OPINION
ROSS, JudgeA jury in án assault trial heard testimony that -appellant Travis Andersen punched his girlfriend A.A. -in the face and that he had. previously assaulted, her, A.A.’s treating physician’s assistant testified that she’ examined A.A.’s x-rays and concluded that. A.A. suffered a broken nose, and the prosecutor introduced a ra-r diologist’s report confirming- that A.A.’s nose was broken. The jury found Andersen guilty. Andersen argues that the district court abused its discretion by allowing A.A. to testify about Andersen’s prior abuse and violated his constitutional right to confront witnesses by admitting the radiologist’s report. We affirm because the prior-abuse, testimony'was admissible relationship evidence and' because the radiologist’s report, was not testimonial in nature.
FACTS
The state charged Travis Andersen with domestic assault and third-degree-assault, *440among other things, after A.A. reported in July 2015 that he punched her several times while she was driving. The jury in Andersen’s March 2016 trial heard from A.A. and the physician’s assistant who treated her.
A.A. testified that she and Andersen began dating in August 2014. Andersen soon began verbally abusing her. By October 2014, A.A. moved in with Andersen and his parents. Sometime in late October or early November, Andersen and A.A. got into an argument during which Andersen hit A.A. in the face. The district court cautioned the jury to avoid misusing this testimony. A.A. told the jury that Andersen went to prison in December 2014.
A.A. said that Andersen stopped verbally abusing her only for a time,. She decided to move out of Andersen’s parents’ home and end the relationship in the spring of 2015, while Andersen was still in prison. But she saw him on July 2, 2015, after he was released. She invited Andersen to stay with her, and on July 7, 2015, she picked Andersen up from his parents’ home. They had an argument during the drive and Andersen grabbed the steering wheel. A.A. stopped in a parking lot, and Andersen yelled at her and threw a beverage at her. A.A. testified that she slapped Andersen and pulled out of the parking lot. They continued arguing, and she hit him once more.
Andersen retaliated. He punched A.A. in the face “four to five times” while she drove. A.A.’s nose bled and she thought it was broken. A.A. took Andersen to his parents’ house and, after Andersen insisted, went inside with him “because [she] was scared since he was still with [her] and [she] was confused.” A.A. said that she tried to leave the house several times but Andersen repeatedly grabbed her wrist and pulled her back.
She testified that the following morning, July 8, 2015, she again attempted to leave and again Andersen prevented her. He called her names, put his hand around her neck, and raised his fist. After A.A. eventually left, she went to her home and then to the hospital, where she told medical staff that she had suffered a domestic assault.
Chelsea Eernisse, an emergency-room physician’s assistant, also testified. Eer-nisse told the jury that she is trained in radiology. She said she had examined hundreds of broken noses, and she was qualified to order x-ray scans and independently interpret them. She was present at the hospital when A.A. came for treatment on July 8, 2015, and after she viewed A.A.’s x-rays she concluded that A.A. suffered “an acute non-displaced nasal bone fracture.” A medical-center employee summoned the police to the hospital. Eernisse explained to the district court that it was “standard practice to send every film ... to a radiologist for a second opinion.” Before police arrived, Eernisse told A.A. that “the radiologist would also be reviewing the x-ray and confirming [her] findings as well.”
The x-ray scans were automatically sent to the radiologist, Dr. Curtis Binder, for review. Dr. Binder reviewed A.A.’s x-rays and completed a report in which he also concluded that A.A. had suffered a nasal fracture. Dr. Binder did not testify at trial. Over Andersen’s objection, the district court admitted Dr. Binder’s report into evidence. It held that the report did not implicate Andersen’s right to confront witnesses and that it satisfied the business-records hearsay exception.
The jury found Andersen guilty of domestic assault, third-degree assault, and obstruction of legal process. The district court entered a conviction for the third-degree assault and sentenced him to 28 months in prison. Andersen appeals.
