OPINION
{1} Defendant was indicted for three counts of criminal sexual contact of a minor (CSCM). Defendant subsequently filed a pretrial motion to dismiss, arguing that the corpus delicti of CSCM could not be proved solely by his extrajudicial statements and that the State would fail to present any independent evidence that would otherwise establish the corpus delicti. The district court denied Defendant’s motion to dismiss, finding that Defendant’s statements were trustworthy and that additional independent evidence helped to establish the corpus delicti of CSCM. We reverse the district court’s denial of Defendant’s motion to dismiss. We also take this opportunity to clarify the law regarding extrajudicial statements and the application of the trustworthiness standard in New Mexico.
FACTS AND PROCEEDINGS BELOW
{2} Although the parties present slightly different chronologies of the events in this case, the material facts are undisputed. In 2001, Defendant was diagnosed with Huntington’s Disease. Huntington’s Disease is a *95fatal hereditary disease that affects motor skills and also causes judgment, memory, and other cognitive functions to deteriorate into dementia. Not long after he was diagnosed, Defendant told his girlfriend (Mother) that God was punishing him with Huntington’s Disease because of what he did to their daughter. At that time, their daughter was two years old.
{3} When pressed by Mother, Defendant admitted to inappropriately touching their daughter and making their daughter touch him, but he did not disclose the exact areas of body contact. Defendant made similar statements to a counselor at the Veteran’s Administration (VA) Hospital. The VA Hospital reported the allegations to the New Mexico Children, Youth and Families Department (CYFD). CYFD investigated the matter and criminal charges were filed against Defendant. During the criminal investigation of Defendant’s statements, Defendant again made statements concerning his daughter to the police officers who interviewed him in his home.
{4} Not long after filing a police report, Mother took the daughter to the S.A.N.E. (Sexual Assault Nurse Examination) Program at St. Vincent’s Hospital where the daughter was examined by a nurse. The examination revealed no physical evidence of sexual abuse. However, Mother did report to the S.A.N.E. nurse that the daughter was exhibiting several changes in her behavior, including nightmares and withdrawal from strangers, particularly male strangers. Of the twelve behaviors that could be corroborative of sexual abuse listed on the S.A.N.E. form, the nurse noted that the daughter experienced only these two behavioral symptoms.
{5} On the basis of Defendant’s statements and the daughter’s behavioral symptoms, Defendant was indicted for three counts of CSCM. Defendant subsequently moved to dismiss the criminal charges against him, arguing that the State failed to prove that the CSCM actually occurred. The State argued that Defendant’s statements were themselves sufficiently trustworthy that, in addition to evidence that the daughter was exhibiting various behavior symptoms indicative of abuse, the corpus delicti of CSCM was established in this case. The district court agreed with the State and denied Defendant’s motion to dismiss.
{6} After the denial of his motion to dismiss, Defendant entered a conditional no contest plea to the three counts of CSCM, expressly reserving the right to appeal the district court’s ruling on the motion. This appeal follows.
STANDARD OF REVIEW
{7} To the extent that Defendant admits the veracity of the facts alleged in the State’s pleadings and argues that such facts cannot establish the corpus delicti of CSCM as a matter of law, we will review the district court’s denial of Defendant’s motion to dismiss de novo. See State v. Roman, 1998-NMCA-132, ¶ 8, 125 N.M. 688, 964 P.2d 852; State v. Rendleman, 2003-NMCA-150, ¶ 27, 134 N.M. 744, 82 P.3d 554; State v. Wasson, 1998-NMCA-087, ¶¶ 5-6, 125 N.M. 656, 964 P.2d 820. However, we caution that not every challenge to the corpus delicti in a particular matter will rest on undisputed facts. Thus, where the determination of the corpus delicti rests on disputed facts, we will defer to the district court’s findings of fact, provided that such findings are supported by substantial evidence. See State v. Jason L., 2000-NMSC-018, ¶ 10, 129 N.M. 119, 2 P.3d 856.
DISCUSSION
{8} On appeal, Defendant argues that the district court erred in concluding that the State presented sufficient evidence to establish the corpus delicti of CSCM. Defendant argues that corpus delicti of CSCM cannot be proved solely by his extrajudicial statements regarding his daughter. Rather, Defendant argues that the State must prove that his daughter was the victim of criminal sexual contact by at least some evidence independent of his extrajudicial statements. We agree that the corpus delicti of CSCM was not established in this case, and we reverse the denial of Defendant’s motion to dismiss.
