¶ 1.

Mother appeals from the trial court’s order finding M.M. and C.M. to be children in need of care or supervision (CHINS). We affirm.

¶ 2. M.M. was born in September 2006; C.M. was born on June 25, 2014. The Department for Children and Families (DCF) began working with mother in June 2012. On June 25, 2014, DCF filed a petition alleging that M.M. and C.M. were CHINS. The parties agreed to a conditional care order with mother retaining custody subject to extensive conditions. Father was incarcerated at the time.

¶ 3. Following a January 2015 merits hearing, the court found the children to be CHINS. The trial court made the following findings. Mother became an opiate addict at age eighteen after which she “embarked on an unhappy diaspora of addiction.” She has been on and off Suboxone for nine years. Mother’s addiction issues may include alcohol. In October 2013, mother was arrested for driving while intoxicated. Her BAC Datamaster test result was .159%, roughly twice the legal limit for intoxication, which the court found sufficiently high to show an unhealthy tolerance for alcohol. Daughter M.M., then age seven, was riding in mother’s car at the time of mother’s DWI. She was not in a child safety seat or booster seat but was instead lying on the backseat with a seatbelt draped over her. The arresting officer was very familiar with mother and M.M. and he recognized them on contact. He had previously stopped and warned mother for allowing M.M. to ride in the car without proper restraints. The officer estimated that M.M. weighed fifty pounds at the time of mother’s DWI.

¶ 4. The court found that mother has been on and off treatment over the years. Some of the events that interrupted her prolonged treatment were outside of her control, such as the retirement of one of her doctors. It concluded that others, like losing her license and therefore her transportation to treatment in April 2014, were entirely her own fault. From the late fall of 2013 to May 2014, mother was not in treatment. Mother became pregnant with C.M. in the fall of 2013. During that fall and winter, mother was using street, or unprescribed, buprenorphine (Suboxone).

¶ 5. In the spring of 2014, after mother revealed her street buprenorphine use, a public health nurse told mother that she *543should continue to use the drug to avoid the harmful prospect of intrauterine damage to the unborn child and herself if she suddenly stopped. When mother returned to professional treatment on May 20, 2014, she was prescribed Subutex, a form of Suboxone. As a result of her self-medication — which itself then necessitated her being prescribed Subutex for her last month of pregnancy — C.M. was born opioid-dependent and required two months to be weaned off of opioids.

¶ 6. Despite her recent DWI incident, mother used alcohol on one occasion in February 2014 while five months pregnant with C.M. Mother also smokes cigarettes and did so throughout her pregnancy. The court found that not only was C.M. born significantly underweight at five pounds, but he also had a heart defect. There was no medical opinion provided linking mother’s continued smoking during pregnancy or her episode with alcohol to any medical issues with the child in útero, although mother admitted that she had been told by hospital personnel that her smoking and drug use may have contributed to the baby’s small birth weight.

¶ 7. As indicated above, DCF had been working with mother since June 2012. DCF was attempting to assist mother with issues involving domestic abuse by her male partner, failure to supervise M.M., and her admitted history of substance abuse. Mother was an inconsistent participant in DCF’s efforts to assist her, sometimes refusing to sign releases and disappearing for periods of time, but being cooperative and actively engaged in services at other times. For his part, father had apparently been in and out of jail and not engaged with the children.

¶ 8. At times, mother had responded to DCF concerns. When confronted about significant dental neglect of M.M., for example, mother scheduled a dentist’s appointment and remedied the problem. The court concluded that M.M.’s dental neglect could likely have worsened had DCF not threatened intervention. Fortunately for mother and the children, mother’s father and his girlfriend had consistently stepped up to support mother and protect the children.

¶ 9. Based on its findings, the court concluded that at the time the petition was filed both children were CHINS.1 The court *544determined that M.M. was grossly neglected when on two occasions mother drove while M.M. was unrestrained by an appropriate car seat. In the second incident, mother was driving while highly intoxicated. The court found that mother exposed her young child to a grave and real risk of serious bodily injury by driving drunk with M.M. unrestrained in the car. The court concluded that C.M. was CHINS at the time of the petition as well. He was born, in part due to mother’s substance abuse, addicted to opiates2 and with a low birth weight. He required two months to be weaned off of substances, substances that were present because of mother’s addiction. While all of C.M.’s health problems might not be directly attributable to mother’s addiction (mother claimed that his heart condition was genetic), the court noted that it was uncontroverted that the child was born addicted. The court also noted that while mother was seemingly not ready to take responsibility for that during her testimony, it was simply a fact that could not be disputed.

¶ 10. While the children were CHINS, the court noted some positive progress by mother since C.M.’s birth. Mother was substance free and in treatment. She lived in her own apartment and no longer at her father’s home. She was trying to quit smoking. DCF allowed the children to be placed with mother. In short, the court concluded, mother seemed to be on a positive trajectory toward overcoming her addiction and being a responsible parent.

¶ 11. Mother appeals from the court’s order, raising numerous arguments. She first takes issue with some of the court’s findings, arguing that: C.M. was born at five pounds, fifteen ounces, which is not “significantly underweight”; C.M. was not “addicted” to opiates as that term is technically defined; the court erroneously *545stated that mother became pregnant with C.M. “because” she was using drugs; mother was “in treatment” from the fall of 2013 to May 20, 2014, even though she refused to sign releases; and her high BAC did not demonstrate an “unhealthy tolerance for alcohol.” As to C.M., mother contends that he was not CHINS simply because he was opiate-dependent at birth and that the court construed the CHINS statute too broadly in reaching its decision.3 Mother also asserts that M.M. was not CHINS because the DWI and the dental neglect occurred well before the CHINS petition was filed.

¶ 12. A child is CHINS if he or she “is without proper parental care or subsistence, education, medical, or other care necessary for his or her well-being.” 33 V.S.A. § 5102(3)(B). As we recently reiterated, “the focus of a CHINS proceeding is the welfare of the child. The State must prove, and the court must determine [by a preponderance of the evidence], if the allegations in a CHINS petition have been established.” In re B.R., 2014 VT 37, ¶ 13, 196 Vt. 304, 97 A.3d 867 (quotation and citation omitted). A child does not need to suffer “actual harm” before he or she can be adjudicated CHINS. In re L.M., 2014 VT 17, ¶ 29, 195 Vt. 637, 93 A.3d 553. On review, we will uphold the trial court’s findings unless they are clearly erroneous, and the court’s legal conclusions will stand when supported by the findings. In re D.D., 2013 VT 79, ¶ 34, 194 Vt. 508, 82 A.3d 1143. It is not our role “to second-guess the family court or to reweigh the evidence.” In re S.B., 174 Vt. 427, 429, 800 A.2d 476, 479 (2002) (mem.).

