The defendant, Richard Langill, appeals his conviction on one count of burglary, RSA 635:1 (2007). This is the second time that this case has been before us. See State v. Langill, 157 N.H. 77 (2008). In the present appeal, the defendant contends that the Superior Court {Lewis, J.) erred when it: (1) denied his motion to strike the testimony of the State’s expert witness identifying the defendant’s fingerprint; (2) admitted into evidence a photograph of the latent fingerprint; and (3) denied his motion to dismiss. We reverse and remand.
I
The record supports the following facts. Karen Katz lived with her two children in a first-floor apartment with a balcony. She had stored $1,200, including ten $100 bills, in a locked box in the top drawer of her bedroom dresser. When she left her home on the morning of March 25,2004, she took $20 from her lockbox. She returned home that afternoon and discovered that all of the money in her lockbox was missing. In addition, $5 was missing from a plastic bottle that her daughter used as a bank. There was no evidence of forced entry into the apartment. The apartment manager testified that she had misplaced a set of master keys to the apartment building sometime during the summer of 2003.
At the time of the burglary, the defendant lived in an apartment in the same building as Katz, which he shared with Justine DeCotis and her two children. Katz testified that she did not know the defendant, but that prior to the burglary, she occasionally saw him outside late at night or with the children, and noticed that “when he would walk by your door, if your blinds were open, he would like stare in.”
On the day Katz discovered that her money was missing, Derry police officer James Belanger arrived at her apartment. He took the lockbox and the plastic bottle to be processed for fingerprints. He also spoke with several tenants in the building, including the defendant. The defendant initially told Officer Belanger that he was ill and had not left his apartment all day. However, he then corrected himself and said that he left his apartment to retrieve his mail, which was in the same building as his apartment. Belanger testified at trial that the defendant was nervous and responded to questions slowly.
The next day, Christina Piccirilli, who worked for a salvage yard, sold a 1992 Ford Taurus to an individual whom she identified only as DeCotis’s *221boyfriend. Piccirilli testified that the car was purchased with two $100 bills. On March 30, Detective Joseph Bennett processed the lockbox and the plastic bottle for fingerprints. He found four sets of latent fingerprints on the lockbox and five on the bottle. He lifted the prints using lifters and placed the lifters in sealed evidence bags that could not be opened without being cut. The lifters were tagged and later transported to the New Hampshire Department of Safety Forensic Laboratory (state lab) for further testing.
On April 1, Derry police officer Mike Houle went to Katz’s apartment to process Katz’s bedroom dresser. He found one latent fingerprint on top of the dresser close to the drawers, which he lifted with a two-by-four inch clear lifter. He then placed the lifter into a plastic evidence bag, and completed a transmittal form requesting further analysis from the state lab.
Lisa Corson, a criminalist at the state lab testified at trial about her examination of certain fingerprints. She explained how fingerprint comparison is conducted using the ACE-V methodology, and that ACE-V is an acronym standing for Analysis, Comparison, Evaluation and Verification. Corson testified that the fourth step of the ACE-V process requires that another qualified examiner complete the first three stages of the ACE-V methodology to determine whether the verifying examiner arrived at the same result as the primary examiner.
Corson testified that in May 2004 the state lab received two separate exhibits from the Derry Police Department, one containing five fingerprint lifts and one containing a single fingerprint lift. Corson performed an analysis of the prints in 2005. Regarding the exhibit containing five lifts, Corson testified that none of these lifts contained identifiable latent prints. Regarding the exhibit containing a single lift, Corson determined that it was an identifiable fingerprint, which she was able to further analyze.
At trial, the jury was shown State’s Exhibit 1, a chart Corson had prepared which compared a photograph of the single latent print and a known print. Corson testified that the known print belonged to the defendant. Corson described how she examined the single latent print along with the known print using the ACE-V methodology. The initial step of the examination involved an analysis of the latent print. Based upon her analysis, Corson concluded that the latent and known prints had sufficient quality and enough detail to permit her to continue with the second and third steps of the examination, which involve a comparison and evaluation of the latent print and the known print. Using Exhibit 1, Corson testified about the various similarities she found between the latent print and the known print, and stated that she was able to make an identification between the latent print and the known print.
*222When the prosecutor asked Corson if her work had been verified, the defendant objected, arguing that the statement was hearsay and that it violated his right to cross-examine witnesses against him. The State argued that the testimony was admissible because it would be based upon Corson’s review of documentation from the state lab reflecting that her work had been verified. The trial court agreed that the verification testimony was hearsay, but overruled the defendant’s objection, relying upon the business records exception to the hearsay rule, and allowed the State to present evidence that Corson’s opinion had been verified. The trial court stated:
[S]he doesn’t have to testify to . . . how well the verification was performed, does she, for her to be able to say that she believes, looking at the business records and looking at what she understands, that her conclusion is okay under her methodology, then you can cross-examine her on that. And maybe the State . . . creates a problem for itself if it doesn’t produce the verification person ....