{9} We begin our opinion with a discussion of the term “corpus delicti” and the corpus *96delicti rule, which has traditionally governed the admissibility of extrajudicial confessions. Second, we address the rejection of the corpus delicti rule by the federal courts and numerous state courts in favor of the trustworthiness standard. Third, we discuss New Mexico’s adoption of the trustworthiness standard and how it is applied by our courts. Finally, we apply New Mexico’s standard to the facts of this ease and conclude that the corpus delicti of CSCM was not established by the State.
A. The Traditional Corpus Delicti Rule
{10} The corpus delicti rule provides that “unless the corpus delicti of the offense charged has been otherwise established, a conviction cannot be sustained solely on [the] extrajudicial confessions or admissions of the accused.” State v. Paris, 76 N.M. 291, 294, 414 P.2d 512, 514 (1966). The term “corpus delicti,” which literally means “body of the crime,” refers to the evidence needed to establish that the charged crime was actually committed. Black’s Law Dictionary 369 (8th ed.2004); see also Doe v. State, 94 N.M. 548, 549, 613 P.2d 418, 419 (1980) (“The corpus delicti of a crime requires proof that the crime was committed.”). Such evidence may or may not connect the defendant with the charged crime and thus should not be confused with the evidence required to prove a particular defendant’s guilt beyond a reasonable doubt. See State v. Chaves, 27 N.M. 504, 506, 202 P. 694, 694 (1921). We observe that
[generally the prosecution bears the burden of proving three elements in a criminal trial: (1) that loss or harm occurred, (2) that someone’s criminal agency caused the loss or harm, and (3) that the “someone” was or included the defendant. It is currently the rule among most jurisdictions requiring proof of the corpus delicti as a condition for admitting the defendant’s confession that the “body of the crime” comprises the first two elements and not the third.
Thomas A. Mullen, Rule Without Reason: Requiring Independent Proof of the Corpus Delicti as a Condition of Admitting an Extrajudicial Confession, 27 U.S.F. L.Rev. 385, 388 (1993) (footnotes omitted); cf. State v. Nance, 77 N.M. 39, 44, 419 P.2d 242, 246 (1966) (“The corpus delicti of a particular offense is established simply by proof that the crime was committed; the identity of the perpetrator is not material.”). Thus, the existence of the corpus delicti is demonstrated by the fact that a harm or injury occurred and that the harm or injury was caused by a criminal act. State v. McKenzie, 47 N.M. 449, 453, 144 P.2d 161, 164 (1943); see also State v. Sosa, 2000-NMSC-036, ¶ 20, 129 N.M. 767, 14 P.3d 32; 7 John H. Wigmore, Evidence in Trials at Common Law § 2072, at 524-25 (Chadbourn rev.1978).
{11} For example, in the present case, the corpus delicti of CSCM is established by evidence proving that (1) the daughter’s intimate parts were touched by someone or that the daughter was forced to touch someone’s intimate parts, and (2) that this harm was caused by someone’s criminal agency. See McKenzie, 47 N.M. at 453, 144 P.2d at 164. Whether or not Defendant was the perpetrator is irrelevant for corpus delicti purposes. See Nance, 77 N.M. at 44, 419 P.2d at 246. The question then becomes what type of evidence can be used to establish the corpus delicti.
{12} According to the corpus delicti rule, the corpus delicti of a charged crime can be proved by direct or circumstantial evidence, State v. Maestas, 92 N.M. 135, 145, 584 P.2d 182, 192 (Ct.App.1978), but such evidence must be independent of a defendant’s own extrajudicial statements. Nance, 77 N.M. at 44-45, 419 P.2d at 246. If the state can present sufficient independent evidence, apart from a defendant’s confession, to establish the corpus delicti, the defendant’s confession may then be used to sustain a conviction. See id. (“When there is, in addition to a confession, proof of the corpus delicti established by independent evidence, the defendant’s voluntary confession will support a conviction.”).
