{1} This criminal prosecution under NMSA 1978, Section 30-6-1 (2005) alleged intentional and negligent child abuse causing great bodily harm to an infant. A jury returned a guilty verdict, which our Court of Appeals affirmed in a memorandum opinion. State v. Consaul, No. 29,559, mem. op. at 2 (N.M. Ct. App. Feb. 20, 2012) (non-precedential).
{2} In our certiorari review of that conviction, we hold first that under the particular circumstances of this case, the district court erred when it rejected defense counsel’s request for separate jury instructions for *533intentional and negligent child abuse. Second, we hold that the evidence offered to support the charge of criminally negligent child abuse resulting in great bodily harm (which we hereafter describe as acts in reckless disregard) failed, according to the State’s own witnesses, to prove that Defendant’s actions caused the infant’s injuries. Third, we hold that the evidence of intentional child abuse resulting in great bodily harm — in this case, an allegation that the accused intentionally suffocated the infant — failed to prove that charge beyond a reasonable doubt.
{3} Accordingly, we reverse the district court and order the charges dismissed with prejudice for lack of sufficient evidence. In the course of our opinion we revisit and modify portions of our jurisprudence regarding the crime of child abuse under Section 30-6-1.
BACKGROUND
{4} Defendant Daniel Consaul was convicted of child abuse resulting in great bodily harm after his nephew, Jack Consaul, suffered a devastating neurological injury when he was ten weeks old.
{5} During the early morning hours of October 10, 2005, Daniel was responsible for watching Jack while Jack’s mother, Heidi Consaul, worked a night shift at a local motel. Daniel helped his sister with caretaking duties for Jack and had been doing so since Jack was born. Daniel was also a student at Doña Ana Branch Community College studying fire science.
{6} Daniel began watching Jack during the evening hours of October 9,2005, after Jack’s grandparents dropped him off at the apartment that Daniel and Heidi shared. During this time and into the morning hours of October 10, 2005, Jack became ill. Daniel was the only person with Jack when this occurred, and we derive an account of the events that evening from two statements Daniel gave to law enforcement officers during the subsequent child abuse investigation.
{7} Daniel recounted that on the evening of October 9, 2005, he had been at the New Mexico State University campus visiting friends. He headed home when he learned that he needed to watch Jack because Heidi was going to work early that evening. In his first interview, Daniel stated that he laid Jack down to sleep about 11:30 p.m. In Daniel’s second interview, he informed law enforcement officers that he put Jack “in his crib face down because [Jack] was aggravating me with his crying . . . .” Daniel stated that he bundled Jack in a blanket “a little tighter . . . than I normally did,” and responded affirmatively when asked if he did this because he was frustrated and wanted Jack to fall asleep so he could complete the tasks his sister had given him. When asked by a law enforcement officer to describe how he bundled Jack and put him to sleep that evening, Daniel responded that he put Jack face down in the crib on a pillow.
{8} Around 1:30 a.m., Jack screamed. Daniel discovered that Jack had thrown up and appeared rigid, so he called Heidi and their neighbor. When Heidi arrived home, she, Daniel and the neighbor rushed Jack to the emergency room at Memorial Medical Center in Las Cruces, where Dr. Martin Boyd, an emergency medicine doctor, examined him. Dr. Boyd testified that Jack arrived shivering and lethargic, but he was “appropriately responsive.” Dr. Boyd ordered lab tests and began administering intravenous (IV) fluids. Dr. Boyd thought something was really wrong *534with Jack, so he contacted Dr. Hernán Ciudad, the pediatrician on call, for a consult.
{9} Dr. Ciudad arrived at the hospital around 4:30 a.m. He could not pinpoint exactly what was wrong with Jack, but believed that Jack could be in septic shock. Dr. Ciudad continued to give Jack IV fluids, reviewed his lab results, and administered a spinal tap. Unsure of what was wrong with Jack, Dr. Ciudad called for help, contacting Dr. Dawn Joseph, a critical care specialist at University of New Mexico Hospital (UNMH) in Albuquerque. Both doctors agreed that Jack should be airlifted to UNMH.
{10} When Jack arrived at UNMH, Dr. Joseph examined him. She believed that Jack was in septic shock or suffering from an infection, like meningitis. Dr. Joseph ordered that antibiotics be administered to treat any potential bacterial infection and fluids to treat the shock. During his second day at UNMH, Jack had a seizure. Antiseizure medication did not help control Jack’s seizure activity. According to her testimony, at that point Dr. Joseph was very worried that something else was going on with Jack. Following the seizure, Dr. Joseph ordered a CT scan and an MRI. Upon receiving the results, Dr. Joseph realized that Jack had a severe neurological injury, specifically a hypoxic ischemic injury to his brain1, which is a lack of oxygen and blood to the brain, and cerebral edema, which is a swelling in the brain. Dr. Joseph called Dr. Mary Johnson, a pediatric neurologist at UNMH, for a consult.
{11} Almost immediately, Dr. Johnson became concerned that Jack’s brain injuries were “non accidental,” specifically, the result of suffocation. In other words, Dr. Johnson suspected child abuse. Once she suspected child abuse, Dr. Johnson “felt that the most likely mechanism, if we talk about mechanisms of brain injury in small children in child abuse, the most likely mechanism was that of suffocation.” Dr. Joseph agreed with Dr. Johnson, stating that suffocation “could be the one diagnosis or the one thing that happened to Jack that could explain everything.”
{12} Believing that Jack was a victim of child abuse in the form of “non accidental” suffocation, UNMH staff alerted the New Mexico Children, Youth and Families Department (CYFD), which then notified law enforcement authorities. With a child abuse investigation underway, Detective Mark Myers of the Las Cruces Police Department interviewed Daniel twice, on October 13 and 14, 2005. Based on the October 14 interview and the doctors’ suspicions that Jack was a victim of child abuse, a district judge issued an arrest warrant for Daniel on October 25, 2005.
{13} On October 27, 2005, a grand jury indicted Daniel on charges of negligently caused child abuse resulting in great bodily harm, and in the alternative, intentional child abuse resulting in death or great bodily harm pursuant to Sections 30-6-1 (D) and (E).
{14} Importantly, the theory stated in the indictment for both negligent and intentional child abuse was that Daniel had placed Jack in a situation that endangered Jack’s life or health “by swaddling Jack . . . tightly and leaving him unattended for an extended period of time . . ..” Absent from the indictment was *535any mention of Dr. Johnson’s theory that Daniel had non-accidentally suffocated Jack. At the beginning of Daniel’s trial in March 2008, the State’s sole theory of the case, consistent with the indictment, was that Daniel had swaddled Jack tightly and left him unattended for a period of time. During trial, however, the State introduced an alternative theory: that Daniel had intentionally suffocated Jack by holding his face down in a pillow or by placing his hand over Jack’s mouth, either continuously or in several episodes over a period of time.
{15} The defense countered the State’s argument by suggesting that Jack’s injuries occurred because the doctors at UNMH had administered an excessive amount of fluid in treating Jack’s shock. The defense presented one expert to support its theory that the treatment Jack received at UNMH caused his injuries. The defense also suggested that the doctors’ testimony that Jack suffered intentional suffocation was self-serving in that their motive to suggest child abuse was to avoid being held accountable for Jack’s injuries.
{16} At the close of trial, defense counsel requested separate jury instructions for negligent and intentional child abuse. The district court declined and gave the jury one uniform jury instruction for both intentional and negligently caused child abuse. See UJI 14-602 NMRA (uniform jury instruction delineating the essential elements of intentional or negligently caused child abuse resulting in great bodily harm). In addition, defense counsel requested separate verdict forms for intentional and negligent child abuse, which the district court also declined.