The trial court instructed the prosecutor to rephrase her questions to “dealt ] with the business record exception.”
Thereafter, the prosecutor asked Corson to review State’s Exhibit 5, consisting of Corson’s laboratory worksheets with her analysis of the latent print. Corson testified that in 2005, it was the practice of the state lab for the person verifying the work to note that fact in the records, and that “there was a worksheet where they would sign off.” In describing the worksheet, Corson testified that “[t]he last column is where the other examiner who is given the latent for verification puts their initials and the date that they do their ACE to the case.” She testified that the initials of the verifying examiner were Timothy Jackson’s, and also stated, “In this case my work was verified by Timothy Jackson.” On cross-examination, Corson agreed that she did not actually know how Jackson conducted his verification, and that she could not speak to the quality of his verification. On redirect, she testified that Jackson was certified by the International Association of Identification (IAI) when he verified her work. Jackson did not testify.
The court then permitted State’s Exhibit 5 to be received into evidence, over the defendant’s objection. At a chambers conference regarding the court’s evidentiary ruling on State’s Exhibit 5, the trial court stated:
[lit was allowed to be introduced as a business record through the foundation that it was regularly maintained in the course of business, etcetera. And it was introduced for the purpose of *223showing that a verification occurred. How good it was, whatever was not the point here, but the business reflects that a verification occurred.
That, as I understand it, under the methodology that was described, would allow the witness here, Ms. Corson, to testify as to her opinion under the methodology we’re dealing with. And therefore I allowed it in.
The only disturbing aspect of it I have here is that obviously by just the — telling the jury that there was some verification here, there is a kind of [sub silentio] implication that the verification is consistent with what Ms. Corson said. And I — that’s a major concern I have. And I would entertain arguments as to whether I should be giving them some sort of limited instruction regarding that later on ... .
It’s only being introduced for the fact that someone reputable, in a position to do so marked on this paper verification which triggers the ability of Ms. Corson to give her opinion under this methodology.
Ultimately, the court refused to strike Corson’s testimony, but gave the jury two limiting instructions regarding Corson’s testimony and Jackson’s verification.
At the close of the evidence, the defendant objected to State’s Exhibit 1, the chart Corson had prepared that compared a photograph of the single latent print and a known print, being admitted into evidence. The defendant also moved to strike Corson’s testimony that the latent print matched the defendant’s print, and to dismiss the case on thé grounds that the evidence was legally insufficient to support a conviction. The court allowed State’s Exhibit 1 to be admitted into evidence and denied both motions. The jury found the defendant guilty of burglary. This appeal followed.
II
The defendant first argues that the trial court erred in failing to strike Corson’s identification of the fingerprint and in admitting the verification evidence. The defendant contends that Corson’s testimony that her work had been verified by Jackson constituted inadmissible hearsay and violated *224his right to confrontation. Because we agree that Corson’s testimony-regarding the verification and verification evidence constituted inadmissible hearsay, we do not reach the constitutional argument. See State (Premo Complainant) v. Martineau, 148 N.H. 259, 260 (2002).
“Hearsay is generally defined as an extrajudicial statement offered in court to show the truth of the matter asserted in the statement.” State v. Soldi, 145 N.H. 571, 575 (2000); see N.H. R. Ev. 801(c). The definition of “statement” includes “an oral or written assertion.” N.H. R. Ev. 801(a). In general, such extrajudicial statements, which are not made under oath or subject to cross-examination, are less trustworthy than those made in court. State v. Cole, 139 N.H. 246, 249 (1994). Hearsay is not admissible into evidence unless it falls within an exception. N.H. R. Ev. 802. We will reverse a trial court’s decision that a statement is not hearsay or falls within an exception to the rule only if the defendant demonstrates that the ruling is clearly untenable or unreasonable to the prejudice of his case. State v. Connor, 156 N.H. 544, 546 (2007); State v. Pelletier, 149 N.H. 243, 253 (2003).
The trial court relied upon the business records exception to the hearsay rule. On appeal, however, the State argues not that the evidence was admissible under an exception to the hearsay rule, but that it was not offered for its truth and, therefore, was not hearsay. According to the State, “Corson’s testimony about the verification was not hearsay, but rather, a confirmation that she fulfilled a specific methodology that allowed her to issue her opinion.” We disagree.
