.2d 319, 326 (Tenn. 1993) (arguing that without more guidance, the best-interests standard “risks unwise results, stimulates litigation, permits manipulation and abuse, and allows a level of judicial discretion that is difficult to reconcile with an [sic] historic commitment to the rule of law”).
17. Troxel, 530 U.S. at 65 (finding that a parent’s “care, custody, and control of their children” is a fundamental liberty); see also In re A.T., 25 N.E.3d 1180, 1183 (Ill. App. Ct. 2015) (“A label can encompass a wide spectrum of effects and is not, standing alone, reliably indicative of a person’s level of functionality. . . . Rather, it is the actual conduct and behavior of the parent that is determinative on the question of fitness, not the label associated with such conduct or behavior.”).
18. Richard Krueger, Diagnosis of hypersexual or compulsive sexual behavior can be made using ICD-10 and DSM-5 despite rejection of this diagnosis by the American Psychiatric Association, 111 Addiction 2110, 2110 (2016).
19. See World Health Org., Int’l Classification of Diseases 11th Revision (2019), https://icd.who.int/.
20. See, e.g., Matter of S.D., Jr., 549 P.2d 1190, 1197 (Alaska 1976) (distinguishing alcohol use from intoxication that interferes with parenting duties); In re Phifer, 312 S.E.2d 684, 689 (N.C. Ct. App. 1984) (“A finding of fact that a parent abuses alcohol, without proof of adverse impact upon the child, is not a sufficient basis for an adjudication of termination of parental rights for neglect.”).
21. Whaley v. Whaley, 399 N.E.2d 1270, 1276 (Ohio Ct. App. 1978).
22. Id. at 1273.
23. In re Marriage of Black, 392 P.3d 1041, 1043 (Wash. 2017).
24. Id. at 1048.
25. S. B. v. L.W., 793 So. 2d 656, 658 (Miss. Ct. App. 2001).
26. Shioji v. Shioji, 712 P.2d 197 (Utah 1985).
27. Id. at 200.
28. See, e.g., Whaley, 399 N.E.2d at 1276; In re Marriage of Black, 392 P.3d at 1053.
29. See, e.g., Prince v. Massachusetts, 321 U.S. 158, 166 (1944) (recognizing that while “[i]t is cardinal with us that the custody, care and nurture of the child reside first in the parents, whose primary function and freedom include preparation for obligations the state can neither supply nor hinder[,] . . . the state as parens patriae may restrict the parent’s control” when “[a]cting to guard the general interest in youth’s well[-]being”).
30. Kraft v. Kraft, 29 A.3d 246, 255 (Del. 2011).
31. Id.
32. Matter of RPF v. FG, 47 N.Y.S.3d 666 (Fam. Ct. 2017).
33. Id. at 671 (“[T]his court is not concerned with who stepped outside of the relationship and rather, is solely concerned with what is in the best interest of the children.”).
34. Id. at 683.
35. Id.