{17} In a general verdict, the jury found Daniel guilty of child abuse resulting in death or great bodily harm. The district court sentenced Daniel to the basic term of 18 years imprisonment. Based on the nature of the crime and the resulting harm, the district court determined that this was a serious violent offense pursuant to NMSA 1978, Section 33-2-34 (2006) which specifies which crimes are serious violent offenses and what amount of prison time is eligible for meritorious deductions.
{18} Daniel raises a number of issues on certiorari, two of which we review in this opinion. We determine first whether the district court erred by denying Daniel’s request for separate jury instructions regarding negligent and intentional child abuse. Having determined that the district court erred, we then address whether sufficient evidence at trial supported the jury’s verdict under either theory. Finding these issues dispositive, we do not address the remaining matters submitted in the petition.
DISCUSSION
Under the Circumstances of this Case, Daniel Was Entitled to Separate Jury Instructions Specifying the Act for Which the Jury Found Daniel Guilty
{19} Daniel argues that the district court erred in refusing to give separate jury instructions regarding negligent and intentional child abuse resulting in great bodily harm, thereby denying Daniel a unanimous verdict. The State conceded on appeal that the jury-instruction issue was dispositive. The State based its concession on this Court’s opinion in State v. Cabezuela, 2011-NMSC-041, 150 N.M. 654, 265 P.3d 705. We agree with Daniel that the jury should have received separate jury instructions, *536although not based on our holding in Cabezuela.
{20} In Cabezuela, the defendant was convicted of intentional child abuse resulting in death. Id. ¶ 1. The defendant argued that the jury received an improper instruction regarding the elements of intentional child abuse resulting in death. Id. In Cabezuela we “suggested] that there should be separate instructions for negligent and intentional child abuse.” Id. ¶¶ 36-37.
{21} Despite this language, Cabezuela is not on point with the jury-instruction issue presented in this case which deals with child abuse resulting in great bodily harm pursuant to Section 30-6-1 (E). Cabezuela addressed the statutory differences in punishment between intentional child abuse resulting in death of a child under twelve and negligent child abuse resulting in death of a child under twelve. 201 l-NMSC-041, ¶ 32. The punishment for intentional child abuse resulting in death of a child under twelve is a life sentence, whereas the punishment for negligent child abuse resulting in death of a child under twelve is eighteen years. Id.
{22} In Cabezuela, we focused on the need for separate jury instructions for intentional and negligent child abuse resulting in death of a child under twelve. We also focused on the use of the language “failure to act” regarding intentional child abuse, an issue not present in this case. Id. ¶ 36 (internal quotation marks and citation omitted). In Cabezuela, our focus on the differentpenalties for intentional and negligent child abuse resulting in death of a child under twelve is important, because the differences “indicate[] that the Legislature meant to punish only the most deliberate and reprehensible forms of child abuse under this crime.” Id. ¶ 32; see § 30-6-l(H) (“A person who commits intentional abuse of a child less than twelve years of age that results in the death of the child is guilty of a first degree felony resulting in the death of a child.”), anciNMSA 1978, § 31-18-15(A)(1) (2007) (“[T]hebasic sentence ... for a [noncapital] first degree felony resulting in the death of a child [is] life imprisonment.”).
{23} Here, in contrast, the punishment for child abuse resulting in great bodily harm, whether done knowingly, intentionally, negligently, or recklessly, is the same. Section 31-18-15(A)(3) (“[T]he basic sentence ... for a [noncapital] first degree felony [is] eighteen years imprisonment.”). Notwithstanding this lack of difference in penalty, child abuse resulting in great bodily harm will sometimes also require separate jury instructions, and the case before us presents just such a circumstance. When two or more different or inconsistent acts or courses of conduct are advanced by the State as alternative theories as to how a child’s injuries occurred, then the jury must make an informed and unanimous decision, guided by separate instructions, as to the culpable act the defendant committed and for which he is being punished.
{24} As we now discuss in more detail, the alleged acts of criminally negligent child abuse and intentional child abuse rested on different and inconsistent theories as to what Daniel actually did. Fornegligent child abuse, the State told the jury that Daniel put Jack to bed carelessly (tightly swaddled and placed face down on a pillow), and argued that this act of negligence caused Jack’s injuries. For intentional child abuse, however, the State hypothesized that Daniel did not just put Jack to bed carelessly, but that Daniel actually used a pillow or his hand to suffocate Jack so he could not breathe. But the jury was never *537asked to specify which criminal act Daniel committed. With the jury instruction as given, the jury simply had to agree that Daniel caused Jack to be “placed in a situation which endangered [Jack’s] life or health” or that Daniel “tortured or cruelly punished Jack.” The jury instruction did not require the jury to specify and unanimously agree upon which conduct caused Jack’s injuries.
{25} A statement made during the State’s closing argument to the jury demonstrates the risk of jury confusion if we were to allow these two criminal acts to be commingled in a single instruction. When describing what the jury needed to decide to find Daniel guilty, the prosecutor said:
So you can all go back there and talk about it and you might think, “Gosh, I think he did this on purpose,” or some of you might think, “I just think he’s an idiot and he wanted the baby to shut up, and he put him face down in there,” but everybody knows, and we know he knew better to do that because he wouldn’t tell anyone. If he thought it was just not that big a deal and “Oops, this is what happened,” he would have told immediately.
The prosecutor invited the jury to convict Daniel of child abuse whether or not the jury agreed on what criminal act Daniel actually committed. Jurors should not be left free, let alone encouraged by the prosecutor, each to go his or her own way when it comes to determining what criminal conduct — if more than one act is alleged — caused the child’s harm. The jury needs to agree unanimously on what conduct caused harm to the child.
{26} Accordingly, we agree with the parties that in the particular circumstances of this case Daniel was entitled to separate jury instructions for negligent and intentional child abuse resulting in great bodily harm because the State’s theories of how that harm occurred were different and inconsistent. Failure to give separate jury instructions constituted reversible error.
Criminal Negligence Articulated in Section 30-6-l(A)(3) Means Reckless Disregard
{27} Before we address Daniel’s specific challenge to the evidence, we discuss whether the language currently used in our uniform jury instructions adequately captures the true nature of the crime and the legislative intent behind the statute. We raise this issue sua sponte as a matter of public importance. Although generally, “propositions of law not raised in the trial court cannot be considered sua sponte by the appellate court,” we have previously done so to resolve “questions of a general public nature affecting the interest of the state at large.” State v. Jade G., 2007-NMSC-010, ¶ 24, 141 N.M. 284, 154 P.3d 659 (internal quotation marks and citation omitted). We will also “determine propositions not raised in the trial court where it is necessary to do so in order to protect the fundamental rights of the party.” Id. (internal quotation marks and citation omitted). This is one of those cases.
{28} As we will shortly explain, this Court has struggled over the years to differentiate criminal negligence from civil negligence. Even the legislative definition ofnegligence in the child abuse context contains a confusing mixture of concepts, some more familiar to the civil litigator (e.g., “knew or should have known”) and others more clearly grounded in criminal law (e.g., “reckless disregard”). See § 30-6-l(A)(3) (“‘[Negligently’ refers to *538criminal negligence and means that a person knew or should have known of the danger involved and acted with a reckless disregard for the safety or health of the child.” (emphasis added^). Our own case law has not been a model of clarity. Prosecutors and defense attorneys alike have expressed concern over the potential for jury confusion. We raise this issue now because it potentially impacts people and events all across our state, and the fundamental rights of criminal defendants in particular. Citizens have a right to be fully informed about the contours of activity that society judges to be “morally culpable, not merely inadvertent.” Santillanes v. State, 1993-NMSC-012, ¶ 28, 115 N.M. 215, 849 P.2d 358.