Corson testified at trial that she forms her opinion after following the analysis, comparison, and evaluation steps of the methodology. Once she concludes that the prints match, the identification moves to the verification phase. This fourth step of the ACE-V methodology requires that another qualified examiner complete the first three stages, independent of the primary examiner’s work, to determine whether he or she arrives at the same result as the primary examiner. Specifically, Corson testified that “the latent [print] and the known are given to another qualified examiner who will then conduct their [sic] own analysis, comparison and evaluation on the latent to the known and arrive at the same conclusion that [the initial examiner did].” She testified, after reviewing the State laboratory work papers, that the initials of the verifying examiner found in the appropriate column on the work papers belonged to Jackson, and that “[i]n this case my work was verified by . . . Jackson.” On redirect examination, Corson testified that Jackson was certified by the IAI when he verified her work.
We conclude, as we did in State v. Connor, 156 N.H. at 546, that the primary examiner’s testimony regarding the process undertaken and the *225verifying examiner’s ultimate opinion was offered for its truth, as distinguished from the mere satisfaction of procedure. Connor involved an identification that Jackson made of fingerprints found at a crime scene and Jackson’s testimony that Corson had verified his identification. Id. at 545. The defendant in Connor argued that Jackson’s testimony about Corson’s verification was hearsay and violated his right to confrontation, and that the trial court should have stricken Jackson’s entire testimony. Id. at 546. The State argued that Jackson could testify that Corson had verified his work because it was not hearsay, and, even if it was hearsay, it was admissible under New Hampshire Rule of Evidence 703. Id. at 546, 547. In concluding that Jackson’s testimony regarding Corson’s verification of his work was hearsay because Corson’s ultimate opinion was offered for its truth, we stated:
The verification process, as described by Jackson, supports our conclusion. Jackson’s testimony clearly illustrates that the verification is not conducted to ensure that he had followed the applicable procedures. Corson did not simply check that the equipment or procedure used by Jackson was proper or that Jackson employed the correct number of comparison points in making his determination. Rather, it is clear that her task was to affirm Jackson’s identification by undertaking an independent analysis, comparison and evaluation of the fingerprint, and ultimately forming her own opinion. By its very nature, the purpose of this verification, as described by Jackson, lies in the truth of Corson’s opinion, that is, that her independent ACE procedure resulted in the same conclusion, thus corroborating Jackson’s opinion. Jackson’s testimony relating to the verification process and Corson’s independent opinion extends well beyond establishing Jackson’s compliance with procedure. Under these circumstances, this evidence constitutes inadmissible hearsay unless it falls under an exception to the rule.
Id. at 546-47 (citation omitted); see also People v. Smith, 628 N.E.2d 1176, 1181 (Ill. App. Ct. 1994) (fingerprint technician’s testimony that her identification was verified by another technician was hearsay); State v. Wicker, 832 P.2d 127, 129 (Wash. Ct. App. 1992) (same). The same reasoning applies here.
The State seeks to distinguish our prior decision, in part, on the ground that in Connor, the testifying examiner stated that the verifying examiner had determined that the latent print found “was, in fact, made by the left middle finger from the individual whose name appears on the fingerprint card of David Connor,” Connor, 156 N.H. at 546 (brackets omitted), *226whereas, in the present case, Corson testified only that her work had been verified. We note, however that in ordinary language, the definitions of “verify” include “to prove to be true” “establish the truth of’ and “to swear or affirm the truth of.” WEBSTER’S Third New INTERNATIONAL DICTIONARY 2543 (unabridged ed. 2002). Furthermore, it is clear from Corson’s testimony in this case regarding the ACE-V methodology that the verification step requires the verifying examiner to perform the first three steps of the methodology to determine whether the primary examiner’s conclusions were correct. Therefore, implicit in Corson’s testimony that Jackson had verified her conclusion is the statement that Jackson performed the ACE analysis, and, like Corson, reached the conclusion that the latent print matched the defendant’s print. We see no meaningful distinction between Corson testifying that a second examiner performed the ACE analysis and affirmed the truth of her conclusion that the latent print matched the defendant’s print, and Corson testifying that a second examiner performed the ACE analysis and stated that the latent print matched the defendant’s. As we reasoned in Connor, “By its very nature, the purpose of this verification, as described by [the testifying examiner], lies in the truth of the [verifying examiner’s] opinion, that is, that her independent ACE procedure resulted in the same conclusion, thus corroborating [the testifying examiner’s] opinion.” Connor, 156 N.H. at 547.