¶ 13. None of the alleged factual errors identified by mother are significant and none warrant reversal. See In re R.W., 2011 VT 124, ¶ 17, 191 Vt. 108, 39 A.3d 682 (stating that Supreme Court applies harmless error analysis in juvenile cases, and will reverse judgment only where error has affected substantial rights of party); see also In re D.D., 2013 VT 79, ¶ 34 (reiterating that in juvenile proceedings court’s decision will not be reversed, even *546if some of trial court’s findings are unsupported, “if the remainder of the court’s findings, which are supported by the record, are sufficient to sustain the decision” (quotation and brackets omitted)).

¶ 14. The court stated that C.M. was five pounds at birth, which it characterized as “significantly underweight” and as a “low birth weight.” Mother points to her testimony that C.M. was five pounds, fifteen ounces, when born. Even if the court incorrectly believed that a five or six pound newborn is “underweight,” it was not a significant finding in light of other evidence of the newborn’s condition. We note that mother herself testified that she was told by hospital employees that the child was “a little small,” and that there had been concern during her pregnancy, given her drug use and her smoking throughout the pregnancy, that the child would have a low birth weight.

¶ 15. It cannot reasonably be disputed that C.M. was “addicted” to opioids at birth applying a common-sense definition of that term, or that he needed two months to be weaned off of drugs. It is of no moment that the court used the word “addicted” as opposed to opioid-dependent. It is beyond dispute that C.M. had to be weaned off drugs because mother used drugs while she was pregnant. The reason that mother was put on a prescribed dose of drugs during her last month of pregnancy was because mother had used Suboxone she acquired on the street up to that point in her pregnancy, and thus, the unborn baby would suffer from acute withdrawal if mother were to stop using drugs immediately. Again, mother herself testified to this effect. The court’s finding that mother became pregnant “because” she was using drugs appears to be a simple misstatement. Alternatively, as the State posits, the court may have been commenting that it was unfortunate, because mother was using drugs, that mother became pregnant. Regardless, this statement had no bearing on the court’s decision that C.M. was CHINS.

¶ 16. The court’s findings that mother was in and out of treatment and that she refused to sign releases are also supported by the evidence. The focus of the court’s concern was mother’s inconsistency in addressing her drug problem. It is uncontested that mother has been in treatment for nine years. Mother was DWI in October 2013, and she reported to DCF in December 2013 that she had relapsed and was using opiates. She was seeing a doctor but the doctor had asked her to leave his care. The social *547worker testified that between December 2013 and May 2014 mother was inconsistent in providing urine screens and there was a concern that she was not in treatment. The social worker further testified that mother was not in treatment between February 2014 and May 2014. Mother relapsed with alcohol in February 2014, and she acknowledged being kicked out of her doctor’s care. Mother was supposed to go to a daily clinic in Burlington, but she claimed it was difficult for her to do so. At a May 14, 2014 meeting with DCF, mother acknowledged that she was not in treatment and that she was using Suboxone off the street. On May 20, 2014, mother entered a program at Fletcher Allen.

¶ 17. Mother’s own testimony echoes this sequence of events. She described the retirement of her treating doctor and her search for a new doctor in the fall of 2013. She testified that before she was accepted into the new program in November 2013, she was using Suboxone off the street. She continued to use street drugs as she disclosed to DCF at a May 2014 meeting.

¶ 18. Even if it is not completely accurate to say that mother was not in treatment from the late fall of 2013 to May 2014, the error is harmless. Her participation was inconsistent, at best, and continued to involve the use of street drugs. She enrolled in a course of treatment through a hospital only one month before C.M. was born. Her refusal to sign releases concerning her care also raises concerns.

¶ 19. Based on this and other evidence, the court did not err in concluding that C.M. was CHINS. This is not a case where mother learned of her pregnancy and immediately sought help in a therapeutic setting, following a prescribed course of treatment, as in New Jersey Division of Child Protection, & Permanency v. Y.N., 104 A.3d 244 (N.J. 2014). The New Jersey court also applied a completely different statutory standard in conducting its analysis. Y.N. involved a prosecution of a parent for abusing and neglecting her child, and the state was required to prove that the mother “unreasonably inflicted harm on her newborn and did so, at least, by acting with gross negligence or recklessness.” Id. at 246, 253 (stating that New Jersey statute “makes clear that parental fault is an essential element for a finding of abuse or neglect”). By contrast, “the focus of a CHINS proceeding is the welfare of the child,” In re B.R., 2014 VT 37, ¶ 13 (quotation *548omitted), and the court must determine if a child is without the proper parental care necessary for his or her well-being, 33 V.S.A. § 5102(3)(B).

¶ 20. The New Jersey court held, under the New Jersey statute, that “absent exceptional circumstances, a finding of abuse or neglect cannot be sustained based solely on a newborn’s enduring methadone withdrawal following a mother’s timely participation in a bona fide treatment program prescribed by a licensed healthcare professional to whom she has made full disclosure.” Y.N., 104 A.3d at 246. Its decision turned on the absence of evidence to show parental fault. See id. at 255 (concluding that lower court erred “by concentrating on harm without regard to parental fault”). Even putting aside critical differences in statutory language, mother’s disclosure and enrollment in a bona fide treatment program in the instant case one month before giving birth cannot be construed as “timely.”

¶ 21. The record here shows that mother was kicked out of treatment, accessed unprescribed Suboxone off the street, and did not return to a medically monitored treatment regimen until shortly before the birth of the child. We agree that pregnant women should not run the risk that taking prescribed drugs for addiction treatment will necessarily result in a CHINS finding should their children be born addicted to medications prescribed to treat that addiction. That a child is born addicted to opioids, standing alone, does not necessarily require a finding of CHINS where that addiction is the result of the mother consistently and fully following a bona fide addiction treatment program. To hold otherwise would serve as a disincentive for pregnant women to seek out treatment for their addiction. But such is not the case here. Although access to services may provide barriers in some instances, mother had gotten treatment through approved sources and lost that opportunity due to her own conduct. Mother’s choice to use drugs accessed on the street, without proper medical monitoring, is fraught with danger in a host of ways, including abuse of Suboxone itself.