{29} To adequately evaluate our current instructions, we explore the roots of criminal negligence in the child abuse statute. Since its enactment in 1973, the Legislature has defined child abuse as consisting “of a person knowingly, intentionally or negligently, and without justifiable cause, causing or permitting a child to be: (1) placed in a situation that may endanger the child’s life or health; (2) tortured, cruelly confined or cruelly punished; or (3) exposed to the inclemency of the weather.” Compare Section 30-6-1 (D) with 1973 N.M. Laws, ch. 360, § 10 (containing identical language defining child abuse). The Legislature did not define the terms “knowingly, intentionally, or negligently” within this statute.
{30} In Santillanes, this Courtprovided its first meaningful interpretation of the use of the term “negligently” in the child abuse statute. See Santillanes, 1993-NMSC-012, ¶ 12. We asked, “when the legislature has included but not defined the mens rea element in a criminal statute, here the term ‘negligently,’ what degree of negligence is required}?]” Id. In that case, the defendant appealed the district court’s refusal to give his requested jury instruction explaining the concept of criminal negligence based on the Model Penal Code’s definition of criminal negligence. Id. ¶¶ 1, 3. The district court refused and instead gave the jury a civil negligence instruction. Id. ¶ 3.
{31} In reversing the conviction in Santillanes, we were clear that this Court’s role is to engage in statutory construction and apply a judicial interpretation that fully illuminates the Legislature’s intent in enacting the child abuse statute. Id. ¶ 24. W e stated that “[a] criminal statute may not be applied beyond its intended scope, and it is a fundamental rule of constitutional law that crimes must be defined with appropriate definiteness.” Id. ¶ 25. We construed the statutory language in favor of lenity, and stated that “in the absence of a clear legislative intention that ordinary civil negligence is a sufficient predicate for a felony, we conclude that the civil negligence standard, as applied to the child abuse statute, improperly goes beyond its intended scope and criminalizes conduct that is not morally contemptible.” Id. ¶ 28.
{32} The Santillanes opinion made clear that the Court was not invading the powers of the legislative branch but instead was interpreting “the intended scope of the statute as aiming to punish conduct that is morally culpable, not merely inadvertent.” Id. ¶ 28.
We interpreted] the mens rea element of negligence in the child abuse statute ... to require a showing of criminal negligence instead of ordinary civil negligence. That is, to satisfy the element of negligence in Section 30-6-l(C), we require proof that the defendant *539knew or should have known of the danger involved and acted with a reckless disregard for the safely or health of the child.
Id. ¶ 29. We did not provide any further explanation regarding the phrases “knew or should have known” or “reckless disregard.” See id.
{33} Tracking Santillanes, in 1997 the Legislature amended the child abuse statute to define the term “negligently.” 1997 N.M. Laws, ch. 163, § 1. The Legislature provided that “‘negligently’ refers to criminal negligence and means that a person knew or should have known of the danger involved and acted with a reckless disregard for the safety or health of the child.” Id. (emphasis added). This definition of “negligently” remains in the statute today. Section 30-6-1(A)(3).
{34} In 2000, this Court approved amendments to the uniform jury instructions addressing child abuse resulting in death or great bodily harm. UJI 14-602 NMRA Compiler’s Annotations. Like most jury instructions, UJI 14-602 does not track the statute verbatim, but rather tries to direct the jury’s attention to the essential aspects of the crime as explained in non-legalese. Significantly, UJI 14-602 does not use any variation of the word “negligence” in the body of the instruction but instead requires the jury to determine whether the defendant acted with “reckless disregard,” explaining further that reckless disregard means the defendant “knew or should have known the defendant’s conduct created a substantial and foreseeable risk, the defendant disregarded that risk and the defendant was wholly indifferent to the consequences of the conduct and to the welfare and safety of’ a child. Id.
{35} Despite our best efforts, we remain concerned that our present jury instructions do not faithfully capture the true legislative intent behind Section 30-6-1 and that they potentially contribute to jury confusion, resulting in unjust child abuse convictions. Over the years, we have previously noted these concerns and confusion. See State v. Schoonmaker, 2008-NMSC-010, ¶ 45, 143 N.M. 373, 176 P.3d 1105 (“We also acknowledge that ourUJI 14-602 on negligent child abuse appears to be somewhat inconsistent by using a ‘should have known’ standard and then later requiring that the defendant have ‘disregarded [the] risk and ... [been] wholly indifferent to the consequences.” (alterations in original)); State v. Mascarenas, 2000-NMSC-017, ¶ 13, 129 N.M. 230, 4 P.3d 1221 (“The jury instructions failed to sufficiently define the proper negligence standard for child abuse, and there is no way to determine if the jury based their conviction on the terms ‘knew or should have known,’ language typically associated with a civil negligence standard, or on the proper criminal negligence standard which requires that they find defendant acted in ‘reckless disregard’ of the safety of the child.”); see State v. Magby, 1998-NMSC-042, ¶ 15, 126 N.M. 361, 969 P.2d 965 (“[T]here [was] a distinct possibility that the jury understood the applicable negligence standard to criminalize ‘careless’ conduct or perhaps only ‘extremely careless’ conduct.”), overruled on other grounds,Mascarenas, 2000-NMSC-017, ¶ 27.
{36} Recognizing the confusion, we take this opportunity to clarify certain principles, which we will address in detail in the near future as part of our continuing responsibility for the development of uniform jury instructions. As previously explained, the Legislature did not mean to punish ordinary acts of negligence when it amended the child *540abuse statute to require proof of recklessness. Indeed, “negligence represents the least blameworthy of all the culpable mental states, hence the negligent offender is usually not subjected to severe punishment; and there is a tendency in penal codes to recognize negligence only in limited instances.” 1 Charles Torica, Wharton’s Criminal Law § 27 (15th ed. 2013). The Legislature intended to punish acts done with a reckless state of mind consistent with its objective of punishing morally cirlpable acts and not mere inadvertence.
{37} To avoid the confusion that has plagued this area of the law, we believe that what has long been called “criminally negligent child abuse” should hereafter be labeled “reckless child abuse” without any reference to negligence. The jury should be instructed with this terminology alone. Typical definitions of recklessness require an actor to consciously disregard a substantial and unjustifiable risk of such a nature and degree that its disregard involves a gross deviation from the standard of conduct that a law-abiding person would observe in the actor’s situation. See Model Penal Code § 2.02(2)(c) (definition of “Recklessly”).
{38} In this spirit, we expressly modify prior cases, including our own, in which courts have held that recklessness is not the culpability required for the crime of negligent child abuse. In that regard, we specifically overrule that portion of Schoonmaker in which we stated that “Section 30-6-1 evinces a legislative intent to use the concept of criminal negligence, not recklessness, as the standard for negligent child abuse.” Schoonmaker, 2008-NMSC-010, ¶ 44.
{39} W e also share the concern expressed in our precedent over continued use of the phrase “knew or should have known” in our jury instructions because of its close association with principles of civil negligence and ordinary care. We regret our use of that phrase many years ago in Santillanes, which the Legislature subsequently adopted in its 1997 statutory amendments. We note, however, that in Section 30-6-1 (A)(3) the Legislature appeared to capture two standards of mens rea in one sentence when it defined “negligently” as meaning “that a person knew or should have known of the danger involved and acted with a reckless disregard for the safety or health of the child.” § 30-6-l(A)(3) (emphasis added). The Legislature joined these two distinct states of mind with the conjunctive “and.” Taken literally, the text of the statute refers to both ordinary negligence and criminal recklessness all in a single legislative breath. The Legislature cannot rationally have intended such self-contradiction.
{40} We have already explained why recklessness is the concept closest to what the Legislature had in mind. When a literal, textual reading leaves us with such ambiguity, as this surely does, we traditionally apply the rule of lenity in favor of that interpretation more favorable to the defense, much as we applied lenity to this very statute in Santillanes. Following that line of reasoning, it seems to us that the Legislature intended the term “reckless disregard” to prevail when “knew or should have known” conflicts. Accordingly, we are doubtful about the continued vitality of “knew or should have known” in our instructions, a subject this Court will address in the near future. We now turn to the appropriate remedy — whether we remand for a new trial or the charges against Daniel mustbe dismissed for lack of sufficient evidence to support either theory.