We are not persuaded by the State’s argument that Corson’s testimony was not hearsay because it was “not offered for the truth of the matter asserted because it was offered solely to establish the foundation that allowed Corson to issue her opinion.” Having chosen to introduce fingerprint evidence based upon ACE-V methodology, the State had the burden of presenting evidence that each step in the ACE-V procedure was followed. Cf State v. Dahood, 148 N.H. 723, 735 (2002) (results of HGN test only admissible if State establishes proper foundation, including that test was properly administered). Because the ACE-V methodology itself requires two experts to conclude, independently, that the prints match, the State was required to prove that a second examiner performed the ACE analysis and that he agreed with the primary examiner’s opinion that the latent print matched the known print. Thus, it is clear that Jackson’s out-of-court statement, “I verify this work,” communicates that he agreed that the fingerprints matched, and was offered to prove the truth of the matter asserted, that is, that he agreed that the fingerprints matched. We note that the State did not seek to prove the admissibility of Corson’s opinion outside • the presence of the jury. Accordingly, we express no opinion as to whether the State may prove the foundational requirements necessary for the admissibility of fingerprint opinion testimony based on ACE-V methodology outside the presence of the jury. See N.H. R. Ev. 104.
*227The State argues that the fact that the defendant’s cross-examination of Corson established that Corson did not know the procedure that Jackson followed and that she could not speak to the quality of his verification illustrates that the testimony was both “significantly more limited” than it was in Connor and was admitted only to establish that Corson complied with procedure. We do not agree. Rather, the nature of the cross-examination underscores the problem presented by the hearsay testimony. According to the State’s evidence, under the ACE-V methodology, an examiner’s conclusion that a set of prints match must be verified by a second examiner. Indeed, Corson acknowledged on cross-examination that verification is critical to the methodology. Thus, Corson’s testimony that her work was verified necessarily assumes, and asks the trier of fact to assume, that Jackson performed the ACE analysis, that he did so accurately, and that he agreed with Corson’s opinion that the latent print matched the known print. As the cross-examination makes clear, when the primary examiner testifies about the verification, the defendant is unable to challenge the second examiner’s statement that the prints matched. This is precisely the problem the rule against hearsay is designed to address. See Cole, 139 N.H. at 249.
We also disagree -with the State’s argument that the fact that limiting instructions were given in this case should lead to a result different from the one we reached in Connor. Prior to closing arguments, the court gave the jury the following limiting instruction:
You have heard testimony about Timothy Jackson. Lisa Corson identified initials of “TAJ” ... as those of Timothy Jackson, a criminalist at the laboratory. However Mr. Jackson did not testify in this case. You are instructed that the only opinion before you with regard to fingerprint identification is that of Ms. Corson and not of the verifier, Mr. Jackson. You are not to consider the verifier’s work in this case as an additional opinion or as any way a supplement of Ms. Corson’s opinion. You must consider Ms. Corson’s opinion on its own merits without regard to the verifier’s actions as to this matter.
The next day, following closing arguments, the court read the full set of jury instructions, which included the following limiting instruction:
Now, please be reminded that the only expert opinion before you with regard to fingerprint identification is that of Ms. Corson and not Mr. Jackson. You are not to consider any other examiner’s work in this case as an additional opinion or in any way as a supplement of Ms. Corson’s opinion. You must consider Ms. *228Corson’s opinion on its own merits without regard to any other examiner’s actions in this matter.
The State contends that these instructions prohibited the jury from considering Jackson’s opinion for its truth (that is, that he agreed that the latent print matched the defendant’s print). We do not believe that they can be read this way. While the jury was instructed not to consider the verifier’s work as an “additional opinion,” or as a “supplement to” Corson’s opinion, the instructions still permitted the jury to conclude that the work had been verified. As we have explained, the testimony that Jackson had verified Corson’s conclusion is tantamount to testimony that Jackson had affirmed the truth of Corson’s match and had concluded that the latent print matched the defendant’s print, and is hearsay.
Accordingly, we hold that the trial court erred when it denied the defendant’s motion to strike Corson’s testimony and in admitting the verification evidence. Because the fingerprint evidence was a critical piece of evidence physically linking the defendant to the crime, we conclude that admission of the testimony and the verification evidence prejudiced the defendant. Therefore, we reverse and remand.
Because we reverse his conviction upon his first claim of error, we do not reach the defendant’s argument that the trial court erred when it admitted into evidence State’s Exhibit 1, consisting of a photograph of the latent print and the defendant’s known print. We must, however, address the defendant’s argument that the evidence presented to the trial court was insufficient to reach a guilty verdict because, if the evidence were insufficient, the Double Jeopardy Clauses of the State and Federal Constitutions would preclude a remand for a new trial. State v. Sweeney, 151 N.H. 666, 672-73 (2005).
