OPINION

PAGE, Justice.

Appellant Anthony Phillip Scacchetti was convicted of first-degree criminal sexual conduct, malicious punishment of a child, and third-degree assault against his girlfriend's three-and-one-half-year-old daughter, R.J. At trial, the court determined that the child victim was incompetent to testify and, as a result, the state was allowed to introduce into evidence statements the victim made during medical assessments conducted by a pediatric nurse practitioner for Midwest Children's Resource Center (MCRC). Scacchetti appealed his conviction, and the court of appeals affirmed. On appeal to this court, we remanded to the court of appeals for a determination of whether the statements the victim made to the pediatric nurse practitioner during the assessments were admissible under Crawford v. Washington, 541 U.S. 36, 124 S.Ct. 1354, 158 L.Ed.2d 177 (2004). On remand, the court of appeals held that the statements were not testimonial under Crawford because no reasonable three-year-old would anticipate the use of the statements in a future trial and again affirmed. Because we conclude that the statements made by the victim to the pediatric nurse practitioner during the assessments are not testimonial for the purposes of Crawford, we affirm.

In May 2002, Scacchetti began living with his girlfriend, K.J., and her daughter, R.J. Before Scacchetti moved in, K.J.'s mother or K.J.'s ex-boyfriend cared for R.J. while K.J. was at work. After Scacchetti had lived with K.J. for a few weeks, Scacchetti began caring for R.J. while K.J. was working.

Soon thereafter, K.J. started noticing physical changes in R.J. On one occasion, R.J. told K.J. that Scacchetti had spanked her with a belt, which Scacchetti admitted doing. Shortly after that, K.J. noticed R.J. had marks on her face, which Scacchetti claimed were from R.J. falling down the stairs. R.J. could not explain how the injuries occurred. During May and June of 2002, K.J. noticed that R.J. had become clingy to her and distant toward Scacchetti. In fact, R.J. began to tell K.J. she did not want to be left alone with Scacchetti.

On June 24, 2002, K.J. returned from work to find that R.J.'s body had multiple bruises and burn marks. In addition, her face and one ear were swollen. Scacchetti told K.J. that R.J. had fallen in the bathtub. K.J. also found a number of bloody items, including underwear and a pillowcase belonging to R.J., and a washcloth. R.J. could not explain what had happened to her. K.J. did not go to work the next two days because she did not want to leave R.J. alone with Scacchetti.

On June 26, 2002, K.J. brought R.J. to a Saint Paul woman's shelter, where shelter employees suggested K.J. take R.J. to the hospital to be examined. On June 27, K.J. took R.J. to Minneapolis Children's Hospital, where R.J. told K.J. that Scacchetti had touched her "down there" and that it hurt when she urinated. The doctor who examined R.J. was concerned that her genital exam had indicated an abnormal hymen and called a pediatric nurse practitioner, Laurel Edinburgh, from MCRC to examine R.J.

MCRC is a clinic that assesses children for possible physical abuse, sexual abuse, and neglect. Their assessment consists of an interview and physical examination of the child. After an assessment, Edinburgh links the family with appropriate services. Edinburgh testified that she has seen over 500 patients and has testified in court about 15 to 20 times. She uses a particular protocol when assessing children, which she used when assessing R.J. Dr. Carolyn Levitt, the founder of MCRC, testified at trial regarding the protocol Edinburgh used. Dr. Levitt explained that the protocol is "pretty much based on what I would be doing if I were evaluating a child * * * who had abdominal pain and appendicitis, but it's a medical protocol and it specializes in getting specific details from the child." According to Dr. Levitt, the protocol consists of a verbal interview, an external exam, and a colposcopic exam of the child's genitalia. In explaining why the assessments are videotaped, Levitt stated, "There are many children seen at our center who have videotaped interviews so that the evaluations are then reviewed. The videotape is reviewed by me, the videotape colposcopic examination is reviewed by me."

Edinburgh assessed R.J. two times. At the first assessment, which took place at the hospital on June 27 and was not videotaped, Edinburgh first met with K.J. to gather background information and then met with R.J. to ask questions regarding her injuries. She then did a physical exam, during which K.J. was present. While examining R.J.'s anal and vaginal area, Edinburgh asked R.J., "did anything ever happen to this area right here?" to which R.J. responded "yes." Edinburgh asked "What touched there?" and R.J. responded "Tony's pee-pee." As part of the exam, Edinburgh found an oblong-shaped bruise next to R.J.'s anal opening.

Edinburgh conducted a follow-up assessment of R.J. the next day. That assessment took place in the MCRC office in Saint Paul. The office has a typical doctor's office layout, with a waiting area and two exam rooms. The assessment took place in one of the exam rooms and was videotaped. At one point during the assessment, R.J. became distraught and asked for her mother.