*541Remanding for a New Trial Depends on Whether the Evidence at the First Trial Was Sufficient to Support Either Theory Leading to Daniel’s Conviction
{41} Daniel argues that there was not sufficient evidence at trial to support his conviction under either theory. As a matter of law under well-settled precedent, a “[djefendant would be entitled to a dismissal of the charges on remand if the evidence adduced at trial was insufficient to support [his or ]her conviction. . . .” State v. Jojola, 2005-NMCA-119, ¶ 2, 138 N.M. 459, 122 P.3d 43, aff’d, 2006-NMSC-048, 140 N.M. 660, 146 P.3d 305. To avoid any double jeopardy concerns, we review the evidence presented at the first trial to determine whether it was sufficient to warrant a second trial. See Cabezuela, 2011-NMSC-041, ¶ 40 (“We . . . address Defendant’s sufficiency of the evidence claim to determine whether a retrial would implicate double jeopardy protections.”); State v. Rosaire, 1996-NMCA-115, ¶ 20, 123 N.M. 250, 939 P.2d 597 (“[0]ur review of the sufficiency of the evidence is analytically independent from the issue of the defect in the jury instruction.” (citation omitted)), aff'd, 1997-NMSC-034, 127 N.M. 701, 945 P.2d 66.
{42} When reviewing a verdict for sufficient evidence, we employ a deferential standard in favor of the jury’s verdict. State v. Dowling, 2011-NMSC-016, ¶ 20, 150 N.M. 110, 257 P.3d 930. We view the evidence “in the light most favorable to the State, resolving all conflicts and making all permissible inferences in favor of the jury’s verdict.” Id. “It is our duty to determine whether any rational jury could have found the essential facts to establish each element of the crime beyond a reasonable doubt.” Id. W e must also determine “whether substantial evidence of either a direct or circumstantial nature exists to support a verdict of guilt beyond a reasonable doubt with respect to every element essential to a conviction.” State v. Sena, 2008-NMSC-053, ¶ 10, 144 N.M. 821, 192 P.3d 1198 (internal quotation marks and citation omitted). In doing so we “take into account both the jury’s fundamental role as factfinder in our system of justice and the independent responsibility of the courts to ensure that the jury’s decisions are supportable by evidence in the record, rather than mere guess or conjecture.” State v. Flores, 2010-NMSC-002, ¶ 2, 147 N.M. 542, 226 P.3d 641.
The State Failed to Present Sufficient Evidence of Causation in Regard to Its Theory of Criminally Reckless Conduct
{43} To prove the crime of “criminally negligent child abuse,” or reckless disregard, the State had to establish that Daniel’s actions resulted in great bodily harm to Jack. As argued by the State, that conduct supposedly was Daniel’s freely admitted act of swaddling Jack tightly in a blanket and placing him face down on a pillow in the crib. In other words, the State had to prove that the act of swaddling Jack tightly in a blanket and placing him face down on a pillow actually caused the great bodily harm that Jack suffered. The State also had to offer evidence that such conduct amounted to a reckless disregard for Jack’s welfare.
{44} We only address the element of causation because as we discuss below, the State’s theory breaks down due to lack of supporting evidence. The State’s own experts testified almost unanimously that the mere act of swaddling Jack and placing him face down on a pillow would not have caused the severe brain injuries they observed in Jack.
*542{45} Dr. Johnson, a pediatric neurologist at UNMH, testified for the State about whether swaddling Jack and placing him face down on a pillow could have caused his neurological injuries. The State asked, “[w]ould placing Jack — wrapping him tightly in a blanket, swaddling him, so that his arms are down, and placing him face down on an adult sized pillow for more than an hour, could that cause the kind of injuries that Jack had?” Dr. Johnson responded, “[t]hat was not the initial history” and that she “would be doubtful of that.” When asked by the State whether intentionally placing a child face down on a pillow would result in oxygen loss to the brain, Dr. Johnson responded, “I think it would be kind of unlikely, but this is a two- and-half-month-old who was able to lift their [sic] head and move their [sic] head back and forth, had had a previous- two-month well-child checkup, and he was developing normally.” Dr. Johnson also stated that Jack had good head control, “[a]nd to have this kind of significant injury, I think, would be very, very unusual in a child this age by just having him placed face down.”
{46} Dr. Denise Coleman, a pediatric critical care specialist and child abuse expert, also testified for the State that swaddling Jack and placing him face down on a pillow would not be enough to cause his brain injuries. Dr. Coleman testified:
[I]t’s — the putting him down face down in the pillow for a short amount of time in a baby that can lift their [sic] head up, would not have given this severity of injury. We would have found other injuries in his body. Meaning, his liver would have been more damaged, his kidneys would have been more damaged, there would have been more hypoxic injury to the rest of his body.
(Emphasis added.) Dr. Coleman explained that she did not think that swaddling Jack and placing him face down
sufficiently explained] the amount of injury that we found. . . . [I]f he was placed face down in the pillow for an hour-and-a-half or an hour, and enough to cause this kind of injury to his brain, it would have been a near-miss SIDS [sudden infant death syndrome]. I mean, he would have been found down and unresponsive. Meaning, you would not have been able to wake him up. He would not have woken up crying.
{47} Doctors Johnson and Coleman were two of the State’s principal witnesses called to support its theory that swaddling could have caused Jack’s injury, but neither of them did so. However, we note that Dr. Blaine Hart, a neuroradiologist but not a child critical care specialist, examined Jack’s brain scans, and he was the only doctor to provide minimal support for the State’s theory. When the State asked Dr. Hart if Jack’s brain injury “[wjould ... be consistent with a baby being placed face down on a pillow for about an hour or so,” Dr. Hart responded, “[i]t could be. There are a variety of possible causes, but a suffocation, inadequate air supply is a possibility, yes.” The State never asked Dr. Hart about Jack’s ability at ten weeks old to lift his head and avoid suffocation or the fact that when Jack woke up he was crying, contrary to the expected symptoms of sudden infant death syndrome. In fact, Dr. Hart testified that he did not know exactly what caused Jack’s hypoxic ischemia. Dr. Hart further acknowledged that he was “not a *543treating physician .. . and it’s not appropriate for me to address the clinical care because that’s not what I do. I haven’t done it for 25 ' years.”
{48} Clearly the State failed to prove causation by anything approaching substantial evidence. The State even admitted during a bench conference that its own medical evidence disproved causation, stating that “[t]he intentional act is based on the medical evidence that you can’t do this merely by accident, merely by setting a baby upside down.”
{49} Without any proof of causation, the charge of criminal negligence (or now criminal recklessness) completely fails for lack of substantial evidence, and we need not examine any evidence offered that Daniel was allegedly negligent — or reckless — in the way he put Jack to sleep. We observe, however, that the State utterly failed to produce any such evidence. In fact, no one testified to any kind of conduct that parents and caretakers would normally observe when putting a child to sleep. If the State’s position was that swaddling and placing Jack face down on a pillow was morally and criminally wrong, then one would expect at least some evidence to that effect. But the State produced no testimony, no expert opinions, no evidence of any generally accepted conduct, nothing.2
*544Child Abuse Resulting in Great Bodily Harm Caused by Intentional Suffocation
{50} When trial commenced, consistent with the charges in the indictment, the State theorized that Jack’s injuries resulted from being swaddled in a blanket and placed face down in the crib. During its opening statement, the State said that the evidence would show that Daniel “[sjwaddled [Jack]” and placed him “face down on the weapon that ultimately hurt[] Jack. A pillow.” The State told the jury that the evidence would show that:
[Daniel] described that evening as the most frustrated he has ever been with Jack. This is a young man who is stuck with a crabby baby. And frankly ... he was sick of it. And so he put him on his face in a pillow and left him. And that’s how Jack’s brain gets hurt.
