Appeal from an order of the Family Court of *1031Broome County (Connerton, J.), entered June 4, 2004, which granted petitioner’s application, in a proceeding pursuant to Family Ct Act article 10, to adjudicate respondents’ child to be abused and neglected.
In the early morning hours of Tuesday, October 28, 2003, respondents’ two-month-old son was screaming, pale, acting strangely, vomiting, refusing to eat and displaying seizure-like symptoms. In particular, on more than one occasion, the child went limp, his eyes rolled back in his head and he would then stare into space. At this time, he was in the care of his father, respondent Sean K. (hereinafter the father), as his mother, respondent Meghan K. (hereinafter the mother), had been working an all-night shift at her place of employment. Later that morning the infant was brought to the doctor’s office where he was examined by board-certified pediatrician Padmini Sagar, who found him to be “acutely ill” and ordered immediate hospitalization. An MRI ultimately revealed that the child suffered from multiple brain bleeds. He was thereafter examined by another board-certified pediatrician, John Andrake, who opined that the injuries were consistent with shaken baby syndrome.
Following an investigation, and based on the lack of any explanation from respondents concerning the injuries to the child, petitioner commenced this proceeding alleging that they abused and neglected him. Following a hearing, Family Court found that respondents abused their child and that the father neglected him by failing to obtain prompt medical care. The child was ultimately placed with his paternal grandmother for a 12-month period and this appeal by respondents ensued.
As relevant here, in order to obtain a determination of abuse, a petitioning agency must show by a preponderance of the evidence that a respondent either “inflict[ed] or allow[ed] to be inflicted upon [a] child physical injury by other than accidental means” (Family Ct Act § 1012 [e] [i]). Furthermore, Family Ct Act § 1046 (a) (ii) “provides that a prima facie case of child abuse or neglect may be established by evidence of (1) an injury to a child which would ordinarily not occur absent an act or omission of [the] respondents, and (2) that [the] respondents were the caretakers of the child at the time the injury occurred” (Matter of Philip M., 82 NY2d 238, 243 [1993]). Once a prima facie case is established, the burden of going forward shifts to the respondent to rebut the evidence of parental culpability (see id. at 244). To be sure, the burden of going forward does not shift the burden of proof, which always rests with the petitioning agency (see id.). Upon our review of the evidence, we are *1032satisfied that petitioner presented a prima facie case of abuse against respondents, that respondents failed to rebut this presumption and that petitioner met its burden of proving both abuse and neglect by a preponderance of the evidence (see Family Ct Act § 1046 [b] [i]). Accordingly, we affirm.
According to Andrake, the child’s MRI showed multiple areas of bleeding in the brain. Such findings, Andrake opined, were evidence of a serious traumatic injury that ordinarily would not occur in an infant. He further opined that the injury was the result of a nonaccidental event for which no explanation was offered and was consistent with shaken baby syndrome. Petitioner further established that the child was in the care and custody of respondents during the time period when the injury occurred and that they were responsible for his care. In our view, this evidence was sufficient to make out a prima facie case of abuse against them (see Matter of Ashley RR., 30 AD3d 699, 700 [2006]; Matter of Keone J., 309 AD2d 684, 686 [2003]; Matter of Magnolia A., 272 AD2d 115, 116 [2000], lv dismissed 95 NY2d 902 [2000]; Matter of Tyeasia C., 227 AD2d 165, 166 [1996], lv dismissed 88 NY2d 1017 [1996]; Matter of Westchester County Dept, of Social Servs. v Felicia R., 215 AD2d 671, 672 [1995], lv denied 86 NY2d 708 [1995]; Matter of F. Children, 178 AD2d 246 [1991]; Matter of Cynthia V., 94 AD2d 773 [1983]).1
In addition to this evidence, petitioner presented further proof to support its allegation against respondents. Andrake testified that symptoms of shaken baby syndrome would have been manifest, if not immediately, then within a few days of the event that caused them. This testimony was consistent with that of Sagar, who surmised that a child with this type of injury would display symptoms within 24 to 48 hours. Here, Lucinda Wingerter, the mother’s stepmother, watched the child for most of the day on Sunday, October 26, 2003 and from thereafter he was in the exclusive care of one or both of the respondents.2
According to Wingerter, the child was happy, smiling a lot, not fussy and able to sleep on that Sunday. Wingerter denied that she ever left the child alone with anyone that day or that anything untoward happened to him. According to both *1033Wingerter and the mother, the child was sleeping soundly when picked up around midnight. Moreover, according to the mother, the child slept through the night that night and did not wake up screaming. Notably, Family Court expressly found that Wingerter did not inflict the injuries on the child and, according due deference to this credibility assessment of the court as the trier of fact, we find no basis to disturb it (see e.g. Matter of Amanda M., 28 AD3d 813, 814 [2006]; Matter of Brandyn P., 278 AD2d 533, 535 [2000]; Matter of Zachary MM., 276 AD2d 876, 881 [2000]).