¶ 22. Finally, it was reasonable for the court to consider that a person who can operate a car with a BAC of .159 — almost twice the legal limit — may have an “unhealthy tolerance for alcohol.” Even if this were not a reasonable assumption, the error is harmless. It is undisputed that mother was driving while *549significantly impaired with her minor daughter unrestrained in the backseat. This was the focus of the court’s concern, not mother’s relative tolerance for alcohol. Mother’s consumption of alcohol on one occasion during her pregnancy does not rise to the level of concern necessary to support a CHINS determination. More concerning is the timing of her resumption of alcohol use, coming just a few months after a DUI arrest where her BAC was nearly twice the legal limit and during a time when she was not in treatment for her opioid addiction. The use of alcohol under those conditions, to say nothing of the fact of her pregnancy, is evidence that mother’s stability concerning substances was perilous. There was sufficient evidence for the court to find C.M. was CHINS as of the date the petition was filed.

¶ 23. We thus turn to mother’s remaining arguments. Mother argues that the court’s findings concerning M.M. were too far removed from the date that the CHINS petition was filed. She maintains, and we agree, that a child must be CHINS “at the time of the filing of the petition,” In re D.T., 170 Vt. 148, 156, 743 A.2d 1077, 1084 (1999), and she asserts that the “stale evidence” here cannot support a CHINS ruling.

¶ 24. We reject this argument. As mother acknowledges, we held in In re L.M. that a court’s CHINS analysis is not “limited only to the child’s well-being on the precise day that the CHINS petition was filed.” 2014 VT 17, ¶ 20. We explained that:

Obviously, the circumstances leading up to the filing of the CHINS petition are relevant in the court’s assessment. This allows the court to have a full picture of the child’s well-being and to base its decision on all relevant information; it promotes the care and protection of the child, while not unfairly undermining parents’ interest in maintaining family integrity.

Id.

¶ 25. In this case, the court was mindful of mother’s inconsistent record in addressing her opioid addiction, addressing M.M.’s basic needs such as dental care, and participating in DCF’s efforts to assist her in addressing these and other problems. The record before the court showed that mother had been on and off Suboxone for nine years. She was using this drug off the street, while pregnant, until a month before the CHINS *550petition was filed. She was in and out of treatment for her opioid addiction and she suffered several relapses before the CHINS petition was filed. She neglected M.M.’s dental needs to the point where DCF threatened to intervene. She drove with M.M. unrestrained in the car twice. She had been warned by a police officer that M.M. must be properly restrained, yet she allowed the child to ride in the back seat again unrestrained while she was driving the car in a highly intoxicated state. As the trial court found, this placed M.M. at a grave risk of harm. Mother pled guilty to DWI in April 2014, shortly before the CHINS petition was filed. This evidence was relevant to an assessment of M.M.’s well-being, and the trial court could properly consider it in reaching its conclusion. We reject mother’s argument that this evidence was too “stale” to support a finding, by a preponderance of the evidence, that on June 25, 2014, M.M. was without proper parental care necessary for her well-being.

¶ 26. Finally, we reject mother’s suggestion that the court here construed the CHINS statute too broadly. The court applied the plain language of the statute and appropriately assessed whether the children were without proper parental care necessary for their well-being as required by 33 V.S.A. § 5102(3)(B). That mother disagrees with the trial court’s conclusion does not render its decision unconstitutional.

Affirmed.

¶ 27.

concurring in part and dissenting in part. The primary bases for the trial court’s finding that C.M. is a child in need of care or supervision (CHINS), and the majority’s affirmation thereof, are that mother had a history of struggling with opioid addiction;4 mother was addicted to opioids at the time she became pregnant and continued to take nonprescription buprenorphine during her pregnancy; on the advice of a public health nurse, mother entered into a medically supervised opioid-*551assisted treatment program during the last month of her pregnancy; and C.M. was born with an opioid dependence5 that required that he undergo two months of medically supervised weaning.6

¶ 28. In affirming the trial court’s determination that C.M. is CHINS, the majority’s reasoning is essentially as follows: mother’s conduct before C.M. was born placed C.M. at risk; after birth C.M. required medical treatment as a result of the opioid dependence the fetus developed in útero; therefore, C.M. is a child in need of care or supervision. Alternatively, the majority’s opinion could be read to hold that mother had a history of opioid addiction; she had been in treatment for only a month at the time the CHINS petition was filed; and therefore, C.M. is CHINS.

¶ 29. The majority’s analysis either relies on the view that a mother’s prenatal neglect of a fetus that causes injury supports a *552CHINS finding, or on the view that when a parent is opioid-dependent and has not been in treatment for very long, the child is CHINS. Each of these analytic paths is fraught. Neither can support the majority’s affirmance of the court’s CHINS decision.

I.

¶ 30. I consider first the primary path: mother consumed opioids during her pregnancy, C.M. was born opioid-dependent and required medical treatment, and therefore C.M. is CHINS. I cannot agree that the fact that C.M. required medical treatment after birth on account of mother’s conduct while pregnant is a sufficient basis for finding that C.M. is CHINS. I suspect that most parents whose children are born opioid-dependent are demonstrably unable to provide adequate care for those children once they are born. See infra, ¶¶ 49-50. But we should not presume that every child born opioid-dependent is by definition CHINS on account of the fact that the child developed an opioid dependence in útero.

¶ 31. I note at the outset that the current standard of care for pregnant women with opioid dependence is opioid-assisted therapy with methadone, although emerging evidence suggests that buprenorphine also should be considered.7 “Abrupt discontinuation of opioids in an opioid-dependent pregnant woman can result in preterm labor, fetal distress, or fetal demise.” ACOG Comm. Op., at 1. Opioid-assisted therapy during pregnancy can “prevent complications of illicit opioid use and narcotic withdrawal, encourage prenatal care and drug treatment, reduce criminal activity, and avoid risks to the patient of associating with a drug culture.” Id. at 3. Accordingly, the American College of Obstetricians and Gynecologists recommends against withdrawal — even medically supervised withdrawal — from opioids in opioid-dependent women *553during pregnancy. Id. at 4; see also W. Kraft & J. van den Anker, Pharmacologic Management of the Opioid Neonatal Abstinence Syndroms, 59 Ped. Clinics of N. Am. 1147, 1150 (2012) (“[F]or newborns, the benefits of maternal opioid therapy during pregnancy using methadone in a structured program clearly outweigh no therapy.”).