*441ISSUES
I. Did the district court abuse its discretion by allowing A.A. to testify that Andersen previously abused her?
II. Did admitting Dr. Binder’s report violate Andersen’s confrontation right?
ANALYSIS
Andersen asks us to reverse his assault conviction and remand for a new trial. He maintains that the district court abused its discretion by allowing A.A. to testify about his previous abuse, and that the district court violated his constitutional right to confront witnesses who testify against him by admitting Dr. Binder’s report into evidence. Neither argument is convincing.
I
Andersen faults the district court for permitting A.A. to testify that he previously verbally and physically abused her. We review the district court’s decision to admit relationship evidence-for an abuse of discretion. State v. Loving, 775 N.W.2d 872, 879 (Minn. 2009). To secure a reversal, an appellant must demonstrate that the district court erred by admitting the evidence and that the erroneously admitted evidence substantially influenced the jury’s decision. Id.
Andersen argues that the district court abused its discretion because the danger of unfair prejudice outweighed any probative value. A district court has discretion to admit presumptively admissible relationship evidence in domestic-abuse cases:
Evidence of domestic conduct by the accused against the victim of domestic conduct, or against other family or household members, is admissible unless the probative value is substantially outweighed by the danger of unfair prejudice. ...
Minn. Stat. § 634.20 (2014) (emphasis added). This kind of evidence is relevant “to illuminate the history of the relationship” so as to put the charged crime in context. State v. McCoy, 682 N.W.2d 153, 159 (Minn. 2004).
Andersen argues that A.A.’s testimony about his prior abuse falls short of this relevancy objective because a key issue was whether A.A. fabricated her assault allegations and the testimony does not bear on that issue. He believes that the challenged testimony was relevant instead only to establish “that A.A. alleges assaul-tive behavior—not that her allegations in this case were true.” The argument misses the mark. Relevant evidence is any evidence that tends to make the existence of any consequential fact more or less probable. Minn. R. Evid. 401. That Andersen verbally and physically abused A.A. earlier in their relationship tends to illuminate the volatility of the relationship and put the July 7, 2015 incident in the context of the couple’s interaction with each other. And relationship evidence, like Spreigl evidence, can also be admitted to demonstrate motive and intent. See Loving, 775 N.W.2d at 880. The testimony was not chiefly that A.A. alleged prior abusive behavior, but that Andersen had actually engaged in prior abusive behavior. Her testimony informed the jury of the nature of their relationship, the times that she felt afraid of Andersen, and the times that Andersen attempted to manipulate, control, and restrain her. This testimony has obvious probative value.
As for. the counterbalance against admitting this probative evidence—the risk of unfair prejudice—we begin by observing that the district court’s cautionary instructions lessened any probability that the jury would rely improperly on relationship evi*442dence. See State v. Lindsey, 755 N.W.2d 752, 757 (Minn.App. 2008), review denied (Minn. Oct. 29, 2008). Andersen argues that “it was nearly impossible” for the jury to fairly evaluate the charges after it heard the relationship evidence, despite the district court’s cautionary instructions. He would have us hold that the danger of confusing the jury outweighed the testimony’s probative value. But to become inadmissible, the presumptively admissible relationship evidence must be more than merely potentially prejudicial, and it must be more than merely potentially unfairly prejudicial; the evidence becomes inadmissible' only if its danger for unfair prejudice substantially outweighs its probative value. Andersen’s argument does not suggest that A.A.’s testimony reaches that line. And we hold that it does not. The district court properly admitted the relationship evidence.
II
We next consider Andersen’s contention that the district court erroneously determined that admitting Dr. Binder’s report did not violate Andersen’s right to confront witnesses. The Sixth Amendment’s Confrontation Clause provides a criminal defendant the right “to be confronted with the witnesses against him.” U.S. Const. amend. VI; see also Minn. Const. art. I, § 6. Whether admitting evidence violates a defendant’s confrontation right is a question of law that we review de novo. State v. Caulfield, 722 N.W.2d 304, 308 (Minn. 2006). A violation of a defendant’s confrontation right warrants reversal of his conviction unless the violation is harmless beyond a reasonable doubt. See State v. Courtney, 696 N.W.2d 73, 79 (Minn. 2005).