Edinburgh's follow-up assessment followed the standard protocol. Incidentally, during the assessment, R.J. described how Scacchetti had slapped her on her cheek and spanked her with a big belt and a brush. After further questioning, R.J. indicated that Scacchetti had touched her anus (where she "goes poop") with his hands. R.J. stated, "he put his hands right in there" while pointing up with one of her fingers. The following questions and answers ensued:

Q: Did Tony touch here with his pee-pee?

A: Yeah.

Q: Yeah. What did his pee-pee do there?

A: Him — when he was mad in there.

Q: When he was mad there?

A: Yeah. When he was mad at me.

Q: When he was mad at you?

A: Yeah.

Q: What did he do with his pee-pee there?

A: I don't know.

Q: You don't know?

A: Him put in the corner.

Q: What happened to your clothes? What did he do with your clothes?

A: He take them off.

Q: He took them off?

A: Uh-huh.

Q: Okay. What did Tony do with his clothes?

A: Um, he took them — his off too.

Q: He took his off too?

A: Uh-huh.

The court conducted a hearing to determine whether R.J. was competent to testify. The hearing took place with the judge in normal attire. During the hearing, R.J. was unwilling to orally answer questions asked of her. As a result, the court found that R.J. was incompetent to testify and was therefore unavailable. Because R.J. was unavailable, Scacchetti had no opportunity to cross-examine her.

Edinburgh testified at trial regarding both of her assessments of R.J. and statements R.J. made during those assessments, and the videotape of the second assessment was shown to the jury. She also testified that R.J., as a three-and-one-half-year-old, had sexual knowledge that was beyond her age, and that based on Edinburgh's experience R.J. was both physically and sexually abused. In explaining the basis for that conclusion, Edinburgh stated:

I use all of my knowledge of how children disclose, out of the statements that they made during [the] interview, out of what a parent tells me about a medical history, about any histories of trauma, and my physical exam, and I put those, all of those sort of factors go together when I'm making a diagnosis that a child was sexually abused.

Scacchetti testified and admitted spanking R.J. with a belt, but denied sexually abusing her.

On December 12, 2002, a jury found Scacchetti guilty of criminal sexual conduct in the first degree in violation of Minn. Stat. § 609.342, subd. 1 (2004), malicious punishment of a child in violation of Minn. Stat. § 609.377, subd. 4 (2004), and assault in the third degree in violation of Minn. Stat. § 609.223, subd. 3 (2004). Scacchetti appealed his conviction and the court of appeals affirmed on June 1, 2004. State v. Scacchetti, No. A03-301, 2004 WL 1191666 (Minn.App. June 1, 2004). Crawford was decided while Scacchetti's appeal was pending. The court of appeals did not consider the Crawford decision when it affirmed Scacchetti's conviction. Scacchetti petitioned this court for review.

We granted the petition and remanded to the court of appeals for reconsideration in light of Crawford. On remand, the court of appeals affirmed, determining that the statements R.J. made to Edinburgh were not testimonial for purposes of Crawford. State v. Scacchetti, 690 N.W.2d 393, 397 (Minn.App. 2005). Scacchetti again petitioned for review and again we granted the petition.

The Sixth Amendment to the United States Constitution ensures a criminal defendant's right "to be confronted with the witnesses against him." Until Crawford was decided, Confrontation Clause issues were determined according to Ohio v. Roberts, 448 U.S. 56, 100 S.Ct. 2531, 65 L.Ed.2d 597 (1980). Under Roberts, an unavailable witness's statements were admissible if the statements bore adequate indicia of reliability. Id. at 66, 100 S.Ct. 2531. The evidence was reliable if it fell within a firmly rooted hearsay exception or bore "particularized guarantees of trustworthiness." Id. In Crawford, the Supreme Court rejected the Roberts test, explaining that "the principal evil at which the Confrontation Clause was directed was the civil-law mode of criminal procedure, and particularly its use of ex parte examinations as evidence against the accused." Crawford, 541 U.S. at 50, 124 S.Ct. 1354. Therefore, when a witness is unavailable, testimonial statements made by the witness are inadmissible at the defendant's trial unless the defendant had an opportunity to cross-examine the witness. Id. at 53-54, 124 S.Ct. 1354. The rule "applies to `witnesses' against the accused — in other words, those who `bear testimony.'" Id. at 51, 124 S.Ct. 1354. The Court explained that testimony is typically "[a] solemn declaration or affirmation made for the purpose of establishing or proving some fact." Id. Thus, the determination of whether a statement violates the Confrontation Clause is dependent upon whether that statement is testimonial.