{13} The corpus delicti rule first developed in England during the seventeenth and eighteenth centuries. David A. Moran, In Defense of the Corpus Delicti Rule, 64 Ohio St. L.J. 817 (2003); Mullen, swpra, at 399. The rule has its roots in a notorious English case *97in which an individual was convicted of murder solely on the basis of a confession, only to have the murder victim later appear alive. See Mullen, supra, at 400 (describing Perry’s Case, 14 How. St. Tr. 1311 (1660)). A similar case in Vermont led to the adoption of the rule in the United States. See Moran, supra, at 829-31 (describing The Trial of Stephen and Jesse Boom, 6 Am. St. Tr. 73 (1819)). While the corpus delicti rule only applied to homicide crimes in England, its application was expanded in the United States to cover most crimes. State v. Mauchley, 67 P.3d 477, 481-82 (Utah 2003).
{14} The initial purpose behind the corpus delicti rule was “to prevent the conviction of those who confessed to non-existent crimes as a result of coercion or mental illness.” Moran, supra, at 817. Those who support the rule’s continued application argue that the rule serves an additional purpose as well: “promoting better police work by requiring the prosecution to prove its case without the aid of confessions.” Mullen, supra, at 401. The rule’s opponents point out that modern constitutional and procedural safeguards now serve these same purposes, arguably better than the corpus delicti rule, and assert that the corpus delicti rule is a relic of the common law that should be abandoned. See, e.g., People v. McMahan, 451 Mich. 543, 548 N.W.2d 199, 203-05 (1996) (Boyle, J., dissenting); State v. Parker, 315 N.C. 222, 337 S.E.2d 487, 493-94 (1985); Mauchley, 67 P.3d at 485-88; State v. Ray, 130 Wash.2d 673, 926 P.2d 904, 908-09 (1996) (Talmadge, J., concurring); Mullen, supra, at 401-07. Based on this criticism, the federal courts and an increasing number of state courts have rejected the corpus delicti rule. Mauchley, 67 P.3d at 482-83.
B. The Trustworthiness Doctrine
{15} In recent decades, the corpus delicti rule has been abandoned in the federal courts and in a number of state courts. Moran, supra, at 818. In Opper v. United States, 348 U.S. 84, 93, 75 S.Ct. 158, 99 L.Ed. 101 (1954), and Smith v. United States, 348 U.S. 147, 156, 75 S.Ct. 194, 99 L.Ed. 192 (1954), the United States Supreme Court rejected the corpus delicti rule in favor of the “trustworthiness” doctrine. Mauchley, 67 P.3d at 488; Ray, 926 P.2d at 909 (Talmadge, J., concurring). Under the trustworthiness doctrine, “corroborative evidence need not be sufficient, independent of the [defendant’s] statements, to establish the corpus delicti.” Opper, 348 U.S. at 93, 75 S.Ct. 158. Instead, “[i]t is sufficient if the corroboration supports the essential facts admitted sufficiently to justify a jury inference of their truth.” Id. Further, “ ‘[t]here is no necessity that [the] proof [independent of the defendant’s confession] touch the corpus delicti at all.’ ” Parker, 337 S.E.2d at 492 (alteration in original) (quoting Opper, 348 U.S. at 92, 75 S.Ct. 158). Rather, “proof of any corroborating circumstances is adequate which goes to fortify the truth of the confession or tends to prove facts embraced in the confession.” Opper, 348 U.S. at 92, 75 S.Ct. 158; see also United States v. Johnson, 589 F.2d 716, 718-19 (D.C.Cir.1978) (stating that under the trustworthiness doctrine, “the adequacy of corroborating proof is measured not by its tendency to establish the corpus delicti but by the extent to which it supports the trustworthiness of the [defendant’s] admissions”). A number of state courts have rejected the corpus delicti rule in favor of the federal court’s trustworthiness doctrine. Moran, supra, at 832 n. 103 (citing cases).