The Confrontation Clause applies only to testimonial evidence.. See Davis v. Washington, 547 U.S. 813, 821, 126 S.Ct. 2266, 2273, 165 L.Ed.2d 224 (2006); Crawford v. Washington, 541 U.S. 36, 51, 124 S.Ct. 1354, 1364, 158 L.Ed.2d 177 (2004), The district court may admit testimonial evidence “only where the de-clarant is unavailable, and only where the defendant has had a prior opportunity to cross-examine.” Crawford, 541 U.S. at 59, 124 S.Ct. at 1369; see also State v. Swaney, 787 N.W.2d 541, 552 (Minn. 2010). We must therefore decide whether Dr. Binder’s report constitutes testimonial evidence.
The United States Supreme Court has clarified what constitutes “testimonial” evidence. The Crawford Court identified three broad categories: “ex parte in-court testimony or its functional equivalent .. extrajudicial statements contained in formalized testimonial materials ... [and] statements that were made under circumstances which would lead an objective witness reasonably to believe that the statement would be available for use at a later trial.” 541 U.S. at 51-52, 124 S.Ct. at 1364 (citations omitted). In cases involving police interaction, the Supreme Court has since emphasized the “primary purpose” of a statement,' concluding that testimonial statements are essentially trial-testimony substitutes relevant to proving facts for prosecution, while nontestimonial statements are those made to police to assist in an ongoing emergency. See Ohio v. Clark, - U.S. -, 135 S.Ct. 2173, 2183, 192 L.Ed.2d 306 (2015); Michigan v. Bryant, 562 U.S. 344, 358-59, 131 S.Ct. 1143, 1155, 179 L.Ed.2d 93 (2011); Davis, 547 U.S. at 822, 126 S.Ct. at 2273-74.
Andersen relies on Minnesota cases involving scientific reports made during the course of criminal investigations to support his argument that the report was prepared for trial. Each differs materially from this .case.
*443In Caulfield, the Minnesota Supreme Court determined that a Bureau of Criminal Apprehension laboratory report identifying cocaine was testimonial evidence that was admitted in violation of the appellant’s confrontation right. 722 N.W.2d at 306-07, 310. The court recognized that the “critical determinative factor” in determining testimonial nature is whether a statement was prepared for litigation. Id. at 309. It concluded that the report was prepared for litigation, bearing “characteristics of each of the three generic [Crawford] descriptions” because the report was akin to testimony, prepared at police request to aid in a prosecution, and offered into evidence to prove an element of the charged crime. Id.
In State v. Weaver, we held that laboratory results obtained during the course of an autopsy were testimonial. 733 N.W.2d 793, 799-800 (Minn.App. 2007), review denied (Minn. Sept. 18, 2007). .The following facts were determinative: the test results were obtained at the medical examiner’s request during an autopsy that occurred during a homicide investigation; the doctor relied on the results in reaching a conclusion on the cause of death; the underlying information was relayed to the jury “in lieu of testimony” at the trial; blood samples were sent to the laboratory after the medical examiner “preliminarily determined that arson had occurred and after appellant had been arrested;” and the technician performing the tests “would have known that the medical examiner’s office was a medical-legal operation.” Id.
And in State v. Johnson, we determined that the district court plainly erred by allowing the state to present an autopsy report through a doctor who was not one of the medical examiners who performed the autopsy. 756 N.W.2d 883, 892 (Minn. App. 2008), review denied (Minn. Dec. 23, 2008). We rejected the idea that the medical examiner’s statutory autopsy duties were “sufficiently independent” of a criminal investigation to render the autopsy report nontestimonial. Id. at 889-90. We pointed out that “Johnson was arrested at the scene. And the autopsy was not performed until approximately 33 hours after death, by which time a homicide investigation presumably had begun.” Id. at 890.
Relying on those cases, Andersen emphasizes that he was arrested later in the same morning that A.A. went to receive the treatment that prompted Dr. Binder’s report. But Andersen mistakenly asserts that Dr. Binder did not create his report until that night, long aftér the state’s investigation was “well underway” and after “authorities were, gathering evidence to support the [sjtate’s prosecution.” In fact, while the record indicates that Dr. Binder’s report was drafted at 10:34 without any reference to a.m. or p.m., the report also indicates that the specified time is represented in military style, which would designate 22:34, not 10:34, if the report were drafted at night.