Our first post- Crawford Confrontation Clause case was State v. Wright, 701 N.W.2d 802 (Minn. 2005), in which we held that statements made to a 911 operator and statements made to police officers during a field investigation were not testimonial. In doing so, we concluded that statements the Supreme Court has not explicitly defined as testimonial "are best analyzed in terms of the particular facts and circumstances under which they are made in order to determine their testimonial or nontestimonial nature." Id. at 814.

We set forth eight considerations relevant to determining whether a declarant's statements are testimonial:

(1) whether the declarant was a victim or an observer; (2) the declarant's purpose in speaking with the officer (e.g., to obtain assistance); (3) whether it was the police or the declarant who initiated the conversation; (4) the location where the statements were made (e.g., the declarant's home, a squad car, or the police station); (5) the declarant's emotional state when the statements were made; (6) the level of formality and structure of the conversation between the officer and declarant; (7) the officers' purpose in speaking with the declarant (e.g., to secure the scene, determine what happened, or collect evidence); and (8) if and how the statements were recorded.

Id. at 812-13.

In State v. Bobadilla, we again applied the eight factors considered in Wright for determining whether a declarant's statement is testimonial. 709 N.W.2d 243, 250 (Minn. 2006). We clarified that, of the factors, the central considerations are "the purpose of the statements from the perspective of the declarant and from the perspective of the government questioner," in other words, "whether either a declarant or government questioner is acting, to a substantial degree, in order to produce a statement for trial." Id. at 250, 252. We noted that the other six factors are probative of those two. Id. at 250.

In Bobadilla, we held that statements made by a three-year-old child sexual abuse victim in an interview with a county child protection worker with a law enforcement officer present in accordance with Minn.Stat. § 626.556, subd. 10(a) (2004), were not testimonial. Bobadilla, 709 N.W.2d at 257. The victim in that case informed his mother that his uncle had touched him inappropriately. Id. at 246. The mother took the victim to the hospital, where law enforcement was contacted. Id. The interview took place five days later and, as noted above, was conducted by the child protection worker in the presence of a police officer. Id. at 247. Because the child was determined to be incompetent to testify at trial, the child's statements were admitted at trial through the testimony of his mother and the child protection worker and through the videotape of the interview. Id. at 248. We concluded that the child protection worker was not acting substantially to preserve testimony for trial, but was conducting the interview for the health and welfare of the child. Id. at 255. We further concluded that the three-year-old victim was not acting to preserve testimony for trial because children of that age are "simply unable to understand the legal system and the consequences of statements made during the legal process." Id. at 255-56. On that basis, we held that the victim's statements were not testimonial. Id. at 256.

In addition, we stated that the intent of the questioner, the seventh Wright factor, is one of the most important factors in determining whether statements are testimonial. Id. at 250. We also consider the identity of the questioner to be an indication of the questioner's intent. As the Supreme Court stated in Crawford, "Involvement of government officers in the production of testimony with an eye toward trial presents unique potential for prosecutorial abuse." Crawford, 541 U.S. at 56 n. 7, 124 S.Ct. 1354 (emphasis added). Therefore, the fact that the questioner is a government officer is probative of the questioner's purpose for eliciting the statements in question. With respect to statements made to nongovernment questioners, we must determine whether the questioner was acting in concert with or as an agent of the government.

We begin our analysis by examining the identity of and the purpose of the questioner. It is significant in this case that no government actor initiated, participated, or was involved in any way with the assessments of R.J. that resulted in the statements at issue. In Bobadilla, even though police were involved in organizing the child's interview, were present at the interview, the interview was conducted at the police station, and the interviewer was a government questioner, we held that the child's statements were not testimonial. 709 N.W.2d at 246-47, 256. Unlike the questioner in Bobadilla, the record here indicates that Edinburgh is not a government questioner. Edinburgh is a nurse practitioner employed by MCRC, which, as Dr. Levitt testified at trial, is a department of Children's Hospital. Also, unlike Bobadilla, R.J.'s statements were made to Edinburgh at the hospital and in a doctor's office, not at a law enforcement center. Further, the police were not involved in organizing R.J.'s assessments and were not present for and did not participate in the assessments. Therefore, we conclude, based on the record presented, that Edinburgh was not a government questioner and that she was not acting in concert with or as an agent of the government. Having concluded that neither the police or any government actor, nor anyone acting in concert with or as an agent of the government was involved in R.J.'s assessments, we further conclude that R.J.'s statements to Edinburgh were not testimonial.

