¶ 32. dissenting. The majority’s decision fails to reconcile the family court’s ultimate conclusion that parents did not abuse Z.L. with the court’s finding that the State presented much more probable evidence than parents as to how Z.L.’s injury arose. Because the court found that Z.L. suffered a severe nonaccidental injury while in parents’ exclusive care, the only possible conclusion that the trial court could have reached is that it is more probable than not that parents were responsible for inflicting serious abuse on Z.L. The State met its burden of demonstrating by a preponderance of the evidence that Z.L. is a child in need of care or supervision (CHINS). The evidence of parents’ apparent good character did not rise to a level sufficient to rebut the court’s findings on the credibility of the medical evidence, which the trial court found conclusively established that Z.L. was injured by nonaccidental means. Because I believe that the standard of proof was met in this case, I dissent.
¶ 33. Our primary concern in a CHINS proceeding is to protect a child from future harm. See E.J.R. v. Young, 162 Vt. 219, 222-23, 646 A.2d 1284, 1286 (1994) (explaining that CHINS adjudications are both remedial and preventative). Given the State’s strong interest in “protecting a child from the risk of serious and potentially irrevocable harm,” In re A.D., 143 Vt. 432, 436, 467 A.2d 121, 124 (1983) (quotation omitted), the State has the burden of proving its case “by a preponderance of the evidence.” In re D.T., 170 Vt. 148, 156, 743 A.2d 1077, 1084 (1999). “The State may do so by establishing any combination of statutory factors, including abandonment, abuse, or lack of parental care or subsistence necessary for the child’s well-being.” In re J.R., 164 Vt. 267, 270, *305668 A.2d 670, 673 (1995); see 33 V.S.A. § 5502(a)(12)(A)-(B)3 (delineating that a child is a CHINS if he has been abused or neglected by a parent). In this case, the sole impetus for the CHINS filings was Z.L.’s severe skull and brain injuries. The State filed petitions on behalf of both children, alleging that Z.L. was a CHINS because he was abused by his parents, and that M.L. was without proper parental care. 33 V.S.A. § 5502(a)(12)(A)-(B).
¶ 34. That Z.L. suffered severe and extensive injuries was not disputed. Z.L. had two skull fractures, one running from ear to ear and the other extending from behind his right ear to the base of his skull. He also had a large acute subdural hematoma on the left side of his brain, and a chronic subdural hematoma on the right side of his brain. The main issue at trial was the cause of Z.L.’s injuries, and, as the trial court recognized, the State and parents presented “two possible scenarios.” Either Z.L. was abused by his parents, or he was injured during an accidental fall involving his sister M.L. According to the State, Z.L. was injured during a series of nonaccidental injuries. The State presented the testimony of six experts, five of whom were treating physicians — four doctors who treated Z.L. in the hospital, the children’s regular pediatrician, and a pediatrician with extensive experience in child abuse. All of these experts agreed and unequivocally stated that Z.L.’s injuries were caused by severe head trauma. Because Z.L. had two skull fractures in different orientations, and both old and fresh subdural hematomas, the State’s experts explained that Z.L. had suffered injury on more than one occasion. The State’s experts all agreed that Z.L.’s injuries could not have resulted from an accident with his older sister, and the trial court did not reject this evidence.4 The experts testified that Z.L.’s injuries were simply too severe, and the injury did not fit *306a compression-type accident, as would have resulted from a fall with M.L. In addition, they concluded that Z.L. suffered injuries at distinctly different points in time and that the parents’ evidence did not adequately explain Z.L.’s multiple injuries over time. The State’s experts agreed that the injuries were the result of abuse.
¶ 35. Parents’ explanation was that some of Z.L.’s injuries were caused by his sister falling while holding him and crushing his head between her and the floor, and that Z.L.’s chronic subdural hematoma was the result of preexisting medical conditions that made Z.L. more prone to brain injury. Parents’ experts allowed that parents’ explanation was a possibility, although they could not confirm it, and the experts therefore did not conclude that Z.L. actually had a preexisting medical condition that would have resulted in a chronic subdural hematoma.