¶ 32. Neonatal abstinence syndrome (NAS) “is an expected and treatable condition that follows prenatal exposure to opioid agonists.” ACOG Comm. Op. at 1. NAS is characterized by hyperactivity of the central and autonomic nervous systems. Id. at 5. The syndrome may cause the infant to become irritable, to have “uncoordinated sucking reflexes leading to poor feeding,” and to voice a high-pitched cry. Id. These symptoms generally develop within twelve to forty-eight hours of birth, peak at seventy-two to ninety-six hours, and generally resolve within seven days. Id. NAS symptoms severe enough to require pharmacologic treatment occur in 55-94% of infants born to opioid-dependent mothers. See Kraft & Anker, supra, at 1148.8

¶ 33. In short, late in her pregnancy, but before C.M. was born, mother — who had struggled with opioid addiction before and during her pregnancy — took the very steps she should have taken to treat her own addiction and provide the best outcome for C.M. upon birth.

¶ 34. The majority does not deny this, but essentially argues that the child is CHINS because the choices mother made before the child’s birth — in fact, prior to the child’s conception insofar as mother was addicted from the outset — led the child to be born with physical opioid dependency that required medical treatment after birth. The majority places great weight on the harm inflicted on the fetus before birth, and the risk to which mother exposed the fetus, as grounds for concluding that the child, once born, is CHINS. I am extremely wary of relying on the fact that mother’s prepartum conduct caused the fetus to become opioid-*554dependent as the basis for a CHINS finding. Although a parent’s conduct prior to a child’s birth may support inferences about the parent’s ability to care for the child upon birth, I do not believe that we can predicate a CHINS finding on harm inflicted on a fetus before birth.

¶ 35. I reach this conclusion for several reasons. First, the statutes governing child abuse and neglect cases are forward-looking. They are to be construed to “provide for the care, protection, education, and healthy mental, physical, and social development of children.” 33 V.S.A. § 5101(a)(1). Although a parent’s prepartum conduct — like other past behavior, such as conduct toward a child’s sibling — may be relevant insofar as it supports an inference that the child, once born, is at risk of neglect or abuse, the civil child-protection statutes are not designed to punish prepartum conduct. As one commentator explained:

Removal of a child from parental custody or otherwise intervening in the child’s and parent’s family life must be accomplished to keep the child safe, not to punish a parent for her evil deeds. Punishment of a parent’s actions towards a child ought to be left strictly to the criminal law.

L. Nelson & M. Marshall, Ethical and Legal Analyses of Three Coercive Policies Aimed at Substance Abuse by Pregnant Women 106-07 (1998). As another has elaborated:

Past parental conduct will support a finding of neglect or abuse only if a court determines that a child is likely to face future harm. It is the prediction that a child is likely to come to harm in the future that underlies the decision to intervene, not the past parental conduct itself. Thus, state intervention is appropriate only when the likelihood of future harm to the child is great — that is, when past parental conduct correlates strongly with future conduct that is likely to cause specific harm to the child. . . . Whether prior parental conduct is blameworthy or repulsive should not be of concern to the child welfare system.

C. Clarke, FINS, PINS, CHIPS, & CHINS: A Reasoned Approach to the Problem of Drug Use During Pregnancy, 29 Seton Hall L. Rev. 634, 668 (1998).

*555¶ 36. Accordingly, “while it should be acknowledged that a woman’s prenatal use of drugs, alcohol, or other substances might affect the child’s health and development or make her care more challenging and difficult, ‘only drug or alcohol use after the child’s birth will affect the mother’s ability to provide appropriate care.’ ” Nelson & Marshall, supra, at 104 (citing New York Legal Aid Soc’y, Juvenile Rights Div., Position Paper, Governmental Action in Cases of In Utero Drug or Alcohol Exposure: The Role and, Responsibilities of Child Protective Authorities and the Family Court [hereinafter Juvenile Rights Div. Position Paper] (Dec. 1997) at 8. As the Juvenile Rights Division of the Legal Aid Society of New York9 explained in a 1997 position paper:

Drug and alcohol abuse during pregnancy is insupportable behavior because of its potential for harming the developing fetus. Yet, in a child-protective context, the salient characteristic of drug or alcohol use is its effect on the parent’s ability to provide adequate care for the child after birth.

Juvenile Rights Div. Position Paper at 1. See also B. Robin-Vergeer, Note, The Problem of the Drug-Exposed Newborn: A Return to Principled Intervention, 42 Stan. L. Rev. 745, 774-75 (1990) (“Given . . . that the child welfare system is currently concerned with dealing with the drug-exposed infant after [birth], the focus on the harm inflicted upon the child in útero, as if the child had been battered, is clearly misplaced, except as it bears on the mother’s ability to care for her child in the future.”) (emphasis added).

¶ 37. The question is not what harm mother’s prepartum consumption of opioids caused to the developing fetus; instead, the question is what her conduct tells us about her ability to adequately care for the child after birth. Our own case law buttresses this understanding. As we have explained:

“The issue before the family court at the merits stage of a CHINS proceeding is a determination of whether, at the time of the filing of the petition, the juvenile is a *556child in need of care and supervision.” . . . This does not mean that the court’s analysis is limited only to the child’s well-being on the precise day that the CHINS petition was filed. Obviously, the circumstances leading up to the filing of the CHINS petition are relevant in the court’s assessment. This allows the court to have a full picture of the child’s well-being and to base its decision on all relevant information; it promotes the care and protection of the child, while not unfairly undermining parents’ interest in maintaining family integrity.

In re L.M., 2014 VT 17, ¶ 20, 195 Vt. 637, 93 A.3d 553 (emphasis added and footnote omitted) (quoting In re D.T., 170 Vt. 148, 156, 743 A.2d 1077, 1084 (1999)).

¶ 38. To the extent that the majority focuses on the postpartum consequences of the harm experienced by the fetus in útero, it departs from the purpose of our child-protection laws, and substitutes a judgment about mother’s opioid consumption and her delay in seeking treatment for her addiction for an evidence-based analysis of the prospective risk of harm faced by C.M.