Petitioner further supported its allegations with proof that the father, a young and first-time parent, would at times become frustrated when the child would cry and could not be consoled. Indeed, testimony established that the child was crankier than usual on Monday. Petitioner also presented evidence, through the testimony of the mother’s stepsister, that respondents had a conversation following the Tuesday morning doctor’s appointment wherein the issue of the father shaking the baby was discussed.3 Significantly, Family Court expressly credited the testimony of the stepsister concerning this conversation.
Respondents could not offer any explanation for the child’s injury (compare Matter of Brandyn P, supra at 534) and attempted to rebut the presumption by each denying that he or she did anything to cause it. Notably, Family Court did not find them credible on this critical point. In holding that both were responsible for abuse of the child, Family Court expressed the following: “Neither of [respondents] betrayed emotion [throughout the numerous court appearances and three days of hearings] including during Dr. Andrake’s testimony. Their affect is flat. Their testimony was delivered matter-of-factly. It is not possible ... to discern which of them is responsible for [the child’s] trauma since neither of them admits to wrong-doing.”
Respondents also attempted to rebut the presumption by pointing out that various members of their extended family were with the child during the days leading up to October 28, 2003. Family Court was ultimately unpersuaded (compare Matter of Ashley RR., 30 AD3d 699, 700-702 [2006], supra; Matter of Zachary MM., supra at 881), and we find no basis upon which to disturb this finding, which was based primarily on the court’s resolution of credibility. While proof that a child is not in a *1034parent’s care when an injury occurred is indeed one way by which the presumption can be rebutted (see Matter of Philip M., 82 NY2d 238, 244 [1993], supra), respondents proof in this regard fell short. First, they presented no expert medical proof to pinpoint the timing of the child’s injury to a time period when neither of them was with the child.4
Next, at the hearing, respondents never specifically accused any particular adult family member of inflicting the injury (see Matter of Westchester County Dept. of Social Servs. v Felicia R., 215 AD2d 671, 672 [1995], supra; compare Matter of Ashley RR., supra). They did, however, present evidence attempting to imply that the mother’s seven-year-old stepbrother, who suffers from attention deficit disorder and has behaved inappropriately with his parents and peers in the past, may have injured the child. However, there was not one shred of credible evidence to suggest that this seven-year-old child had the opportunity or propensity to injure any infant, let alone the subject child. Indeed, there was no evidence that this seven year old had ever even held the child, let alone shook him with such force as to cause his brain to bleed. Moreover, his mother (i.e., Wingerter) unequivocally testified that he was never alone with the child on October 26, 2003. Notably, Family Court found her testimony on this issue to be “straightforward and credible.”
With respect to Wingerter herself and whether she may have inflicted the injury on the child, Family Court specifically found that “[respondents never suggested, much less proved that Lucinda Wingerter intentionally injured [the child] on October 26, 2003.” The record fully supports this finding as it reveals that respondents themselves were perfectly comfortable with her caring for their child. In fact, the mother looked to Wingerter, who was a registered nurse and herself a mother of three, for parenting advice on more than one occasion during the infant’s first few months of life. Indeed, it was at Wingerter’s suggestion on the morning the child was first taken to the doctor that an immediate referral for a CT scan be ordered. Thus, while Family Court declined to credit certain portions of Wingerter’s testimony, the court expressly discounted any notion that she was responsible for the child’s injury. Additionally, while there was testimony that another family member was with the child on the Saturday preceding his hospitalization, this fact, without more, is wholly insufficient to rebut the evidence offered by petitioner.
*1035In short, we are satisfied, as was Family Court, that respondents failed to rebut the prima facie case against them. We are further satisfied, upon our review of the entire record and according due deference to the court’s resolution of credibility determinations, that the finding that the child was abused by respondents is based on a preponderance of the evidence (see Family Ct Act § 1046 [b] [i]). Furthermore, by failing to secure prompt medical treatment for the child on the morning of October 28, 2003 despite the severity of the symptoms he was then exhibiting, the father did not act as “an ordinarily prudent and loving parent, ‘solicitous for the welfare of his child and anxious to promote [the child’s] recovery’ ” (Matter of Hofbauer, 47 NY2d 648, 655 [1979]), and thus Family Court also properly found that he neglected the child.