¶ 36. The court recounted the expert testimony on both sides and concluded: “if this was simply a battle of the experts the court would agree with the state that the injuries to Z.L. were probably the result of non-accidental means.” Ultimately, the court found the State’s medical evidence more probable than that submitted by parents. Nonetheless, the court concluded that other nonmedical evidence, which was more circumstantial and less direct than the State’s witnesses, persuaded it that Z.L. was not abused by his parents. The evidence the court relied on included: parents’ stable relationship; a lack of history of abuse in the family or involvement with DCF; a lack of evidence of other instances of abuse of either M.L. or Z.L.; and reports from teachers and caregivers that M.L. is well-adjusted and has no behavioral problems. The court thus denied the CHINS petition.
¶ 37. The court’s conclusions, however, are not consistent "with its findings, and at a minimum the case should be remanded for the trial court to attempt to reconcile the discordancies between *307its findings and conclusions. Without further explanation and on the record we have, the decision cannot be affirmed.
¶ 38. In a strikingly similar case, a Louisiana court of appeals reversed a trial court’s conclusion that an infant was not abused where the trial court based its decision on nonmedical evidence of the parents’ loving nature. State ex rel. W.H.V. v. J.A.V., 811 So. 2d 189 (La. Ct. App. 2002). In that case, a three-week-old infant was brought to the hospital after the mother noticed the baby’s body was twitching on one side. Scans and x-rays revealed an intracranial hemorrhage, fractured ribs, a fractured ankle, and a fracture in the thumb. The Department of Social Services sought to adjudicate the child as one in need of care based on parental abuse, and the court held a trial on the matter. As in Vermont, the standard of proof in Louisiana at such a proceeding is a preponderance of the evidence. Medical experts testified that the brain bleed was caused by trauma, that it could not have been caused by the child’s birth, and that the child was a victim of shaken baby syndrome. The baby’s parents and grandmother testified that the infant was in their exclusive care since birth, and they could provide no explanation for the injuries other than hypothesizing that they were caused by a difficult birth. The trial court concluded that the Department had failed to prove that the child was in need of care, citing the lack of certainty of the medical evidence, and nonmedical factors such as the loving nature of the parents, lack of evidence of other abuse, and parents’ prompt action in seeking medical attention. Id. at 197.
¶ 39. The appellate court reversed, noting that even though the medical experts could not explain exactly how the infant was injured, they were in agreement that the injury was caused by abuse. Id. The court emphasized that the uncontroverted medical evidence demonstrated that “some type of external trauma injured this infant on more than one occasion while in the exclusive care and control of his parents and grandmother.” Id. As to the nonmedical evidence, the appellate court explained: ‘We must not allow a trial court’s belief in the ‘goodness’ of the parent(s) to overshadow the overwhelming medical evidence that this child sustained serious and even life-threatening injuries sometime between his release from the hospital following his birth and nearly four weeks later . . . .” Id.
¶ 40. The facts compel a similar conclusion in this case. The State presented medical proof that Z.L. was injured by *308nonaccidental means. As factfinder, the trial court certainly could have found that the State’s experts were not credible or that the parents’ medical experts were more convincing. The court concluded that it was not, however, persuaded by the parents’ experts and that “a fall involving M.L. was probably not the cause of the extensive acute damage to the skull and brain.” The court further concluded that the State’s experts were more persuasive and that Z.L.’s injuries “were probably the result of non-accidental means.” Having reached this conclusion, it was beyond the court’s discretion to ignore its findings because the parents also appeared to be good people. Under our standard of proof, competent, credible medical evidence of abuse cannot be overcome by testimony that parents appear to be caring and attentive. See In re Marcus H., 697 N.E.2d 862, 867 (Ill. App. Ct. 1998) (reversing trial court’s conclusion that infant was not abused even though the court found medical evidence credible that baby’s injuries were not consistent with an accident and explaining that once the trial court found that the injuries were sustained by nonaccidental means the court could not disregard this medical evidence). The conclusions the court reached based upon the direct medical evidence are the most important in its analysis. As outlined above, a CHINS finding is compelled once the court finds that a child has suffered nonaccidental injuries while in a parent’s care. See In re J.R., 164 Vt. at 270, 668 A.2d at 673. On these facts, the court’s finding that the State’s account of Z.L.’s injury was more probable must result in a CHINS conclusion. See Livanovitch v. Livanovitch, 99 Vt. 327, 328, 131 A. 799, 800 (1926) (explaining that a preponderance-of-the-evidence standard is met when “the beam inclines toward him who has the burden, however slightly”). No conclusion other than that Z.L. was abused is possible from this finding.