¶ 39. A second reason that I do not believe a CHINS determination can be based on prepartum harm inflicted on a fetus by mother is that our CHINS statute, on its face, is limited to protecting children, not fetuses in útero. A “child in need of care or supervision (CHINS)” is, by definition, a child. 33 V.S.A. § 5102(3). A “child,” in turn, is defined as an individual of a specified age for specified purposes. Id. § 5102(2). The statute does not contemplate inclusion of fetuses as in need of care or supervision. Any act or risk of abuse or neglect must exist postpartum to be recognized as such by our statutes. For similar reasons, other courts have refused to apply state child abuse and neglect laws to harms inflicted on a fetus prepartum. For example, in declining to affirm the termination of a mother’s parental rights based on her injection of cocaine shortly before she went to the hospital to give birth to the child in question, the Connecticut Supreme Court explained:

We also note that we do not endorse the moral quality of the conduct of the respondent in this case. Certainly no one approves of the intravenous injection of cocaine by a pregnant woman, who had been warned of the risks to her fetus, at any time during her pregnancy let alone *557just before the onset of labor. Nor, on the other hand, are we here to condemn her for succumbing to what may well have been the unyielding demands of her addiction. Our task, rather, is to determine whether the legislature, in enacting [the termination of parental rights statute], intended it to apply in a case such as this. We do not believe that it did.

In re Valerie D., 613 A.2d 748, 759 (Conn. 1992).

¶ 40. The court went on to examine the applicable child-protection statute:

“‘Child’ means any person under sixteen years of age.” The ordinary usage of the term “parent,” insofar as it applies to the female, suggests that, unless the context requires otherwise, it means “one [who] . . . brings forth offspring.” Thus, in ordinary parlance, until the child in this case was born, or was “brought forth,” the respondent was not her “parent” and the conduct of the respondent with respect to her was not “parental” conduct. Similarly, the definition of “child” as a person “under sixteen years of age” suggests a limitation on the applicability of that definition to a person who has been born, since that is the ordinary beginning point of one’s “age.” Thus, until the moment of birth, Valerie was not a “child” within the meaning of [the applicable statute] and, therefore, the “act ... of parental commission” that took place before that moment cannot be considered to be parental conduct that “denied [her] ... the care . . . necessary for [her] physical . . . well-being.

Id. at 760 (citations omitted).10

*558¶ 41. Some courts have concluded otherwise — authorizing state intervention to protect a newborn child pursuant to general child protection statutes based solely on prior prepartum injury to a developing fetus. For example, in In re Baby Boy Blackshear, a divided Ohio Supreme Court concluded that “when a newborn child’s toxicology screen yields a positive result for an illegal drug due to prenatal maternal drug abuse, the newborn is, for purposes of [the applicable child protection statute] per se an abused child.” 736 N.E.2d 462, 465 (Ohio 2000). The court concluded that once born, the child fit within the abuse and neglect statute, and that the mother’s actions caused the child to experience injury both before and after birth. Id. at 464-65. It also stated that “there can be no more sacred or precious right of a newborn infant than the right to life and to begin that life, where medically possible, healthy, and uninjured.” Id. at 465; see also In re Baby X, 293 N.W.2d 736, 738-39 (Mich. Ct. App. 1980) (noting that, “[w]hile there is no wholesale recognition of fetuses as persons, fetuses have been accorded rights under certain limited circumstances. . . . when it is for the child’s best interest,” and concluding that “a newborn suffering narcotics withdrawal symptoms as a consequence of prenatal maternal drug addiction may properly be considered a neglected child” (citations omitted)); In re U.P., 105 S.W.3d 222, 234-36 (Tex. Ct. App. 2003) (affirming termination of parental rights of father who, among other things, provided illegal drugs to mother during her pregnancy, thereby endangering the child, and explaining, “Under the Texas Family Code, a ‘child’ is defined as a person under 18 years of age . . . . However, harm done to a child in útero has generally been recognized in Texas when the child has been born alive.”).

¶ 42. Other states have enacted statutes expressly extending the protections of the child abuse and neglect statutes to infants who were exposed to alcohol or controlled substances in útero. See, e.g., Fla. Stat. Ann. § 39.01(30)(g) (defining “harm [to child’s health],” for purposes of neglect or abuse finding, to include newborn infant’s physical dependency on a controlled drug); Ga. Code Ann. § 15-11-2(2), (56) (including “prenatal abuse” in definition of abuse and defining “prenatal abuse” to include prenatal *559exposure to chronic or severe use of alcohol or controlled substances resulting in specified harms or consequences); 750 Ill. Comp. Stat. 50/l(D)(t) (defining “unfit person” to include biological mother of a child born with “any amount of a controlled substance” in his or her blood, urine, or meconium, if biological mother had at least one other child previously adjudicated as a neglected minor and an opportunity to enroll in treatment); La. Child. Code Ann. art. 603(18), (24) (providing that “[njeglect includes prenatal neglect” and defining “prenatal neglect” as “the unlawful use by a mother during pregnancy of a controlled dangerous substance . . . which results in symptoms of withdrawal in the infant or the presence of a controlled substance in the infant’s body”); Minn. Stat. Ann. § 626.556 subdiv. 2(g)(6) (defining “neglect” to include “prenatal exposure to a controlled substance . . . used by the mother for a nonmedical purpose, as evidenced by withdrawal symptoms in the child at birth, [or] results of a toxicology test performed on the mother at delivery or the child at birth”).

¶ 43. Given the Vermont Legislature’s recent focus on the epidemic of opioid addiction in Vermont and its impact on Vermont families, see, e.g., An act relating to strengthening Vermont’s response to opioid addiction and methamphetamine abuse, 2013, No. 75 (eff. July 1, 2013), as well as its recent attention to Vermont’s response to child abuse and neglect, see, e.g., An act relating to improving Vermont’s system for protecting children from abuse and neglect, 2015, No. 60 (eff. July 1, 2015), the Legislature is well-positioned to determine, within constitutional limitations, the best approach to addressing the serious problems afflicting children born with conditions arising from in útero exposure to alcohol, nicotine, and controlled substances. I would not preempt that conversation by holding that an infant who is suffering as a result of harms in útero is, on that basis alone, CHINS. See Med. Ctr. Hosp. of Vt. v. Lorrain, 165 Vt. 12, 16, 675 A.2d 1326, 1329 (1996) (“The Legislature, not this Court, is better equipped to assemble the facts and determine the appropriate remedies in an arena fraught with social policy . . . .”); see also Paris Adult Theatre I v. Slaton, 413 U.S. 49, 64 (1973) (“We do not sit as a super-legislature to determine the wisdom, need, and propriety of laws that touch . . . social conditions.” (quotation omitted)).