To the extent not discussed, respondents’ remaining contentions have been examined and found to be unpersuasive.
Cardona, EJ. and Rose, J., concur.
Crew III, J. (concurring in part and dissenting in part). We have no quarrel with the majority’s finding that respondent Sean K. (hereinafter the father) neglected the child by failing to secure prompt medical treatment on the morning of October 28, 2003. Plainly, any reasonably prudent parent (see e.g. Matter of Joseph DD., 214 AD2d 794, 794-795 [1995]; Matter of Robert YY., 199 AD2d 690, 692 [1993]) confronted with a two-month-old child who was screaming, pale, acting strangely, vomiting, refusing to eat and displaying seizure-like symptoms would have summoned emergency medical aid and would not have waited approximately 90 minutes for the child’s other parent to arrive home from work to assess the situation.
We reach a contrary conclusion, however, with regard to the finding that the father and respondent Meghan K. (hereinafter the mother) abused their child. To sustain such finding, the majority relies upon the presumption set forth in Family Ct Act § 1046 (a) (ii), which “provides that a prima facie case of child abuse or neglect may be established by evidence of (1) an injury to a child which would ordinarily not occur absent an act or omission of [the] respondents, and (2) that [the] respondents were the caretakers of the child at the time the injury occurred” (Matter of Philip M., 82 NY2d 238, 243 [1993]; see Matter of Keone J., 309 AD2d 684, 686 [2003]). In this regard, we agree with the majority that the testimony offered by John Andrake is sufficient to satisfy the first prong of the statutory test— namely, that respondents’ child sustained a serious traumatic injury that ordinarily would not occur in an infant. We do not, however, agree that the record as a whole establishes that the *1036child’s injury occurred while he was in the care of respondents. In our view, the second prong of the statutory presumption has not been met and, hence, the presumption cannot apply under the particular facts of this case. Moreover, even assuming that the presumption indeed applies here, we nonetheless are persuaded, for the reasons that follow, that such presumption was rebutted and, further, that the record as a whole simply does not support a finding that respondents abused their child. Accordingly, to that degree, we respectfully dissent.
As a starting point, the medical testimony adduced at the hearing does not establish, with any degree of precision, when the child’s injury occurred. In this regard, the record reflects that the underlying MRI study was conducted on October 30, 2003 and that Andrake examined the child and reviewed such results upon the child’s admission to the hospital on October 31, 2003. As noted previously, that MRI study revealed multiple brain bleeds which, Andrake acknowledged, could have been there for “at least a few days.” When questioned regarding the interval between the infliction of the child’s injury and the onset of symptoms, Andrake testified that this depended upon the severity of the injury sustained. When asked whether the interval between the injury and the symptoms could be as many as three days, Andrake replied, “Yes.”1 This testimony was consistent with that offered by Padmini Sagar, the physician who initially examined the child in the pediatrician’s office, who surmised that a child with this type of injury would display symptoms within 24 to 48 hours.
*1037Adopting what it apparently construed as a definitive three-day time line, Family Court, working back from Andrake’s evaluation of the child on October 31, 2003, reasoned that the child’s injury occurred on or after October 28, 2003 and, hence, occurred while he was in the exclusive care and custody of respondents—specifically, the father. The flaw in this analysis is twofold. First, although Andrake examined the child on October 31, 2003, the MRI study was conducted the previous day. Using the MRI study as a starting point and working “a few days” back from that date potentially places the child’s injury on or about October 27, 2003. Second, Family Court’s analysis and time line ignores the fact that the child plainly was in acute distress on the morning of October 28, 2003, screaming, vomiting and displaying seizure-like symptoms. While such symptoms most assuredly could be evidence of a recently inflicted injury, i.e., one that occurred that very morning, Andrake’s and Sagar’s respective testimony roughly establishes a 24 to 72-hour window within which the child’s injury could have occurred, which means that the child’s injury could have been inflicted as early as October 25, 2003.
In this regard, although the record plainly reflects that the child was with one or both of his parents from 12:00 a.m. on Monday, October 27, 2003 until the time of his admission to the local hospital on Tuesday, October 28, 2003, the record also indicates that the child was being cared for by Lucinda Wingerter, the mother’s stepmother, from at least 1:00 p.m. on Sunday, October 26, 20032 until the mother retrieved the child at 12:00 a.m. on Monday, October 27, 2003. Wingerter’s oldest son, David, also watched the child for a period of time on Saturday, October 25, 2003. As to respondents’ respective whereabouts during this time period, the father was either working or out of town from 7:00 a.m. on Sunday, October 26, 2003 until approximately 4:30 p.m. on Monday, October 27, 2003. The father and the mother were home with the child from this point until the mother left for work at approximately 10:45 p.m. on Monday, October 27, 2003, after which the father was alone with the child until the mother returned home from work around 7:00 a.m. on Tuesday, October 28, 2003. As noted previously, the child was with Wingerter from at least 1:00 p.m. on Sunday, October 26, 2003 to 12:00 a.m. on Monday, October 27, 2003, at which point the mother was alone with the child until the father’s return around 4:30 p.m. that afternoon.