{51} However, as previously described, the State changed its theory of the case during trial after its experts testified that swaddling Jack and placing him face down on a pillow would not likely have caused Jack’s injuries. The doctors who testified for the State instead hypothesized that Jack must have been “non-accidentally” smothered or suffocated.
{52} The State’s pivot from a theory of careless bed-positioning to one of intentional suffocation caught defense counsel by surprise. During a bench conference at trial, defense counsel complained to the judge that “[t]hings have changed a lot. The statement from ... the witness stand ... the neurologist [Dr. Johnson] to say this was an intentional act, changed completely the nature of the action that was charged to the jury.” Defense counsel went on to state:
things have changed. . . . [T]heir entire case when they present it to the jury was one of a negligent act. The theory has changed completely. I think we should get a mistrial based on the fact that they’ve switched horses on us because our discovery didn’t relate to that, and nobody even suggested it. The grand jury was never even [sic] suggested it, Your Honor.
The district judge acknowledged that the State’s theory had changed, referencing the indictment which based the negligent and intentional child abuse charges on the act of swaddling. Inexplicably, at the end of the bench conference, the district judge told the prosecutor, “I’m not throwing out your indictment. I justpointed out that the theory of the State, as [defense counsel] was speaking of, has changed.”
{53} Defense counsel’s complaint about the State’s change of theory occurred after the State had already presented medical testimony from Dr. Johnson about smothering, testimony to which defense counsel did not object. Accordingly, our review on appeal is limited to the sufficiency of that evidence, so as to determine whether the State should have another opportunity to try Daniel for child abuse based on the non-accidental suffocation *545theory, or the charge must be dismissed for lack of sufficient evidence at trial.
{54} Our sufficiency-of-the-evidence review of the conviction for intentional child abuse based on smothering is the analytical opposite of the conviction for negligent “or reckless” child abuse. For “negligent child abuse” Daniel’s actions were not in dispute; he freely admitted to “bundling]” Jack and placing him face down on a pillow. The jury then had to decide whether those actions were culpable and caused Jack’s injuries. With respect to intentional child abuse resulting in great bodily harm, however, we are faced with the opposite situation. The act of smothering Jack — if proven — would be sufficient evidence of intentional child abuse. The unanswered question is whether Daniel actually committed such an act, an act to which he never admitted and which no observer testified that he did. We next examine whether the State presented sufficient evidence to prove that Daniel smothered Jack.
{55} Child abuse prosecutions are unusual in that sometimes medical-opinion testimony provides the only evidence that a wrongful act occurred or that the accused committed a wrongful act. See Deborah Tuerkheimer, The Next Innocence Project: Shaken Baby Syndrome and the Criminal Courts, 87 Wash. U. L. Rev. 1, 5 (2009) (“With rare exception, the [shaken baby syndrome] case turns on the testimony of medical experts. Unlike any other category of prosecution, all elements of the crime — mens rea and actus reus (which includes both the act itself and causation ofthe resulting harm) — are proven by the science.”).
{56} In this case, expert medical testimony provided the only evidence that Jack may have been smothered — that a crime had occurred — and that Jack had not been injured by other, noncriminal causes. Expert witnesses function to assist the trier of fact in understanding evidence and the factual issues in a case. See Rule 11-702 NMRA (“A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue.”). Trial courts have broad discretion in admitting expert testimony as long as the expert (1) is qualified, (2) provides testimony that will assist the trier of fact, and (3) provides testimony regarding “scientific, technical, or other specialized knowledge with a reliable basis.” State v. Downey, 2008-NMSC-061, ¶¶ 25, 26, 145 N.M. 232, 195 P.3d 1244.
{57} Although the admissibility of medical testimony in this case is not under review — defense counsel did not object to either the qualifications of the medical experts or the vast majority of their testimony — orrr inquiry must go deeper than mere admissibility. Analytically, there is a vital difference between expert testimony that may be admissible to assist the trier of fact and expert testimony that, unassisted by any other evidence, sufficiently establishes guilt beyond a reasonable doubt. Our review here focuses on whether this expert testimony alone — testimony in this case based solely on a deduction from an absence of other causes that a certain event likely occurred — is sufficient to support a criminal conviction. We now review the testimony of expert witnesses, seen through the prism of proof beyond a reasonable doubt, to assess its sufficiency.
*546The Medical Process of Elimination Leading to Suffocation as a ‘Likely’ Cause Failed to Prove Guilt Beyond a Reasonable Doubt
{58} In this case, the doctors used a process of differential diagnosis to determine what was likely causing Jack’s condition. The doctors then used, as we shall explain, what we call differential etiology to determine the likely non-medical, external cause of Jack’s hypoxemic ischemic injury, in this case non-accidental suffocation. We examine that process in more detail to understand what it purports to be, as well as its limitations in the context of a criminal trial which demands evidence sufficient to prove guilt beyond a reasonable doubt.
{59} In medicine, differential diagnosis is “the determination of which of two or more diseases with similar symptoms is the one from which the patient is suffering, by a systematic comparison and contrasting of the clinical findings.” John B. Wong, et al., Federal Judicial Center, Reference Manual on Scientific Evidence, Reference Guide on Medical Testimony, 687, 691 n.9 (3d ed. 2011) (quoting Steadman's Medical Dictionary 531 (28th ed. 2006)). Differential diagnosis is a process widely used by doctors in modern medicine. Edward J. Imwinkelried, The Admissibility and Legal Sufficiency of Testimony about Differential Diagnosis (Etiology): Of Under — And Over — Estimations, 56 Baylor L. Rev. 391, 392 (2004). One court has described the process this way:
[Differential diagnosis involves a process by which a medical expert first “rules in” various potential causes and then “rules out” those causes one by one (to the extent possible) by analyzing the patient’s condition until the expert can identify the likely cause from among those remaining. Thus, “differential diagnosis” is a general description of a methodology or process accepted by the medical community as a means of determining causation (or condition), rather than a specific scientific technique or test. . . .
Marcum v. Adventist Health System/W., 193 P.3d 1, 6 (Or. 2008) (en banc.).
{60} To understand the process whereby doctors attempt to determine the external, non-medical cause of the injury, we explore a legal construct called differential etiology. Differential etiology is “a process that identifies a list of external agents . . . that potentially caused the disease.” Ian S. Spechler, Physicians at the Gates ofDaubert: A Look at the Admissibility of Differential Diagnosis Testimony to Show External Causation in Toxic Tort Litigation, 26 Rev. Litig. 739, 740 (2007). It is a “[t]erm used by the court or witnesses to establish or refute external causation for a plaintiffs condition. For physicians, etiology refers to cause.” Wong, supra at 743. Significantly, we note that“[i]n medicine, etiology refers to the study of causation in disease, but differential etiology is a legal invention not used by physicians.”Id. at 691 (footnote omitted). The difference between differential diagnosis and differential etiology is that “[differential diagnosis refers to the clinical process by which doctors determine the internal disease that is causing a patient’s suffering; differential etiology is used for determining the external causes of the problems.” Spechler, supra at 743.
{61} When a physician conducts a *547differential analysis, “he or she brainstorms and attempts to think of all the possible or eligible causes of the plaintiff’s illness. It may be impossible for the physician to explore every conceivable cause, but the physician must undertake a serious investigation of at least the obvious plausible alternative causes.” Imwinkelried, supra at 403-404 (footnotes omitted). Importantly, “physicians receive more formal training in differential diagnosis than in differential etiology. . . . [Practicing physicians have more experience working with the differential diagnosis technique, since in many cases the cause of an illness is irrelevant to thepatient’s treatment.”Id. at405 (footnote omitted). Our Court of Appeals has further acknowledged that “in many cases, including toxic tort cases ... the determination of the external cause of a patient’s disease is a complex process that is unrelated to diagnosis and treatment, and which requires specialized scientific knowledge regarding the external agents involved.” Parkhill v. Alderman-Cave Milling & Grain Co. of N.M., 2010-NMCA-110, ¶ 23, 149 N.M. 140, 245 P.3d 585.