¶ 44. Another reason that I resist the inference that a child born with NAS is presumptively CHINS is that such a policy *560would deter pregnant opioid-addicted women from taking the steps best calculated to protect their own health and that of their children. An approach that incentivized pregnant women to attempt an abrupt opioid withdrawal in order to avoid the birth of a child with a physical opioid dependence would fly in the face of ACOG’s recommended course for pregnant opioid-addicted women. Moreover, studies have shown that fear of punitive responses and loss of custody is a deterrent to pregnant women seeking treatment for drug addiction. See, e.g., M. Jessup et al., Extrinsic Barriers to Substance Abuse Treatment Among Pregnant Drug Dependent Women, 33 J. Drug Issues 285, 291-92 (2003) (identifying fear of detection and prosecution or loss of custody as a major barrier to pregnant and parenting women seeking treatment for drug addiction). An approach that focuses on a forward-looking analysis of the risks a born child faces would incentivize opioid-addicted pregnant mothers to seek treatment, rather than avoid it.11

¶ 45. Finally, I am concerned about the breadth of an approach that relies on harms sustained by a fetus in útero to support a CHINS determination. Although the majority asserts that the fact that a child is born dependent on opioids, standing alone, does not necessarily require a finding of CHINS where the mother consistently and fully followed a bona fide treatment program, ante, ¶ 21, the majority’s reasoning suggests otherwise. To the extent the child’s in útero exposure to opioids and ensuing physical dependence is a basis for a CHINS finding, that exposure and dependence exists whether or not the mother is in medically supervised treatment, and regardless of the duration of that treatment.12 Our judgments about the mother’s conduct may differ depending on how long it took her to seek medical care, but there *561is no finding that mother’s reliance on nonprescription buprenor-phine for much of her pregnancy, as opposed to prescribed bupre-norphine for the last month, impacted the extent of C.M.’s physical dependence at birth. I believe the majority’s reasoning — that the harm to the fetus in útero can support a CHINS determination at birth — applies in virtually every case in which a child is born with an opioid dependency.

¶ 46. Moreover, the court’s approach has ramifications far beyond the use of opioids. Medical research has established that prenatal alcohol abuse has strong effects on fetal growth and anomalies, as well as on long-term growth, behavior, cognition, and achievement. Behnke & Smith, supra, at 1016. That nicotine has long-term effects on behavior, cognition, language and achievement is likewise well-established. Id. If harm sustained by a fetus in útero can support a CHINS finding upon birth, then children born to mothers who smoke cigarettes or drink during pregnancy are similarly subject to a CHINS finding. In the United States, about 15.4% of pregnant women reported cigarette smoking, 9.4% reported current consumption of alcohol, and 2.3% reported binge drinking. U.S. Dep’t of Health & Human Srvs. et al., NSDUH Series H-48, HHS Pub. No. (SMA) 14863, Results from, the 2013 National Survey on Drug Use and Health: Summary of National Findings, at 37, 51 (2014), http://www.samhsa.gov/data/sites/default/ files/NSDUHresultsPDFWHTML2013/Web/NSDUHresults2013.pdf [perma.cc/7BTH-FV2J].

¶ 47. Reliance on prenatal injury as a basis for a CHINS finding after birth could have consequences outside of the realm of substance abuse as well. Conduct that is ill-advised for a pregnant woman, but that tells us little about her ability to safely parent postpartum, could likewise support a CHINS finding. As the Kentucky Supreme Court noted in concluding that the criminal child abuse statute did not extend to the self-abuse of the pregnant-mother-to-be:

The mother was a drug addict. But, for that matter, she could have been a pregnant alcoholic, causing fetal alcohol syndrome; or she could have been addicted to self abuse by smoking, or by abusing prescription painkillers, or over-the-counter medicine; or for that matter she *562could have been addicted to downhill skiing or some other sport creating serious risk of prenatal injury, risk which the mother wantonly disregarded as a matter of self-indulgence. What if a pregnant woman drives over the speed limit, or as a matter of vanity doesn’t wear the prescription lenses she knows she needs to see the dangers of the road?

Commonwealth v. Welch, 864 S.W.2d 280, 283 (Ky. 1998).

¶ 48. As I did at the outset, I emphasize that I am not arguing that a pregnant mother’s consumption of opioids during pregnancy is not relevant in a CHINS case. Combined with evidence of how the mother’s opioid consumption affects her ability to parent, the evidence may be highly relevant. But I am arguing against a particular analytical path to a CHINS determination — one that relies on the harm to a fetus in útero, and does not focus on the harm or risk of harm experienced by the child after birth.

II.

¶ 49. A second way to interpret the majority’s opinion is that mother’s addiction, untreated until near the end of her pregnancy, renders her incapable of providing adequate care to C.M. after birth. An opioid-addicted parent may well be unable to adequately care for a child, especially a newborn, but I do not believe that identifying a caregiving parent as opioid-addicted alone establishes that a child is CHINS. This is especially true when a parent’s opioid dependence is being successfully managed through medically supervised treatment.13

*563¶ 50. But even if a parent is not in treatment, or has not been in treatment for a long enough time that his or her condition is stable, I do not believe that identifying that parent as opioid-dependent, without more, is sufficient to support a CHINS order. See In re B.R., 2014 VT 37, ¶ 85, 196 Vt. 304, 97 A.3d 867 (Robinson, J., dissenting) (presumption that one drug-addicted parent in the household is sufficient to render the child CHINS “is way overbroad and cannot substitute for actual evidence that ... as a result of that parent’s addiction, the child’s needs are not being met”); In re L.M., 2014 VT 17, ¶ 39 (Robinson, J., dissenting) (“[I]f the State is to take the awesome step of interposing itself into the parent-child relationship, it cannot rely on broad generalizations or per se rules; it must have some individualized evidence that a child is without proper parental care necessary for the child’s well-being.”).

¶ 51. This is a closer question. I suspect that in the vast majority of cases children in the sole care of an opioid-addicted parent or parents likely are CHINS. See L.M., 2014 VT 17, ¶ 38 (Robinson, J., dissenting) (“[A]n active opiate addiction can cloud even the most caring and thoughtful parent’s judgment and capacity to properly care for his or her child.”). In these cases, I expect the State can muster evidence that the opioid addiction has interfered with the parent or parents’ ability to make safe and healthy choices for the child in question, other children, or even for the parent’s own self-care. The State can present evidence of the parent’s behavior arising from or related to the opioid dependence to provide a sense of the impact on that parent of the addiction, whether manifested in an inability to hold a job, to provide safe housing for himself or herself and the child, to feed and clothe the child and ensure minimum standards of hygiene, to get the child to school, to ensure that the child receives age-appropriate attention and care, or some other evidence. But I do believe this added component of evidence, and associated findings — relating not just to the fact of a parent’s opioid dependence or drug addiction, but to the impact of that condition on the safety and well-being of the child — is necessary. I hold this view for several reasons.