*1038Simply put, given the somewhat elastic time period within which the child’s injury potentially could have been inflicted, and in view of the fact that the child was cared for by at least four adults—the mother, the father, Wingerter and her oldest son, David—between Saturday, October 25, 2003 and Tuesday, October 28, 2003, we cannot conclude that petitioner established that the child’s injury was inflicted while he was being cared for by respondents (see Matter of Ashley RR., 30 AD3d 699, 701 [2006])—or more particularly, as the majority apparently believes, by the father. Such a finding, in our view, would be based upon pure conjecture. Our conclusion in this regard does not stand for the proposition that the presumption outlined in Family Ct Act § 1046 (a) (ii) only applies to cases in which a child is never out of the presence of his or her parents during the critical time period when an injury is sustained. However, where, as here, any number of individuals have access to the child during the relevant time period and it is equally likely that the underlying injuries could have been inflicted by any one of those individuals as the other, the presumption simply cannot be invoked. Thus, inasmuch as petitioner failed to establish that “respondents were the caretakers of the child at the time the injury occurred” (Matter of Philip M., 82 NY2d 238, 243 [1993], supra), the statutory presumption should not have been applied and the finding of abuse cannot be sustained.
Moreover, even assuming application of the presumption indeed is compelled here, we nonetheless conclude, for all the reasons just discussed, that the record as a whole fails to establish, by a preponderance of the evidence (see Family Ct Act § 1046 [b] [i]), that respondents abused their child. In reaching this conclusion, we have not considered the testimony offered by Wingerter and her daughter regarding conversations wherein they allegedly overheard the mother question the father as to whether the child had been shaken, as such testimony fails to shed any light upon when the child’s injury was inflicted. Accordingly, we would modify Family Court’s order by reversing so much thereof as adjudicated respondents’ child to be abused.
Mugglin, J., concurs. Ordered that the order is affirmed, without costs.
. Indeed, Family Ct Act § 1046 (a) (ii) is closely analogous to the “res ipsa loquitur” rule in the law of negligence (see Matter of Philip M., supra at 244).
. While Andrake did testify that symptoms of shaken baby syndrome could take up to three days to manifest themselves, his testimony makes relatively clear that this interval only applies when a minor injury is inflicted. Here, no physician described the child’s injury as minor. To the contrary, both Andrake and Sagar described the child’s injury as serious with Sagar specifically testifying that the child was acutely ill requiring immediate hospitalization.
. The stepsister specifically testified that, on the way home from the child’s doctor appointment on October 28, 2003, the mother told the father that she “really wanted to tell the doctor that you did it, just so that— . . . just so we don’t get in trouble,” or words to that effect. The father’s only response to the mother’s comment was “oh.”
. While Andrake was regrettably never asked by any attorney when he believed the subject child’s injuries were inflicted, the absence of such a question and answer is not fatal to a finding of abuse against respondents.
. In our view, the relative clarity of Andrake’s testimony with regard to the time frame at issue is best assessed by his actual statements. “In part it depends on the degree of injury. There’s sometimes when the injury is severe enough from the shaking to the presentation is immediate, so bad that they stop breathing then and there and if they don’t get medical attention they don’t live. There’s others who are more mild, and they can progress with more minor symptoms over the course of days before it’s recognized.”
Further questioning of Andrake on this point elicited the following testimony.
“Q. All right. Now, you mentioned that depending on the severity of the trauma, the symptoms could show up immediately or they could show up days after the trauma occurred?”
“A. Correct.”
“Q. It could be as many as three days?”
“A. Yes.”
While no one questions that the child’s injuries indeed were serious, Andrake never was asked, based upon his observations of the child, the history given and the symptoms presented, when he believed the injuries were inflicted. While the majority ascribes the failure to pinpoint the timing of the child’s injury to respondents, the fact remains that petitioner bore the burden of proof here.
. Wingerter testified that the mother dropped the child off at her residence around 1:00 P.M. on Sunday, October 26, 2003, whereas the mother testified that Wingerter picked the child up at respondents’ residence at 10:00 A.M. that day.