{16} Adherents to the trustworthiness doctrine believe that the doctrine serves many of the same purposes of the corpus delicti rule, only better. Parker, 337 S.E.2d at 493-94. Thus, the trustworthiness doctrine still works to ensure that individuals are not convicted of crimes that did not in fact occur. Id.; see also Mauchley, 67 P.3d at 488. Additionally, the trustworthiness doctrine’s proponents argue that by focusing on whether the defendant’s confession is trustworthy, the doctrine “provides greater assurance against the use of an unreliable confession to prove the defendant’s guilt than does the corpus delicti version.” Parker, 337 S.E.2d at 494; see also McMahan, 548 N.W.2d at 206-07 (Boyle, J., dissenting). This is “because the latter approach is directed only to preventing convictions for a crime which has not occurred,” whereas the trustworthiness doctrine ensures that the defendant confessing to the crime is actually *98the guilty party. Parker, 337 S.E.2d at 494; see also McMahan, 548 N.W.2d at 206-07 (Boyle, J., dissenting). However, critics of the trustworthiness doctrine argue that the doctrine “is so malleable that almost any independent evidence of anything can serve to ‘corroborate’ the confession or make it ‘trustworthy.’ ” Moran, supra, at 852. Perhaps mindful of such criticism, numerous states have not adopted the trustworthiness doctrine in its entirety, but rather have simply “modified their treatment of the corpus delicti rule in light of the federal trustworthiness doctrine.” Ray, 926 P.2d at 910 (Talmadge, J., concurring); see also Moran, supra, at 833-35 & n. 107.
C. New Mexico’s Adoption of the Trustworthiness Doctrine
{17} Ten years after the decisions in Op-per and Smith were announced, the New Mexico Supreme Court announced that it had adopted the trustworthiness doctrine and rejected the corpus delicti rule. See Paris, 76 N.M. at 295, 414 P.2d at 515. However, the court in Paris also appeared to adopt the reasoning of State v. Lucas, 30 N.J. 37, 152 A.2d 50 (1959), which stated that “[n]o greater burden should be required of the State than independent corroborative proof tending to establish that when the defendant confessed he was telling the truth, plus independent proof of the loss or injury.” Paris, 76 N.M. at 296, 414 P.2d at 515 (emphasis added) (quoting Lucas, 152 A.2d at 61). Contrary to Lucas, the trustworthiness standard, as described in Opper, does not require independent proof of the corpus delicti when a confession is shown to be trustworthy. Opper, 348 U.S. at 92-93, 75 S.Ct. 158. Further, the evidence establishing the trustworthiness of a confession need not touch on the corpus delicti either. Id. The Lucas rule, however, requires the state to introduce evidence establishing the trustworthiness of the defendant’s confession, plus independent evidence that establishes the first prong of the corpus delicti, which is proof of loss or injury. Lucas, 152 A.2d at 61; see also State v. Reddish, 181 N.J. 553, 859 A.2d 1173, 1211-12 (2004) (explaining the Lucas rule). The purpose behind such a formulation of the trustworthiness standard is “[t]o avoid the danger of convicting a defendant solely out of his own mouth of a crime that never occurred or a crime committed by someone else.” Reddish, 859 A.2d at 1211 (internal quotation marks and citation omitted); see also United States v. Lee, 315 F.Supp.2d 1038, 1043 (D.Ariz.2003) (discussing a similar approach).
{18} We therefore conclude that Paris did not adopt the trustworthiness standard in its entirety, because the rule still seems to require that at least some evidence independent of a defendant’s confession actually touch upon the corpus delicti. See State v. Buchanan, 76 N.M. 141, 143, 412 P.2d 565, 566-67 (1966) (holding one week after the Paris decision that an extrajudicial confession, “when established as trustworthy, may properly be considered together with independent or corroborative evidence as proof that the crime charged was committed”). The rule announced in Paris thus states that an extrajudicial statement may be used to establish the corpus delicti where the statement is shown to be trustworthy and where there is some independent evidence to confirm the existence of the alleged loss or injury. See Paris, 76 N.M. at 296, 414 P.2d at 515.
{19} Subsequent decisions by our courts raise questions as to whether the Paris rule is still applicable in New Mexico. For example, just five months after the decision in Paris, the Court in Nance cited to pre-Paris cases to conclude that “[wjhen there is, in addition to a confession, proof of the corpus delicti established by independent evidence, the defendant’s voluntary confession will support a conviction.” Nance, 77 N.M. at 44-45, 419 P.2d at 246. This statement suggests a return to the corpus delicti rule. See, e.g., Chaves, 27 N.M. at 510, 202 P. at 696 (“The law is that the corpus delicti must be shown independently, but that, once shown, the other elements necessary to a conviction might be proved by the confession alone.”). Notably, the court in Nance did not cite to Paris at all. Similarly, we observe that our court seemed to follow the corpus delicti rule in State v. Gruender, 83 N.M. 327, 330, 491 P.2d 1082, 1085 (Ct.App.1971), but more recently we have appeared to follow the Paris rule. *99 See State v. Sanchez, 109 N.M. 718, 719, 790 P.2d 515, 516 (Ct.App.1990).