{62} In this case, the doctors used differential diagnosis to first rule in various causes of Jack’s condition, and then, by ruling out other explanations, concluded that oxygen deprivation was the likely internal cause of Jack’s brain injury. The doctors testified that they were initially concerned that Jack was suffering from an infection, septic shock, or inborn error of metabolism. Essentially, the doctors believed that these were competing hypotheses that could explain Jack’s condition. Oxygen deprivation had not at this point entered in to the differential diagnosis. After tests came back negative for the doctors’ possible causes and Jack suffered a seizure, the doctors worried that Jack had a traumatic brain injury and called Dr. Johnson for a neurological examination. The brain scans confirmed that Jack had a brain injury, and it was at this point that the doctors looked to oxygen deprivation as the likely cause of that brain injury. Eventually, the doctors also considered the external, non-medical causes of the oxygen deprivation and settled upon suffocation as a likely cause. Dr. Johnson was the leader of this process, and she settled quickly upon suffocation.
{63} As Dr. Joseph explained, Dr. Johnson was immediately “concerned about suffocation.” Dr. Joseph also testified that “initially, [Dr. Johnson] didn’t say necessarily what the mechanism might have been, but said that she has seen several cases of suffocation in her 20-some-odd years of practice, and that this could most definitely be a case of suffocation.” (Emphasis added.) Dr. Joseph testified that when she first saw Jack she was not ready to commit fully to suffocation as the explanation for his neurological injuries, but that “definitely the possibility of child abuse [by suffocation] entered my differential diagnosis.” (Emphasis added.) Once Dr. Johnson suggested suffocation as a possible cause of Jack’s neurological injuries, however, Dr. Joseph concurred. “[Ajlthough I had never seen a case of suffocation in my practice, I could... understand that that diagnosis or that event could explain everything that I was seeing . .. with Jack. ... It was all explained, especially in the absence of infection and inborn error of metabolism.” (Emphasis added.)
{64} Dr. Joseph testified about other possible causes of suffocation, including carbon monoxide poisoning and drowning, but it was clear to the doctors that Jack was a victim of neither. Dr. Johnson also testified:
[A]t that time, having reviewed the history and looked at the CT scan, *548my major concern was that this was a child who had been submitted to abuse, and it was non accidental. . . [bjecause there was no explanation for what was fairly severe manifestations of his — in his clinical history. And there wasn’t a car accident, or a fall, or anything that would explain that. And that’s when we become very concerned that — because we see a fair amount of child abuse in child neurology, that that’s just about always on our list of possibilities, whatever the presentation.
(Emphasis added.) Dr. Johnson elaborated:
State: Based on the findings that you saw and your thought that perhaps this was non accidental abuse, did you have a method that happened by?
Dr. Johnson: Yes. I felt that at that point in time, with the CT scan evidence, that this was most likely a lack of oxygen. ... I felt that the most likely mechanism, if we talk about mechanisms of brain injury in small children in child abuse, the most likely mechanism was that of suffocation.
(Emphasis added.) After prompting by the district court, the State asked Dr. Johnson what physical manifestation from her examination led her to believe that Jack had been smothered. Dr. Johnson responded that Jack’s seizures were an indicator, as was the lack of any other physical manifestations. “There is often nothing that can be seen, nothing on the face, no bruises, no bleeding. No petechiae. Greater than 50 percent of the time, there is absolutely nothing on the skin or on the baby that would indicate the problem.”3
{65} Dr. Johnson stated at the conclusion of her direct examination, “I believe that this was a prolonged enough event where [Jack] was smothered.” Dr. Coleman agreed with Dr. Joseph and Dr. Johnson. She testified that for Jack’s type of injuries to occur, “[i]t would have had to have been a significant suffocation, and not where he could get out of it. I’m not saying it was one episode, it could have been several repeated episodes over a short period of time.” The State followed up:
State: If you put him down and then hold him down maybe for a little while?
Dr. Coleman: To put him down and hold him down, yes. If you would have held him face down in something, something over his face, or your hand over his face, or some other object over his face.
Overall, Dr. Coleman testified that “I also believe he was suffocated” as a result “[o]f an intentional suffocation. So it’s child abuse.”
{66} As previously acknowledged, this evidence was as admitted without objection. We pause for a moment, however, to consider the vital difference between the mere admissibility of expert testimony and the subsequent review courts must conduct as to whether that testimony, plus any other evidence, is sufficient to support a criminal verdict beyond a reasonable doubt.
*549{67} Admissibility is a minimal standard for individual items of evidence, including expert opinions. Generally, to be admissible an item of evidence need only add something to the debate; it need only have a “tendency to make a fact more or less probable than it would be without the evidence” Rule 11-401 NMRA. Generally, no single piece of evidence has to carry the entire burden of proof especially in a criminal trial. An expert witness may give an opinion if it “will help the trier of fact to understand the evidence.” Rule 11-702.
{68} An expert may testify that an observation or an opinion is consistent with an underlying fact, even though that same observation may be consistent with other facts as well. Bloodshot and watery eyes may be consistent with intoxication, but that same observation is consistent with other, noncriminal explanations as well. It is doubtful whether that observation, considered alone withoirt any additional evidence, would be sufficient to convict someone beyond a reasonable doubt of the crime of driving while intoxicated.
{69} Unlike the present case, doctors usually testify as to what caused a patient’s condition using phrases like “to a reasonable medical probability” or “to a reasonable medical certainty,” phrases that demonstrate a sufficient degree of conviction to be probative. Wong, supra at 693. These phrases “are also terms of art in the law that have no analog for a practicing physician.” Id. at 691. Essentially, these phrases satisfy a minimal standard of probability, and therefore admissibility, that an opinion is more likely than not true. See Renfro v. San Juan Hosp., 1965-NMSC-067, ¶ 9, 75 N.M. 235, 403 P.2d 681 (observing that the two phrases both refer to a probability standard); Black’s Law Dictionary 1380 (9th ed. 2009) (defining “reasonable medical probability” as “a showing that the injury was more likely than not caused by a particular stimulus, based on the general consensus of recognized medical thought. — Also termed reasonable medical certainty.”) (emphasis in original). We recognize that some of the medical opinions offered in this case spoke only in terms of “possibilities” or “concerns,” which essentially proves nothing. Others spoke in terms of “probabilities” or “likelihoods,” sufficient to be admitted. Id. at 692. (“Although most courts have interpreted ‘reasonable medical certainty’ to mean a preponderance of the evidence, physicians often work with multiple hypotheses while diagnosing and treating a patient without any ‘standard of proof to satisfy.” (footnote omitted)). No opinions were offered demonstrating their validity to more than either an unquantified possibility or at most, a bare probability.
{70} In a criminal trial, however, unlike a medical differential diagnosis, the jury must determine beyond a reasonable doubt that a defendant is guilty of the crime charged. The jury must have a sufficient evidentiary basis to conclude that the defendant actually committed the criminal act he is accused of, not just that he may have done it among a range of possibilities or that it cannot be “ruled out” among other possible explanations, or even that it is more likely than not. At Daniel’s trial, the jury was instructed that to find Daniel guilty of child abuse resulting in death or great bodily harm, the State had to prove beyond a reasonable doubt that Daniel caused Jack “to be placed in a situation which endangered the life or health of Jack Consaul” or that Daniel “tortured or cruelly punished Jack Consaul.” The jury also had to determine whether Daniel acted “intentionally.”
