¶1 Following trial, a jury found Curtis Pouncy to be a sexually violent predator. Pouncy seeks a new trial, claiming that the trial court erred when it failed to instruct the jury on the definition of “personality disorder” and when it allowed the State to impeach Pouncy’s expert using judicial findings about the expert set forth in a previous, unrelated trial. We hold that the trial court erred by refusing to instruct on the definition of “personality disorder” and that the error was not harmless. Accordingly, a new trial is required. While this issue is dispositive, we *386also address the impeachment evidence issue because it is one that is likely to arise on retrial. We agree with the Court of Appeals and hold that the trial court erred when it allowed the impeachment evidence.

FACTS AND PROCEDURAL HISTORY

¶2 Curtis Pouncy has a lengthy incarceration history for multiple sexual assaults. In April 2003, the State filed a petition seeking to have Pouncy committed as a sexually violent predator (SVP) pursuant to chapter 71.09 RCW. In 2006, a jury determined that Pouncy met the definition of an SVP and his commitment was ordered.

¶3 To establish that an individual is an SVP, the State must prove that the individual is one who “has been convicted of or charged with a crime of sexual violence and who suffers from a mental abnormality or personality disorder which makes the person likely to engage in predatory acts of sexual violence if not confined in a secure facility.” RCW 71.09.020(18) (Laws of 2009, ch. 409, § l)1 (emphasis added).

¶4 At trial, both the State and Pouncy offered testimony from forensic psychologists. The State’s expert, Dr. Richard Packer, opined that Pouncy suffered from an antisocial personality disorder and a mental abnormality. Pouncy’s expert, Dr. Richard Wollert, disputed the conclusion that Pouncy suffered from either a personality disorder or a mental abnormality.

¶5 On cross-examination, the State vigorously attacked Dr. Wollert’s credibility. As part of its cross-examination, the State introduced as an exhibit a trial court opinion from an unrelated proceeding finding that Dr. Wollert’s methodologies did not enjoy general acceptance in the community of mental health professionals. Pouncy’s counsel objected to the evidence on the basis of foundation, but the objection was overruled.

*387¶6 The line of questioning began with the State asking Dr. Wollert about what other mental health professionals thought of his work. The questioning then moved to discussion of the previous trial court decision:

Q. [PROSECUTION] You testified in a case of In Re: Robinson, right?

A. [DR. WOLLERT] Yes.

Q. Do you recall that case?

A. It’s quite some time ago. I recall — I recall portions of it, but certainly not all of it.

Q. In that case you testified about the Null hypothesis testing approach that you used, right?

A. Yes.

Q. And I asked you about this in your deposition, right?

A. Meaning what?

Q. I asked you about the Robinson case in your deposition?

A. Yes, you asked me about the Robinson case, yes.

Q. And when you say this was some time ago, the findings and conclusions came out in March of 2006, right?

A. Right. The testimony was some time before that.

Q. And in the Robinson case — I’m going to hand you what’s been marked Exhibit 156. I want you to look at finding of fact number 19, which is on page four.

A. Yes.

Q. And it states, Dr. Wollert’s methods of assessing the impact of age on recidivism are generally not accepted in the—

[DEFENSE COUNSEL] Objection, Your Honor, foundation.

THE COURT: On that basis the objection is overruled.

Q. [PROSECUTION] In the community of mental health professionals who evaluate and assess persons in SVP matters. This includes his use of Bayes theorem and Null hypothesis, right?[2]

*388A. Yes, that’s what the judge signed.

Q. And that’s the finding of fact in this case, that your methodologies are not generally accepted in the scientific community, right?

A. That is what the judge signed.

Q. And when I asked you about that in your deposition, you said, geez, I didn’t even know anything about that, right?

A. Right. I had not received a copy of that, so that it was new information to me, yes.

Verbatim Report of Proceedings (VRP) (Oct. 10, 2006) at 159-61. The next day, the State again referred to the Yakima judge’s3 findings on continued cross-examination. Later that day, the State sought to introduce credibility findings from another unrelated proceeding about a different witness. The trial court refused to allow the evidence on the ground that it was not relevant and noted that it would have sustained a relevancy objection about the judicial findings regarding Dr. Wollert had one been made.4 The next day, the trial court apparently decided of its own volition to instruct the jury that the “question of weight to be given to the testimony of any witness is for this jury to decide, based on all evidence introduced in this case. The jury is not to consider the findings of fact [regarding Dr. Wollert] apparently entered in the prior case. Evidence of such finding is stricken from the record.” VRP (Oct. 12, 2006) at 4. However, for reasons that are not clear from the record, the trial court did not so instruct the jury despite its stated intention to do so. Nor did Pouncy ask for a curative instruction.

¶7 Pouncy proposed an instruction defining “personality disorder”:

A Personality Disorder is an enduring pattern of inner experience and behavior that deviates markedly from the *389expectations of the individual’s culture, is pervasive and inflexible, has an onset in adolescence or early adulthood, is stable over time, and leads to distress or impairment.