¶ 52. First, I believe a presumption that the child of an opioid-addicted parent is CHINS shifts the focus from the well-*564being of the child to the status of the parent. This Court has emphasized that “the focus of a CHINS proceeding is on the child’s welfare.” In re B.R., 2014 VT 37, ¶ 20 (quotation omitted); see also L.M., 2014 VT 17, ¶ 19 (noting that the focus at CHINS merit hearing is on whether child is “without proper parental care or subsistence, education, medical, or other care necessary for his or her well-being,” and that parental unfitness is an issue to be resolved at the disposition stage) (quotation omitted); In re S.G., 153 Vt. 466, 472, 571 A.2d 677, 682 (1990) (“[T]he issue in juvenile proceedings is not whether the parent did a particular act or acted in conformity with a particular character trait but instead whether the child has proper care and his or her well-being is protected.”).

¶ 53. The connection between the parent’s opioid addiction and the child’s health and safety depends on the inference that a child of an opioid addicted parent “is without proper parental care or subsistence, education, medical, or other care necessary for his or her well-being.” 33 V.S.A. § 5102(3)(B). It does not rest on any direct consideration of the subsistence, education, medical or other care provided or available to the child, harm suffered by the child, or risk of harm — except to the extent the risk is inferred from the fact of the parent’s addiction.

¶ 54. That takes me to my second, and closely related, concern about the conclusion that the child of an opioid-addicted parent is necessarily CHINS. It rests on the kind of presumption disfavored in the law. The United States Supreme Court considered a due-process challenge to a more obviously antiquated state-law presumption that unmarried fathers are not fit to raise their children. Stanley v. Illinois, 405 U.S. 645 (1972). In striking down the statute that made children of an unwed mother wards of the state upon the death of the mother, the Court explained:

It may be, as the State insists, that most unmarried fathers are unsuitable and neglectful parents. It also may be that Stanley is such a parent and that his children should be placed in other hands. But all unmarried fathers are not in this category; some are wholly suited to have custody of their children.

Id. at 654 (footnotes omitted). The Court acknowledged that “[pjrocedure by presumption is always cheaper and easier than individualized determination,” but concluded that the presumption *565“foreclose^] the determinative issues of competence and care.” Id. at 656-57. See L.M., 2014 VT 17, ¶ 41 (Robinson, J., dissenting) (“[FJather’s untreated longstanding opiate addiction is undoubtedly a red flag. But without more, I cannot tell whether and how it put the child at risk. . . . [I]t cannot be that the child of every parent with an admitted opiate addiction is presumed CHINS without any individualized showing.”); B.C. v. Dep’t of Children & Families, 846 So. 2d 1273, 1275 (Fla. Dist. Ct. App. 2003) (reversing adjudication of dependency on basis of father’s drug and alcohol abuse where there was “no testimony that the father failed to meet the child’s needs while the child was in his care, no testimony that physical harm had come to the child while in the father’s care, and no testimony that the child had been emotionally or mentally harmed by his father’s drinking and drug abuse”).

¶ 55. Just as I suspect that most children of opioid-addicted parents lack adequate parental care necessary for their well-being, I suspect that some opioid-addicted parents are able to provide adequate care. See, e.g., S. Boyd, Mothers and Illicit Drugs: Transcending the Myths 14-17 (1999) (reviewing fifteen studies demonstrating that women who use illegal drugs can be fit parents); Juvenile Rights Div. Position Paper at 14 (“Many children are apparently unaffected by their pre-natal exposure to drugs or alcohol, and some mothers are able to provide an adequate environment for their drug/alcohol-exposed children, especially if a solid support system is available to them.”); B. Smith & M. Testa, The Risk of Subsequent Maltreatment Allegations in Families with Substance-exposed Infants, 26 Child Abuse & Neglect 97, 97 (2002) (“Parents in the [substance-exposed-infant] group are not more likely to incur other types of allegations such as physical abuse or lack of supervision.”).

¶ 56. In the end, my personal suspicion is an inadequate foundation to support either proposition. I have no basis for assessing the scientific soundness of the presumption that the child of an opioid-addicted parent is CHINS. Are all individuals diagnosed as opioid-addicted so impaired by the addiction that they cannot adequately parent? Does it matter whether the opioid in question generates a “high,” like heroin, or merely dulls the cravings, like buprenorphine? Does it matter how frequently a person uses, or the amount of the doses? Does it matter how long the individual has been afflicted by addiction? In developing the record in this case, the parties, reasonably, did not purport to *566address these fundamental and weighty issues. Nor have we squarely considered such issues in a case on appeal, informed by thorough briefing on the scientific literature. I am unwilling to yield to general assertions of “common sense,” even my own, with respect to a disease — drug addiction — that has historically been stigmatized and has in some cases given rise to reflexive and less-than-fully-informed reactions.

¶ 57. That leads to my third point. At least in recent years, the policy of the State of Vermont, as articulated by the Legislature, has been to treat opioid addiction as, first and foremost, a disease, rather than a moral failing or an indication of a criminal disposition. See, e.g., 2013, No. 75, § 1(a) (eff. July 1, 2013) (“This act is intended to provide a comprehensive approach to combating opioid addiction and methamphetamine abuse in Vermont through strategies that address prevention, treatment, and recovery, and increase community safety by reducing drug-related crime.”); id. § 1(b) (directing that legislative initiatives concerning methamphetamine abuse and opioid addiction “be integrated to the extent possible with the Blueprint for Health and Vermont’s health care system and health care reform initiatives”).

¶ 58. I cannot think of any other disease afflicting a parent that we have identified, or would identify, as per se evidence that a child is CHINS. I can think of plenty of medical and psychological conditions that may cause a parent to be unable to care for a child properly, but would expect individualized evidence of the effect of any of those conditions on a particular parent before concluding that the parent’s child is CHINS. In re G.C., 170 Vt. 329, 335, 749 A.2d 28, 33 (2000) (“[T]he fact [of] mother’s mental illness . . . does not necessarily, in and of itself, satisfy the State’s burden in the disposition phase of the dependency proceedings.”); In re B.S., 166 Vt. 345, 352, 693 A.2d 716, 720 (1997) (“Mental retardation is not, by itself, a ground for terminating parental rights.”). If we are to be true to the understanding of opioid addiction as a disease, then we ought to treat it as such. Moreover, an across-the-board presumption that a child in the care of an opioid-addicted parent is CHINS would create a tremendous disincentive for all such parents — not just pregnant mothers — to seek the medical treatment that can help them heal and become better parents.