*550{71} Essentially, the doctors in this case testified in various ways, and with various degrees of conviction, that they suspected child abuse, that they could not rule out child abuse, that they could not think of other explanations for Jack’s injuries, or that child abuse was a likely cause. Thus, the jury had before it some opinions about “possibilities,” other opinions about “probabilities” or “likelihoods.” The best these opinions could offer was that, to a preponderance of the evidence, Jack was likely suffocated. In most cases, additional non-opinion evidence, such as forensic evidence, supplements expert opinion so that a jury may draw supporting inferences and reason from the totality of the evidence to find proof beyond a reasonable doubt. Here, however, the jury had no such help.
{72} No doctor offered any kind of probability analysis or said anything more than they thought it was child abuse. While proof beyond a reasonable doubt is not required for admissibility of an opinion, it is essential to support a jury’s finding of guilt. As a reviewing court, we must decide whether a reasonable jury could “reason” from the available evidence to the point of finding guilt beyond a reasonable doubt. In this case, there was no substantial evidence pointing to Daniel’s guilt other than whatever could be said of the medical testimony, and when that evidence is analyzed, it falls short of establishing proof beyond a reasonable doubt.
{73} Recognizing the challenge this case presents, we offer the following guidance for future courts and litigants. Medical testimony of this type, as a matter of evidentiary foundation, should describe in detail the methodology utilized first to “rule-in” possible causes and then to “rule-out” all but one. Based on that process of elimination, described in detail to the jury, a doctor then should be able to offer an opinion on causation to a reasonable degree of medical probability which satisfies a minimal standard for admissibility. In a criminal trial, to meet a standard of proof beyond a reasonable doubt, prosecutors can usually point to additional, non-opinion evidence, so that when considered cumulatively all the evidence is sufficient to support a verdict beyond a reasonable doubt. If, however, the prosecution is to rely solely on medical opinion, it must go beyond the mere probable causation required for admissibility, which alone is really nothing more than establishment of a preponderance. The medical testimony should establish for the factfinder, the trial court, and reviewing courts why the expert opinions are sufficient in themselves to establish guilt beyond a reasonable doubt. There must be assurances that the jury has before it sufficient evidence, considered as a whole and without passion or speculation, to enable it reasonably to find guilt beyond a reasonable doubt. Those measures were absent from this case.4
*551Daniel’s Alleged Change of Story and His Frustration with Jack
{74} Perhaps sensing the peril in relying solely - upon a medical theory of “likely” suffocation, the State offered two additional theories that the doctors developed. First, as evidence that a crime had occurred, Dr. Johnson specifically testified that she believed Daniel had changed his story about the night Jack was injured, which she characterized as “classic in the child abuse cases, and I’ve seen all too many of them.” Second, as evidence of criminal intent, the State noted Daniel’s acknowledgment that he was frustrated with Jack that night. During a bench conference regarding defense counsel’s surprise at the prosecution’s change of theory mid-trial, the prosecutor responded that “[t]he frustration and irritation that the defendant describes is, in fact, the basis for [the] intentional act.” We next examine the evidentiary proof of these allegations.
{75} In her trial testimony, Dr. Johnson characterized Daniel’s account of the night of Jack’s injury as a “change in story” which, in her opinion, corroborated her theory of intentional suffocation. Dr. Johnson testified that based on her review of the record, “the thing that stood out was this change in story.” Dr. Johnson’s conclusion did not come from anything Daniel directly told her. Dr. Johnson did not interview either Daniel or Heidi, who is Jack’s mother and Daniel’s sister. Dr. Coleman talked with Heidi but not with Daniel.
{76} Dr. Coleman prepared a Child Abuse Response Team (CART) report regarding Jack. According to Dr. Coleman, a CART report documents a patient’s medical records, studies, x-rays and other images, laboratory reports, and family history to determine whether the child may be the victim of abuse. Dr. Coleman’s report in this case described her impressions from Jack’s medical records, laboratory reports, and the history she obtained from Heidi during Jack’s admission. Dr. Johnson used the information contained in Dr. Coleman’s CART report and the information the doctors later learned about Daniel swaddling Jack and placing him face down to conclude that Daniel had changed his story. We examine the relevant portions of those documents in some detail.
{77} Dr. Coleman’s CART report contains two relevant statements from Heidi. During her discussion with Heidi, Dr. Coleman recorded Heidi’s statement that “Jack is always put to sleep on his back but that he likes to pull the blanket up on to his face.” Regarding the night in question, Dr. Coleman noted that “Heidi said that Daniel found Jack *552on his back in the crib [and] had vomited,” based apparently on what Daniel had told Heidi. Dr. Coleman also noted that Dr. Johnson, even at that early point in time and before there had been an alleged change of story, “felt that child abuse by suffocation could not be ruled out by history and findings thus far.” Dr. Coleman included her impression that, even at that early point, “[i]n absence of a medical cause being identified, suffocation is the most likely mechanism for this injury.” The CART report also recites that Heidi had lost another child several years earlier due to suffocation. At that time Heidi was charged with her child’s death and jailed, although those charges were ultimately dismissed. Nothing in the CART report suggests that Daniel was involved in the earlier matter.
{78} Suspecting that Jack was a victim of child abuse, the Child Abuse Response Team notified CYFD, which then notified the Las Cruces Police Department. Detective Myers testified that Dr. Coleman was the only doctor at UNMH with whom he spoke. Dr. Coleman initially informed Detective Myers that the doctors at UNMH had not found any medical cause for Jack’s injuries. When asked if Dr. Coleman told him that there might be other possible causes for Jack’s injuries, Detective Myers stated, “[s]he told me there was no other possible cause.” Based on what UNMH staff and CYFD social workers had said about child abuse, Detective Myers initiated his investigation, which included two interviews with Daniel. Detective Myers ultimately reported what he had learned to Dr. Coleman, and at trial he testified about his two interviews with Daniel. The district court admitted the transcripts of both of Daniel’s interviews into evidence.
{79} During his first interview with Detective Myers, Daniel described wrapping Jack in a blanket. Daniel said he placed the blanket on Heidi’s bed and “folded it over so that [Jack would] have his little head rest and I set him in there, folded it, I folded it up then over and then I wrapped it around and tucked it in and I picked him up and sat him in his bed.” Detective Myers did not ask any followup questions regarding the manner in which Daniel put Jack to bed, or whether it deviated in any way from Jack’s normal bedtime routine. Finally, Detective Myers asked Daniel if anything “happened that night that . . . would explain [Jack’s] condition,” to which Daniel responded “[n]o.”
{80} During his second interview with Daniel, Detective Myers asked more probing questions about the manner in which Daniel placed Jack in the crib to sleep that night. When Detective Myers asked Daniel whether he was frustrated by Jack’s crying and fussiness that particular night, Daniel responded that he was. Later in the interview, Detective Myers asked, “So, that was like the . . . most frustrated you [have] ever been with [Jack]?” and Daniel agreed. Detective Myers also asked Daniel to describe exactly how he bundled Jack in the blanket and placed Jack in the crib. Daniel responded that he bundled Jack more tightly than normal and placed him face down in the crib, because he wanted Jack to stop crying so that he could complete the tasks Heidi had given him that evening.
{81} During cross-examination at trial, Detective Myers acknowledged that it was not until the second interview thathe asked Daniel to describe exactly how he had put Jack to bed that night. Defense counsel asked Detective Myers, “[s]o [Daniel’s] first statement and his second statement aren’t really different statements, they are just more information in the second one because you asked more *553pointed questions?” Detective Myers responded that “there was different information provided to me in the second statement.” Detective Myers acknowledged, however, that defense counsel’s characterization of the evidence was “[c]orrect. Nothing besides the frustration, the way he placed the child in the crib was different.”