{20} In Doe, 94 N.M. at 548, 613 P.2d at 418, a majority of our Supreme Court purportedly applied the Pans rule to a case involving a juvenile’s confession to shoplifting alcohol from an Albertson’s grocery store. In Doe, police officers responded to a report of juveniles drinking alcohol in the parking lot of an Albertson’s store. Id. An officer questioned the four juveniles in the parking lot and they told the officer that a fifth juvenile, Doe, stole the alcohol. Id. at 548-49, 613 P.2d at 418-19. Doe subsequently approached the officers and confessed to stealing the alcohol from the Albertson’s. Id. at 549, 613 P.2d at 419. At trial, the State attempted to corroborate Doe’s confession with evidence that the alcohol had Albertson’s price tags on it, that Doe did not purchase any alcohol from Albertson’s on that day, that the alcohol was being consumed in the proximity of Albertson’s, and that Doe was in the area where the alcohol was found. Id. The majority concluded that the State failed to adequately corroborate Doe’s confession in order to establish the corpus delicti of shoplifting. Id. Specifically, the Court observed that there was no evidence, absent the confession, that actually demonstrated that Doe ever possessed or concealed the alcohol, nor was there any evidence, absent Doe’s confession, that the alcohol was actually shoplifted. Id. Quite simply, the circumstantial evidence was just too tenuous to establish the reliability of Doe’s confession and/or provide adequate proof of the harm or loss. See id.
{21} Justice Easley, with Justice Payne concurring in dissent, argued that the majority incorrectly applied the Paris rule and that there was substantial evidence to corroborate Doe’s confession. Doe, 94 N.M. at 549-50, 613 P.2d at 419-20 (Easley, J., dissenting). The dissent argued that the fact that Doe confessed twice to shoplifting was significant and those two confessions, in addition to the circumstantial evidence described by the majority, were clearly enough independent evidence to substantiate Doe’s confession. Id. at 551, 613 P.2d at 421. The dissent further argued that the majority’s holding was contrary to the rule described in Paris and Opper, and that if the majority refused to consider Doe’s two confessions, Paris should be overruled. Id.
{22} While the dissent in Doe certainly raises questions as to what rule is applied in New Mexico, we are not convinced that the majority decision was a radical departure from the Paris rule. Rather, we interpret the disagreement in Doe not as a dispute over the proper rule, but as a dispute over the quantum of independent evidence needed to corroborate a confession and/or independently establish the alleged loss or injury in a given case. See Brian C. Reeve, Note, Survey of Developments in North Carolina Law, 1985: State v. Parker: North Carolina Adopts the Trustworthiness Doctrine, 64 N.C. L.Rev. 1285, 1299 n. 131 (1986) (interpreting the difference between the majority and dissenting opinions in Doe as a disagreement over how much evidence is needed to establish the trustworthiness of a confession). We therefore conclude that the Paris rule survived Doe. See Sanchez, 109 N.M. at 719, 790 P.2d at 516 (citing Paris as the general rule in New Mexico).
{23} We observe that while the Supreme Court’s most recent decision in Sosa once again appears to signal a return to the corpus delicti rule, we do not interpret that decision to be a repudiation of the Paris rule. In Sosa, a man was shot in the face and died as a result of his wounds. 2000-NMSC-036, ¶ 2, 129 N.M. 767, 14 P.3d 32. The defendant was arrested and subsequently argued that the state failed to establish the corpus delicti of the murder because the state did not present any evidence that connected him with the murder, aside from his own confession. Id. ¶ 16. The Court correctly concluded that the corpus delicti of homicide was established in this case, as the corpus delicti does not require proof that the defendant was the perpetrator. Id. ¶ 20. However, we observe that in coming to this conclusion, the Court cited to Nance rather than Paris as the correct rule, thereby giving the impression that the corpus delicti of a crime must be proven by independent evidence before a confession may be used to sustain a convic*100tion. Id. ¶ 16, 414 P.2d 512 (citing Nance, 11 N.M. at 44-45, 419 P.2d at 246).