Ill
In determining whether there was sufficient evidence to reach a guilty verdict, we consider all the evidence, including evidence that we have concluded was erroneously admitted. State v. Horak, 159 N.H. 576, 582-83 (2010); Lockhart v. Nelson, 488 U.S. 33, 34 (1988) (holding that “where the evidence offered by the State and admitted by the trial court — whether erroneously or not — would have been sufficient to sustain a guilty verdict, the Double Jeopardy Clause does not preclude retrial”).
To prevail on a challenge to the sufficiency of the evidence, a defendant must prove that no rational trier of fact, viewing all of the evidence and all reasonable inferences from it in the light most favorable to the State, could have found guilt beyond a reasonable doubt. State v. Evans, *229150 N.H. 416, 424 (2003). Where, as in this case, the evidence is solely circumstantial, it must exclude all rational conclusions except guilt. Id. Under this standard, however, we still consider the evidence in the light most favorable to the State and examine each evidentiary item in context, not in isolation. Id.
RSA 635:1,1 (2007) provides, in pertinent part, that a person is guilty of burglary “if he enters a building or occupied structure ... with purpose to commit a crime therein, unless the premises are at the time open to the public or the actor is licensed or privileged to enter.”
We conclude, viewing the evidence and reasonable inferences from it in the light most favorable to the State, that the evidence is sufficient to exclude all rational conclusions other than that the defendant was guilty of burglary. The defendant lived in the same apartment complex as Katz, and admitted to being home on the day of the burglary and to leaving his apartment at one point. The defendant could easily access Katz’s apartment because she lived on the first floor of the building and her apartment had a ground-level balcony. Furthermore, the apartment manager had lost the keys to the building the previous summer. Katz testified that the defendant would “stare in” to her apartment if the blinds to her apartment were open. When Officer Belanger questioned the defendant about the incident, the defendant first told the officer that he had been in his apartment all day, and then corrected himself, saying that he left at one point to retrieve his mail. Officer Belanger testified that the defendant was nervous during the questioning, had some difficulty answering questions, and responded to questions slowly. In addition, there was evidence that the defendant purchased a car from a salvage yard the day after the robbery with two $100 dollar bills, and that the money taken from Katz’s apartment came, in part, in the form of ten $100 dollar bills. Finally, and critically, there was testimony that the defendant’s print matched a latent print Officer Houle lifted from Katz’s dresser after the burglary.
The defendant’s argument that the evidence is insufficient because Piccirilli never identified him as the man to whom she sold the 1992 Ford Taurus does not alter our conclusion. Piccirilli sufficiently identified the man as the defendant because she described him as DeCotis’s boyfriend, and the evidence at trial established that the defendant lived with DeCotis at the time of the burglary. We also disagree with the defendant that the evidence was insufficient because there was no evidence that the defendant ever had possession of the master keys for the apartments and that Katz’s testimony that she would see the defendant staring into her apartment was inconclusive. Considered alone, these facts may not exclude all rational *230conclusions except guilt, but our task is to examine each evidentiary item in context, not in isolation. Evans, 150 N.H. at 424.
Finally, we disagree with the defendant that in assessing the sufficiency of the evidence, we “must discount the fingerprint evidence given the lack of a proven connection between the latent print taken and the latent print examined.” The defendant argues that the trial court erred when it admitted into evidence State’s Exhibit 1, the photograph of the latent print and the defendant’s known print, on the grounds that there was “no evidence” linking that latent print Corson examined to the one Officer Houle recovered at the burglary scene. However, because we reverse the defendant’s conviction on the grounds that the trial court erred when it denied his motion to strike Corson’s identification and in admitting the verification evidence, we do not reach this issue. We are unpersuaded by the defendant’s assertion, made without citation to any legal authority, that in determining whether there was sufficient evidence, we should discount the fingerprint evidence given the alleged “lack of a proven connection” between the latent print taken and the latent print examined. Cf. Horak, 159 N.H. at 582-83; Lockhart, 488 U.S. at 34.
Viewing all of the evidence and reasonable inferences from it in the light most favorable to the State, we hold that it was sufficient for the jury to exclude all rational conclusions except guilt. Because, however, the trial court erred when it denied the defendant’s motion to strike Corson’s testimony and in admitting the verification evidence, we reverse the defendant’s conviction and remand.
Reversed and remanded.
DALIANIS, Duggan and Conboy, JJ., concurred; Hicks, J., concurred in part and dissented in part.