{82} In reaching her opinion that Daniel changed his story, Dr. Johnson stated that she relied on the initial history from Heidi as noted by Dr. Coleman in the CART report, “that Daniel found Jack on his back in the crib [and Jack] had vomited.” After prefacing her testimony that a change in story is “very typical of child abuse,” Dr. Johnson stated, “it was only . . . three days later. . . that the first indication of the baby being wrapped or swaddled and placed on his tummy on a pillow was heard . . . after ... the CART report was made.” It appears that Dr. Johnson placed great importance on what she saw as a change in Daniel’s account of how he put Jack to sleep that night — first, on his back and later, on his tummy.5
{83} . The assertion of a “change of story” was based on what Heidi recalled Daniel telling her, which she then repeated to Dr. Coleman, which was then transmitted via the CART report as tertiary hearsay to Dr. Johnson. It is difficult to know how to interpret this so-called contradiction or assess its reliability. Even if Heidi accurately recounted what Daniel had told her, it was that “Daniel found Jack on his back [and] had vomited,” not that he had put Jack to bed in that position. (Emphasis added.)
{84} Dr. Johnson nonetheless characterizes this asserted change in story as a “big change,” which she believed proved her initial theory that Jack had been smothered based on the medical and family history that Heidi provided to Dr. Coleman. During her testimony, Dr. Johnson emphasized that:
The real thing to pay attention to here is that the story came out of the swaddling, the placing of the baby on the tummy on a pillow . . . was very calculated . . . once [Daniel] found out what we’re thinking [is the cause of Jack’s condition] .. . certainly by the 11th [the day Jack was admitted to the hospital], . . . virtually top on the list was non accidental trauma and specifically smothering.
Implicit in Dr. Johnson’s testimony is her speculation that Daniel invented a story in a calculated manner to deflect suspicion away from himself when he knew that a child abuse investigation was likely to occur.
{85} Whatever weight the jury accorded the State’s theory that Daniel changed his story, we will never know. What we do know is that Dr. Johnson’s testimony that Daniel’s alleged change of story was an important element of her calculus, which resulted in her conclusion that Daniel had intentionally suffocated Jack. However, the do'ctors determined that child abuse by suffocation was the likely cause of Jack’s injuries before Daniel had the opportunity to change his story.
*554{86} We are at a loss to understand Dr. Johnson’s conclusion, much less the weight she apparently gave it. Dr. Johnson may have believed that Daniel changed his story between the occurrence of the police interviews and the preparation of the CART report with respect to how he had put Jack to bed that night. However, Daniel was never interviewed for the CART report. The only information in the CART report regarding Daniel’s explanation of what had happened resulted from Dr. Coleman’s interview with Heidi, not Daniel. Heidi described Jack’s usual bed positioning, something that Daniel did not contradict. Heidi described how Daniel had told her that after putting Jack in the crib, he later found Jack on his back, vomiting. Dr. Johnson then read the CART report prepared by Dr. Coleman. The CART contained hearsay stating what Daniel told Heidi, which Heidi then told Dr. Coleman. These levels of hearsay do not provide sufficient evidence to support a conviction of child abuse.
{87} If Dr. Johnson believed Daniel was being evasive, and if that alleged evasion was significant to her opinion, she was free to so testify. Here, however, we are tasked with the legal responsibility of assessing that opinion in light of its probative value, whether it can be described as sufficient evidence of Daniel’s guilt. Dr. Johnson’s view of Daniel’s “story” arguably may go to impeach Daniel’s credibility. However, we are not persuaded that Dr. Johnson’s view ofDaniel’s credibility makes it any more or less probative of her own theory that Daniel intentionally smothered Jack; not, at least, when held up to the searching light of proof beyond a reasonable doubt.
{88} We acknowledge that for doctors, inconsistent accounts from caretakers often indicate abuse. See, e.g., John E.B. Myers, Myers on Evidence of Interpersonal Violence: Child Maltreatment, Intimate Partner Violence, Rape, Stalking and Elder Abuse § 4.10(A)-(E) (5th ed. 2011) (“An abusive caretaker who invents an explanation for a child’s injuries may have difficulty keeping the story straight. When the explanation differs each time the caretaker is interviewed, suspicion escalates.”); Thomas D. Lyon, et al., Medical Evidence of Physical Abuse in Infants and Young Children, 28 Pac. L. J. 93, 96-97 (1996-97) (“Physicians often diagnose abuse on the grounds that the caretakers gave a history that was inconsistent with the injury, vague, or unclear; changed over time; or varied depending upon which family member was asked.’’(emphasis added)). For the inconsistency to amount to sufficient evidence in a criminal case, however, courts have demanded far more than Dr. Johnson elicited here.
{89} Other New Mexico cases discuss situations where caretakers gave different and unreasonable accounts regarding a baby’s injuries. For example, in Cabezuela, the defendant insisted initially that her boyfriend had never hurt the baby. 2011-NMSC-041, ¶¶ 9-10. However, shortly before trial, the defendant reversed course and said that her boyfriend was actually responsible for the baby’s injuries and that she had been covering for her boyfriend. Id. The defendant also gave different accounts of the day and the night that the baby was injured. Id. ¶¶ 11-12. In her earlier statements, the defendant told law enforcement officers that on the day of the incident she was at her storage facility where the baby was lying in her carrier, and the defendant hit the baby on the baby’s head with her open hand and shook the carrier. Id. ¶ 6. In a subsequent interview, the defendant stated that while she was at the storage facility, she *555heard the baby cry from the car where her boyfriend was watching the children, and she noticed a bruise and cut on the baby’s forehead when she returned to the car. Id. ¶ 11. Regarding the night of the baby’s injury, the defendant first stated that she had “pitched” the baby onto the floor and shook her without supporting her head. Id. ¶ 7 (internal quotation marks omitted). The defendant later changed her story, stating that her boyfriend was responsible for the baby’s injuries. Id. ¶ 10; see also Jojola, 2005-NMCA-119, ¶¶ 5, 21 (noting that a defendant’s explanation that child fell from a lower bunk bed was unreasonable in light of child’s significant injuries and testimony at trial).
{90} Any alleged inconsistency inDaniel’s explanation of Jack’s injuries pales by comparison to those cases in which defendants actually change their stories or provide unreasonable explanations for a child’s injuries.
{91} Finally, the State relied on Daniel’s admission that he was frustrated with Jack on the night he was hospitalized — the most frustrated he had ever been with Jack — to establish evidence of intent. During Dr. Coleman’s testimony, the State asked her whether she was told about Daniel’s frustration when Jack was first admitted to the hospital, and if this information would have influenced her diagnosis. Dr. Coleman stated that “[i]t would have been significant” to learn that Daniel had been frustrated with Jack. She further stated that “anybody who has had a baby gets frustrated and upset. How you handle that frustration and upset is a whole different matter.” Dr. Coleman also testified that “[she] was never told that [Daniel] was frustrated or upset, or there was any predisposing event” that could have caused Daniel to “wrap[] [Jack] tightly in a blanket and put him face down in a pillow for an unknown amount of time.”
{92} None of the doctors, including Dr. Coleman, ever spoke with Daniel, meaning that they did not inquire directly with Daniel about whether he was frustrated that night. Frustration can suggest a motive for suffocation, and it was relevant and probative for that purpose. Evidence of motive alone, however, does not establish an intentional act.. There must still be actual evidence — not speculation — of a criminal act, which this case lacks. As Dr. Coleman herself observed, and we agree, every parent or caregiver experiences frustration from time to time, but there are a wide variety of responses to that frustration. Therefore, while the information is relevant, Daniel’s frustration and alleged change of story do not provide proof beyond a reasonable doubt that Daniel suffocated Jack.
CONCLUSION
{93} We reverse Daniel’s child abuse conviction under Section 30-6-1 and order that the charge be dismissed with prejudice.
{94} IT IS SO ORDERED.
RICHARD C. BOSSON, Justice
WE CONCUR:
BARBARA J. VIGIL, Chief Justice
PETRA JIMENEZ MAES, Justice
EDWARD L. CHÁVEZ, Justice
CHARLES W. DANIELS, Justice