¶ 41. The trial court appeared unable to accept that parents, whom witnesses presented as attentive and caring parents, could have been capable of inflicting the injury that the medical evidence indicated Z.L. had suffered. The fact that parents were viewed positively by community members is not irreconcilable, however, with the State’s theory of abuse, as the court appears to suggest. Child abuse is a pernicious societal problem that may afflict families of every strata and commonly occurs in private, closed settings. Individuals who appear to be “good parents” can abuse their children — even if only in a momentary loss of temper — and parents can abuse one child and not the other. *309Indeed, given the findings of the trial court, that is the only rational explanation for what occurred here.
¶ 42. The trial court’s finding that Z.L. was not injured by a fall with M.L. and its concurrent conclusion that Z.L. was not abused were not reconciled by the court and left unanswered the key question of what happened on the day that Z.L. was injured and what caused his previous injuries. Given the court’s determination of the credibility of the medical evidence, the only answer to this critical question is that Z.L. was abused by his parents. General evidence about parents’ apparently good nature cannot diminish the “compelling” nature of the State’s medical evidence regarding the specific injuries inflicted here, or enhance the plausibility of parents’ theory, which the court rejected as merely possible, but not probable. See Porter v. Am. Exp. Lines, Inc., 387 F.2d 409, 410 (3d Cir. 1968) (“Factual determination rests on probability.”).
¶43. In sum, the court’s findings — that Z.L.’s injuries were the result of nonaccidental serious head trauma and that Z.L. was under the exclusive care of his parents during the time of his injuries — support only one conclusion: that it is more probable than not that Z.L. suffered abuse at the hands of his parents. I would therefore reverse the trial court, and order that Z.L. and M.L. are CHINS.
As stated by the majority, the juvenile statutes have recently been amended and this provision now appears at 33 V.S.A. § 5102(3). Ante, ¶ 6 n.2.
The trial court was critical of the State’s lack of any expert testimony on the exact amount of force necessary to break an infant’s skull or created by an eighty-pound child falling on an infant. While none of the State’s experts provided this information in terms of numbers, equivalent evidence was admitted in another form. The State was not limited to proof of the precise numbers that identify the force that would be created by M.L. falling on Z.L. to prove that the fall did not cause Z.L.’s extensive injuries. Instead, the State’s experts concurred that it was simply impossible for an accidental fall involving a child of a weight equal to M.L. to have caused Z.L.’s injuries. The experts likened the amount of force necessary *306to cause Z.L.’s head injuries to the forces created from: a car accident, falling down a long flight of stairs, a ski accident, or shaking or throwing a child. Although the family court may have wished it had more specific evidence of how Z.L. was injured, such detail would be cumulative to the evidence provided by the State’s experts that was certainly adequate to meet our preponderance threshold for the State’s burden of proof, as the majority has correctly defined it. When courts consider the injuries inflicted through child abuse, there frequently is no witness to the event. As another court acknowledged, there are rarely eyewitnesses to child abuse, and therefore these cases often rely on medical expert testimony and circumstantial evidence. See State v. Goblirsch, 246 N.W.2d 12, 14 (Minn. 1976).