Even if we had concluded that Edinburgh was acting in concert with or as an agent of the government, our conclusion that R.J.'s statements to Edinburgh are not testimonial would not change. The record here indicates that Edinburgh's purpose in interviewing and examining R.J. was to assess her medical condition. Both Edinburgh and Dr. Levitt testified that their purpose in evaluating children such as R.J. is to determine whether the child has been abused and, if necessary, to connect the child and family to appropriate services. There is no evidence or other testimony in the record to the contrary. The fact that MCRC generally does not have ongoing contact with the child after the assessment does not minimize the medical purpose for which the assessment is conducted.

Scacchetti argues that, even if the court focuses on the medical purpose for the assessments, the fact that Edinburgh has testified in past child sexual abuse cases with relative frequency indicates that she is acting for the purpose of collecting statements for use in a later prosecution. First, we note that Scacchetti's contention that Edinburgh testifies in child sexual abuse cases with relative frequency is belied by the only evidence in the record on this subject. According to Edinburgh's testimony, she has examined over 500 patients and of those she testified in 15 to 20 cases. We do not believe, on these facts, that testifying 15 to 20 times constitutes relatively frequent testimony.

Moreover, the mere fact that Edinburgh may be called to testify in court regarding sexual abuse cases does not transform the medical purpose of the assessments into a prosecutorial purpose, nor is there any evidence that Edinburgh had a prosecutorial purpose here. Further, in Bobadilla we specifically held that the fact that the interview of the child may have had the secondary purpose of preserving testimony does not make a statement testimonial if that purpose is incidental to a broader purpose of protecting the child's welfare. Bobadilla, 709 N.W.2d at 255. Thus, even if we were to conclude that Edinburgh's assessments of R.J. had, as a secondary purpose, the preservation of testimony for trial, R.J.'s statement would still not be testimonial. Because the broad purpose of Edinburgh's assessments was R.J.'s medical health, any subsequent testimony that Edinburgh was required to give did not change her assessment purpose.

At least three other courts have held that statements made by children are not testimonial when taken for medical assessment purposes without the involvement of government actors. See United States v. Peneaux, 432 F.3d 882, 896 (8th Cir. 2005) (holding statements given to pediatrician for diagnosis or treatment are "presumptively nontestimonial"); State v. Vaught, 268 Neb. 316, 682 N.W.2d 284, 291-92 (2004) (finding statement made by four-year-old to emergency room physician following sexual abuse was not testimonial); State v. Fisher, 130 Wash.App. 1, 108 P.3d 1262, 1269 (2005) (holding statement made by 29-month-old to family physician the morning after child's admission to the hospital was not testimonial). At least one commentator has suggested that, in Crawford, the United States Supreme Court implicitly took a favorable view of statements made for the purposes of medical treatment. Robert P. Mosteller, Crawford v. Washington: Encouraging and Ensuring the Confrontation of Witnesses, 39 U. Rich. L.Rev. 511, 600 (2005).

Further, we also note that the first five Wright factors support the conclusion that R.J.'s statements to Edinburgh are not testimonial. R.J. was the victim in this case, and not a mere observer. Given R.J.'s age at the time Edinburgh assessed her, it is not clear that R.J. knew or understood the purpose of the statements she made to Edinburgh. We do know, however, that R.J. made the statements in response to questions asked during Edinburgh's medical assessments and that the statements were made at a hospital and in a doctor's office. R.J.'s mother initiated R.J.'s medical assessment when she brought R.J. to the hospital and requested that R.J. be examined. And, like the declarants in Wright, R.J. was emotionally distraught when discussing the alleged abuse with Edinburgh.

Finally, we note that the sixth and eighth Wright factors do not clearly support a conclusion one way or the other as to whether the statements are testimonial. Although the questions that Edinburgh asked R.J. followed a particular protocol and were structured to elicit responses from a young child, the questions were designed to elicit answers related to the child's medical health and not responses that would necessarily be useful in prosecuting an offender. Dr. Levitt specifically testified that the structure of the questioning is necessary to elicit useful responses and obtain specific medical details from the child. No evidence to the contrary was presented. Therefore, while the questioning was structured, we conclude that it was not structured for the purpose of creating useful prosecutorial statements. Similarly, although the statements R.J. made during the second assessment were recorded by Edinburgh, the trial record indicates only that the recording was made to allow Dr. Levitt to review the assessment.

Taken separately or together, the absence of government involvement in the questioning of R.J. and the Wright factors, viewed against the backdrop of the record presented, lead us to the conclusion that the statements R.J. made during Edinburgh's assessments are not testimonial. We therefore hold that admission of the statements at trial did not offend Scacchetti's Sixth Amendment right to confront the witnesses against him.

Affirmed.

GILDEA, J., not having been a member of this court at the time of the argument and submission, took no part in the consideration or decision of this case.