The state makes its own mistake about the timing. It asserts that police received the call reporting A.A.’s assault at approximately 11:00 a.m. on July 8. But the testimony the state relies on establishes only that an officer was on duty at 11:00 a.m. when he received a report of a domestic assault, and it was the prosecutor’s question that contained that date and time, not the witness’s answer. And the state’s assumption that Dr. Binder completed his report before police were called is also contradicted by Eernisse’s testimony:
I informed the patient: that I was suspicious for that based on her clinical presentation and the x-ray findings and I told her that while we waited for police to come to do her report that the radiologist would also be revieunng the x-ray apd confirming my findings as well.
*444(Emphasis added.) This testimony does not precisely pinpoint the time of the relevant events, but it suggests that the police received the assault report contemporaneously with Dr. Binder’s reviewing A.A.’s x-rays and writing his findings in the radiology report.
Even if we reasonably assume that police began their investigation before Dr. Binder completed his report, Caulfield, Weaver, and Johnson do not lead us to conclude that his report was testimonial. The “statements” in those cases are distinguished by their timing and nature from Dr. Binder’s report in this case.
The timing of the “statements” is most readily distinguished. In Caulfield, police arrested the defendant, field-tested the seized substance twice, and then sent the substance to the BCA, which only then produced the challenged report. 722 N.W.2d at 307. In Weaver, the homicide investigation was well underway before the relevant testing began: “[T]he blood samples were sent to the laboratory after police and the medical examiner had preliminarily determined that arson had occurred and after appellant had been arrested as a suspect.” 733 N.W.2d at 800 (emphasis added). We observed also “that the laboratory technician would have known that the test results might have legal implications.” Id. Similarly in Johnson, we said that “Johnson was arrested at the scene. And the autopsy was not performed until approximately 33 hours after death, by which time a homicide investigation presumably had begun.” 756 N.W.2d at 890. In each of these cases, the “statements” were made long after an investigation had begun and after the defendant was arrested. Here, the timing of the police investigation and Dr. Binder’s report were essentially simultaneous, and police had not yet arrested Andersen or charged him with any crime.
More important is the material difference in the nature of the challenged “statements.” Caulfield’s BCA drug-analysis report was both offered at trial to prove an element and “prepared at'the request of the Rochester police for the prosecution of Caulfield.” 722 N.W.2d at 309. It “was clearly prepared for litigation” because police had seized the substance as part of a criminal investigation, it was sent to the BCA after police had already discovered its incriminating nature, and the report was created after Caulfield was arrested. Id. In Weaver, the medical examiner knew that a toxicology request included ehain-of-possession notes, he knew the medical examiner’s office was a “medical-legal” operation, and he knew that the results might have legal implications. 733 N.W.2d at 800. And in Johnson, it was “apparent from the police arrival on the scene” that the victim’s death was a homicide, and Johnson had been arrested on the scene. 756 N.W.2d at 890. The testimonial nature of a statement as to the cause of death during a homicide investigation is self-evident. And we observed that the “medical examiner’s statutory duties are not sufficiently independent of a police investigation to make an autopsy report non-testimonial.” Id. at 890; see also Minn. Stat. § 390.11, subd. 9 (2016) (“The coroner or medical examiner shall deliver to the county attorney copies of reports or other information created by the coroner’s or medical examiner’s office in any cases of a potential criminal ñatee.”).
Here, by contrast, Dr. Binder’s diagnostic report was standard practice for providing treatment, not for providing evidence in litigation. We hold that the report was not testimonial. Its admission into evidence therefore did not implicate Andersen’s constitutional right to confront witnesses.
*445DECISION
The district court did not abuse its discretion by admitting the testimony of AA’s relationship with Andersen because its probative value was not substantially outweighed by its potential for unfair prejudice. And the district court did not violate Andersen’s confrontation right by admitting Dr. Binder’s report into evidence because the report does not constitute a testimonial statement.
Affirmed.