{24} Although the Court in Sosa did not cite to Paris in its discussion of the corpus delicti of a homicide, we observe that the corpus delicti was clearly established in Sosa by independent evidence such that the defendant’s extrajudicial confession was unnecessary for purposes of establishing the corpus dehcti. Id. ¶¶ 15-20. Further, we note that the Court did not state that the corpus dehcti of a crime must be established solely by evidence independent of the corpus dehcti, but rather, simply that where there is no independent evidence of the corpus dehcti, a defendant’s extrajudicial statement cannot be used to estabhsh the corpus dehcti. Id. ¶ 16. This is consistent with the rule in Paris, which still requires some independent evidence of a loss or injury before a defendant’s statement can be used to help estabhsh the corpus dehcti. Paris, 16 N.M. at 296, 414 P.2d at 515. The Paris rule is therefore still applicable in New Mexico.
{25} Finally, we observe that neither Defendant nor the State have argued that the rule in Paris is no longer applicable. Additionally, the district court clearly applied the Paris rule, albeit incorrectly, in denying Defendant’s motion to dismiss. We are therefore satisfied, based on our review of New Mexico ease law and the arguments of the parties and court, that the Paris rule is the proper rule to use with respect to Defendant’s extrajudicial statements and the corpus dehcti of CSCM in this case. Having determined the proper rule to apply, we now must address whether the district court erred in concluding that Defendant’s statements were sufficiently corroborated by independent evidence and that the independent evidence produced by the State established that the loss or harm aheged in this case actually occurred.
D. Application of New Mexico’s Rule to the Present Case
{26} Our statutes define CSCM as “the unlawful and intentional touching of or applying force to the intimate parts of a minor or the unlawful and intentional causing of a minor to touch one’s intimate parts.” NMSA 1978, § 30-9-13(A) (2003). As previously mentioned, the corpus dehcti of CSCM is established by evidence proving that (1) the minor’s intimate parts were touched by someone or that the minor was forced to touch someone’s intimate parts, and (2) that this harm was caused by someone’s criminal agency. See McKenzie, 47 N.M. at 453, 144 P.2d at 164. Under the rule in Paris, this can be established through independent evidence establishing the trustworthiness of Defendant’s extrajudicial statements plus independent proof of the loss or injury. See Buchanan, 16 N.M. at 143, 412 P.2d at 566-67.
{27} In concluding that the corpus dehcti of CSCM was established in the present case, the district court found that
[t]he State has identified certain independent circumstantial evidence (be it somewhat tenuous in nature) which tends to verify and support the corpus delicti. This circumstantial evidence (consistent behavioral symptoms by the child/victim) identifies some proof of the corpus delicti and would corroborate the trustworthiness of the Defendant’s confessions.
In order to determine whether Defendant’s statements can be considered sufficiently trustworthy in order to estabhsh the corpus dehcti, we must first decide whether the district court was correct in concluding that the independent evidence presented by the State “tend[s] to estabhsh the trustworthiness of [Defendant’s extrajudicial] statement[s].” Paris, 76 N.M. at 295, 414 P.2d at 515.
{28} Initially, we observe that the State argued below that timing and circumstances of Defendant’s statements tended to estabhsh the statements as trustworthy. More specifically, the State argued that the multiple confessions and Defendant’s disease tend to estabhsh the trustworthiness of Defendant’s statements. Although the district court based its decision on the behavioral symptoms of the daughter, we first briefly address these two arguments in order to determine whether the corpus dehcti of CSCM is met in this case. See Maralex Res., Inc. v. Gilbreath, 2003-NMSC-023, *101¶ 13, 134 N.M. 308, 76 P.3d 626 (“[A]n appellate court will affirm the district court if it is right for any reason and if affirmance is not unfair to the appellant.” (internal quotation marks and citation omitted)).
{29} With respect to the timing and circumstances of Defendant’s statements, Defendant argues that the fact that he made multiple statements concerning his daughter is not in itself sufficient to establish the trustworthiness of an extrajudicial confession. Defendant also argues that his diagnosis with Huntington’s Disease does not bolster his statements, but instead raises additional doubts about the truthfulness of his statements.