¶ 59. The absence of state intervention when a child is at risk can lead to unacceptable consequences for the children that our *567child-protection laws are designed to protect. But state interference where it is not warranted itself poses risks to those same children. In my view, the best way to walk the difficult tightrope between these two realities is to base CHINS decisions on individualized evidence concerning an opioid-addicted parent’s conduct and capabilities, rather than rely solely on broad generalizations about the conduct and capabilities of opioid-addicted individuals — even generalizations that may be, generally, grounded in experience. If a blanket presumption that opioid-addicted individuals cannot adequately parent is well-founded, the State should not have difficulty mustering specific evidence demonstrating that a child in the care of a particular opioid-addicted parent is in need of care or supervision. Evidence of and findings concerning a parent’s abusive conduct toward a child; an unmitigated course of neglectful conduct with respect to other children; neglect of a parent’s own self-care; inability to meet basic needs like safe housing or nourishment; exposure of himself or herself, or children, to violence or danger; or erratic behavior are all examples of conduct that may be fueled by addiction that directly bears on the well-being of a child in that parent’s care.

¶ 60. For these reasons, I cannot affirm the trial court’s CHINS determination based on its finding that mother has a history of opioid addiction, and has only been in treatment for a month, without some findings that move the conversation beyond mother’s diagnosis to the impact of her opioid use on her ability to parent C.M.

III.

¶ 61. In this case, once you strip away the findings that C.M. was born opioid-dependent, and the corollary that mother was addicted to opioids before and during her pregnancy with C.M., and had only been in treatment a short time before the CHINS petition, we are left with findings that mother required prompting to address her older daughter’s dental problems more than a year before the petition; drove while highly intoxicated with her then-seven-year-old daughter in the car more than eight months before the petition; had, prior to that, been warned for allowing the older daughter to ride without proper restraints; and had two cans of Twisted Tea alcoholic beverage while five months pregnant with C.M. I don’t purport to address the question whether these findings would be sufficient to support a CHINS determination *568with respect to C.M., and the subsidiary question of whether the probative value of any of these incidents had receded by virtue of the passage of time before the State’s CHINS petition. It is clear that the trial court’s CHINS determination with respect to C.M. rested almost entirely on its findings that C.M. developed an opioid dependence in útero as a result of mother’s prenatal conduct, and that mother was addicted to opioids and had been in medically supervised treatment for only a month prior to C.M.’s birth. For the above reasons, I believe these findings alone are insufficient to support a CHINS determination with respect to C.M., and I respectfully dissent from the majority’s judgment with respect to C.M.14

¶ 62. Diligence in ensuring that CHINS determinations are based on careful, individualized determinations is especially vital in these times. The current opioid epidemic in Vermont is real, and its consequences for many Vermont children are tragic. See 2014 Vermont State of the State Address (Jan. 8, 2014), C-SPAN/ WCAX-TV, http://www.c-span.org/video/7317076-l/vermont-state-state-address (“In every corner of our state, . . . opiate drug addiction threatens us. . . . When parents struggle, children suffer, and we all pay the price for years to come.”). DCF child abuse and neglect caseloads have skyrocketed in recent years. Vt. Dep’t for Children & Families, Family Serv. Div., 2014 Report on, Child Protection in Vermont (2015), at 5, http://dcf.vermont.gov/sites/dd/ files/pdf/fsd/2014-CP-Report.pdf (reporting that in 2014 DCF “received a record number of child abuse and neglect reports” — a nearly 33% increase from the previous year, with more than two-thirds of the increase involving children under six — “and substance abuse was a factor in about one-third of them”); Casey Family Programs, Assessment of Family Services Division, Safety Decision, Making: Final Report to the Vermont Department for Children, and Families, at 4-5 (Dec. 2014) [hereinafter Casey Report], http://mediad.publicbroadcasting.net/p/vpr/files/201412/ Casey-Report-Full-VPR.pdf (Vermont’s “struggle with a large increase in opioid abuse cases” has led to an “increase in reports of serious child maltreatment,” which “is challenging the service capacity of [DCF] and other agencies in Vermont”).

¶ 63. Increasing caseloads have seriously strained the docket of the family division of the superior court. See Vermont Judiciary *569Annual Statistical Report for FY 2014, at 5, https://www.vermont judiciary.org/jbn/Shared%20Documents/Annual%20Statistical%20 Report%20for%20FY%202014.pdf (reporting that “[t]he increase in CHINS filings over the past few years has been fueled primarily by a dramatic growth in abuse/neglect cases,” with a 62% increase between fiscal years 2010 and 2015, “the largest increase in any case type in any division of the superior court”). The labor-intensive nature of such cases, combined with the “dramatic rise” in their volume, has “put a significant strain on the resources of the trial courts.” Id. at 4.

¶ 64. Exploding caseloads and public scrutiny in the wake of high-profile tragedies have increased the pressure on already overworked and overwhelmed child protection personnel. See, e.g., Casey Report at 10-11 (“Many [DCF] social workers have caseloads and workloads which make it difficult or impossible to complete job tasks on time while doing good quality casework,” and “Many [DCF] social workers feel isolated and unsupported”); A. Burbank & R Achen, Child Protection Workers Face Danger, Criticism, Burlington Free Press, Aug. 17, 2015, http://www. burlingtonfreepress.com/story/news/local/vermont/2015/08/12/ vermont-social-workers-respond/31579459/ [http://perma.cc/E6DL-JS9H] (“[0]ne thing is constant: Vermont child welfare caseworkers always face scrutiny and criticism.”).

¶ 65. These forces have given rise to a perfect storm, exacerbating the already Herculean challenge of determining which circumstances warrant state intervention in families for the protection of children, and which do not. By rejecting presumptions about the effects of opioid addiction on parents and calling for findings concerning the effects of a parent’s addiction in each case, I don’t seek to make the jobs of hard-working state child-protection workers and of the equally hard-working trial courts even more challenging. But these are exactly the kinds of circumstances that require courts to be exceptionally diligent to ensure that, in every case, we ground our rulings in evidence and law, and not supposition and personal judgments.