{30} We do not think that the fact that Defendant made multiple extrajudicial statements concerning his daughter is sufficient to establish the trustworthiness of his statements. “A second confession adds nothing to prove the crime admitted in the first.” Mullen, supra, at 417; see also United States v. Northrup, 482 F.Supp. 1032, 1037 (D.Nev.1980) (“If two admissions, in and of themselves, are untrustworthy, obviously they cannot be bootstrapped together to raise each other to the level of trustworthiness.”); McMillian v. State, 65 Md.App. 21, 499 A.2d 192, 198 (Ct.Spec.App.1985) (“Multiple confessions of an accused cannot be viewed as corroborative of one another.”); cf. State v. Ferry, 2 Utah 2d 371, 275 P.2d 173, 174 (1954) (“To say a second confession, with nothing more, could be used to prove the corpus delicti, obviously would be to devour the rule itself, which is predicated on the fact that one confession alone, with nothing more, cannot prove the corpus delicti,— there being no magic or significance to a repetition thereof.”). We are therefore not convinced that the mere fact that a defendant tells more than one person about his or her alleged acts necessarily establishes the trustworthiness of such statements. Contra Doe, 94 N.M. at 551, 613 P.2d at 421 (Easley, J., dissenting) (arguing in dissent that the defendant’s two confessions, plus other circumstantial evidence, established the corpus delicti in the ease); see Tilley v. State, 1998 OK CR 43, ¶ 18, 963 P.2d 607 (holding that a defendant’s multiple confessions helped to establish the trustworthiness of the initial confession), superceded by statute on other grounds as stated by Coddington v. State, 2006 OK CR 34, ¶ 54, 142 P.3d 437. Rather, we believe that the evidence used to establish the trustworthiness of Defendant’s statements must actually concern the content of his statements, not merely the circumstances surrounding them. See Doe, 94 N.M. at 549, 613 P.2d at 419 (holding that although the defendant confessed twice to stealing alcohol, there was no corroborating evidence to support the allegation that the defendant actually stole the alcohol); see also Opper, 348 U.S. at 93, 75 S.Ct. 158 (“It is sufficient if the corroboration [of the defendant’s statements] supports the essential facts admitted sufficiently to justify a jury inference of their truth.”); Geiger v. State, 907 So.2d 668, 675 (Fla.Dist.Ct.App.2005) (holding that the district court erred when it determined that a defendant’s statements were trustworthy based on the statements themselves, as opposed to “independent evidence corroborating the trustworthiness of the statements or establishing that a crime occurred”).
{31} It is for that same reason that we do not believe that the fact that Defendant was suffering from Huntington’s Disease at the time of his statements helps to bolster the trustworthiness of his statements. As we noted above, evidence of Defendant’s disease merely describes the circumstances surrounding his extrajudicial statements; such evidence does little to corroborate the essential facts of Defendant’s statements. Compare Lee, 315 F.Supp.2d at 1044-45 (finding a defendant’s confession established as trustworthy where his statements were corroborated by witness accounts of the events taking place immediately before and immediately after the abuse), and Parker, 337 S.E.2d at 495 (concluding that the defendant’s statements were sufficiently trustworthy to establish the corpus delicti where the state’s evidence paralleled the defendant’s confession), with Doe, 94 N.M. at 549, 613 P.2d at 419 (holding that a defendant’s confession to stealing alcohol from a grocery store was not corroborated where the independent evidence did not actually connect the defendant to the alcohol). Moreover, the fact that Defendant had Huntington’s *102Disease is undisputed by the parties and is purely a collateral fact that neither proves nor disproves the alleged CSCM of Defendant’s daughter. As such, we do not consider Defendant’s diagnosis with Huntington’s Disease as tending to bolster the trustworthiness of his statements. See Parker, 337 S.E.2d at 495 (“Corroboration of insignificant facts or those unrelated to the commission of the crime will not suffice.”).
{32} Aside from the timing and circumstances of Defendant’s statements, the State argues that the daughter’s behavioral symptoms, i.e., her nightmares and withdrawal from strangers, independently prove the injury to daughter and therefore also establish the trustworthiness of Defendant’s statements. Although the district court found such evidence to be rather “tenuous,” it nevertheless concluded that the daughter’s behavioral symptoms sufficiently bolstered the trustworthiness of Defendant’s confession and provided sufficient independent proof of the harm or injury in this case. We disagree.
{33} In its brief, the State argues that changes in the daughter’s behavior appearing after Defendant’s confession and subsequent arrest are consistent with sexual abuse and therefore prove the alleged injury to the daughter, thereby corroborating Defendant’s confession. Defendant argues that the daughter’s behavioral symptoms are equally consistent with the stress of living with a parent suffering from a debilitating disease or with simply being a toddler. Additionally, Defendant argues that the State’s concession below that the daughter’s behavioral symptoms do not independently establish that the daughter was the victim of sexual abuse lends further support for his argument that the State did not prove the alleged injury to the daughter.
{34} We are not convinced that the daughter’s behavioral symptoms provide adequate proof of the alleged harm to the daughter in this case. In the absence of physical evidence of the abuse, we believe that the circumstantial evidence used to independently prove the loss or injury must be more than a small number of behavioral symptoms susceptible to different inferences. See, e.g., Doe, 94 N.M. at 549, 613 P.2d at 419; Commonwealth v. Leonard, 401 Mass. 470, 517 N.E.2d 157, 159 (1988) (holding no independent evidence that a crime was committed where the proffered evidence was “ambiguous and speculative”).
{35} For example, in Commonwealth v. Villalta-Duarte, 55 Mass.App.Ct. 821, 774 N.E.2d 1144 (2002), the court found that circumstantial evidence proved that a young child had been abused when
at the same time that the defendant was present at his apartment during the day, the preverbal victim began to cry hysterically when brought to the apartment, developed an intractable diaper rash, and had scratches on her face. These symptoms, which were initially repeated at the home of a new day care provider, disappeared after a few weeks at that provider’s home.
Id. at 1148. Similarly, in Lee, 315 F.Supp.2d at 1044, the court concluded that “strong circumstantial evidence corroborated [the defendant's admission that he touched the victim’s genitalia.” Such evidence consisted of witness testimony establishing that the defendant carried the victim into a bedroom and closed the door behind him, that the defendant prevented others from entering the room, and that the victim was found on the bed with a pillow covering her upper body and wearing different clothing than before entering the room. Id. Additionally, the victim’s underwear was on backwards. Id. The court observed that in many cases of sexual abuse of a minor there will be no physical evidence of the crime, but where circumstantial evidence “strongly suggests that the crime did occur,” there is sufficient independent evidence of a loss or injury. Id.
{36} In this case, the daughter’s behavioral symptoms are simply not sufficient circumstantial evidence that a crime did occur, particularly where the symptoms are susceptible to multiple inferences. See Ray, 926 P.2d at 907 (“[C]orpus delecti is not established when independent evidence supports reasonable and logical inferences of both criminal agency and noncriminal cause.” (emphasis omitted and internal quotation marks and citation omitted)); cf. State v. Garcia, 114 N.M. 269, 275, 837 P.2d 862, 868 *103(1992) (“[E]vidence equally consistent with two inferences does not, without more, provide a basis for adopting either one____”). We note that although we typically defer to a district court’s factual findings, including all reasonable inferences therefrom, see State v. Bankert, 117 N.M. 614, 618, 875 P.2d 370, 374 (1994), the district court here found the evidence to be “tenuous.” Evidence such as this, which is equally consistent with sexual abuse and no sexual abuse, does not constitute sufficient evidence of the corpus delicti. Two behaviors out of twelve that could be corroborative of sexual abuse were cheeked on the S.A.N.E. form. While these behavioral symptoms may be suggestive of CSCM, such symptoms could be explained easily by other factors. Therefore, without more, the daughter’s behavioral symptoms are insufficient to establish independent proof of the loss or harm in this case. It is for these same reasons that we also conclude that such evidence does not establish the trustworthiness of Defendant’s confession. Because there was no independent evidence to show the corpus delicti or the trustworthiness of Defendant’s statement, Defendant’s conviction cannot stand.
CONCLUSION
{37}' We reverse the district court’s denial of Defendant’s motion to dismiss and remand this case for further proceedings consistent with this opinion.
{38} IT IS SO ORDERED.
WE CONCUR: JAMES J. WECHSLER and MICHAEL E. VIGIL, Judges.