Clerk’s Papers (CP) at 931. The trial court refused to give this instruction, and no instruction otherwise defined “personality disorder.” The jury found that Pouncy met the criteria for an SVP, but the verdict did not specify whether the jury believed Pouncy suffered from a mental abnormality or a personality disorder, or both.

¶8 Pouncy appealed. The Court of Appeals held that the impeachment evidence was improperly admitted and ordered a new trial. In re Det. of Pouncy, 144 Wn. App. 609, 622-27, 184 P.3d 651 (2008). The Court of Appeals also held that the trial court properly refused Pouncy’s proposed jury instruction defining “personality disorder,” relying on In re Detention of Twining, 77 Wn. App. 882, 894 P.2d 1331 (1995). Pouncy, 144 Wn. App. at 620-21. The State petitioned for review by this court, which we granted at 165 Wn.2d 1007, 198 P.3d 512 (2008). We also granted review on an issue raised in Pouncy’s answer to the State’s petition— whether the trial court erred when it refused to instruct the jury on the definition of “personality disorder.”5

ANALYSIS

I. Personality Disorder Instruction

¶9 In order to prove that Pouncy is an SVP, the State was required to show that Pouncy suffers from either a mental abnormality or a personality disorder. Former RCW 71.09-.020(16) (2006). The jury was instructed as to this requirement. CP at 991. The jury was further instructed on the definition of “mental abnormality,” which is defined by statute. CP at 992. At the time of Pouncy’s trial, “personal*390ity disorder” was not defined by statute.6 Pouncy’s counsel proposed an instruction defining “personality disorder,” but the trial court refused the instruction, to which Pouncy took exception. VRP (Oct. 12, 2006) at 38-39, 43.

¶10 The Court of Appeals rejected Pouncy’s claim of error regarding the missing definitional instruction, noting that the same argument had been made and rejected in Twining. Pouncy, 144 Wn. App. at 620-21. There, Division Three of the Court of Appeals held that a definitional instruction was not required because “personality disorder” was not defined in the SVP statute. Twining, 77 Wn. App. at 896. Pouncy argues Twining was wrongly decided.

¶11 A trial court’s refusal to send a proposed instruction to the jury is a discretionary decision reviewed for abuse of discretion. Seattle W. Indus., Inc. v. David A. Mowat Co., 110 Wn.2d 1, 9, 750 P.2d 245 (1988). “Trial courts must define technical words and expressions used in jury instructions, but need not define words and expressions that are of ordinary understanding or self-explanatory.” State v. Brown, 132 Wn.2d 529, 611-12, 940 P.2d 546 (1997). Whether a word is technical in nature is a question within the discretion of the trial court. State v. Guloy, 104 Wn.2d 412, 417, 705 P.2d 1182 (1985). Just because a word or phrase is included in a statute does not mean it is a technical term. State v. Scott, 110 Wn.2d 682, 691-92, 757 P.2d 492 (1988).

¶12 Twining rejected the contention that “personality disorder” is a technical term in need of definition for the jury, reasoning that “personality disorder” is not a statuto*391rily defined term with a specific legal meaning and each party was allowed to argue its understanding of personality disorder. Twining, 77 Wn. App. at 895-96. In light oí Brown and Scott, this analysis is too simplistic. A word’s designation as technical does not hinge on whether it is statutorily defined, Scott, 110 Wn.2d at 691-92, but rather on whether “it has a meaning that differs from common usage.” Brown, 132 Wn.2d at 611.

¶13 We have previously noted that “the term ‘personality disorder’ has a well-accepted psychological meaning” according to the Diagnostic and Statistical Manual of Mental Disorders (DSM).7 In re Pers. Restraint of Young, 122 Wn.2d 1, 50, 857 P.2d 989 (1993). The phrase “personality disorder” is not one in common usage and is beyond the experience of the average juror. It is a term of art under the DSM that requires definition to ensure jurors are not “forced to find a common denominator among each member’s individual understanding” of the term. State v. Allen, 101 Wn.2d 355, 362, 678 P.2d 798 (1984). We hold that Twining was wrongly decided insofar as it held that “personality disorder” is not a term in need of definition because it is not defined in the SVP statute.

¶14 The failure to instruct on a term’s definition may be harmless error. See, e.g., State v. Thompson, 47 Wn. App. 1, 9-10, 733 P.2d 584 (1987). “A harmless error is an error which is trivial, or formal, or merely academic, and was not prejudicial to the substantial rights of the party assigning it, and in no way affected the final outcome of the case.” State v. Britton, 27 Wn.2d 336, 341, 178 P.2d 341 (1947). Here, the term in question implicated an element of the State’s case. In order to prove Pouncy was an SVP, the State had to show he suffered from either a mental abnormality or a personality disorder. We have no way of know*392ing from the verdict whether the jury found that Pouncy was an SVP because he suffered from a mental abnormality or a personality disorder. CP at 1019. And, if the jury agreed Pouncy suffered from a personality disorder, we have no way of knowing what definition the jury used in reaching this conclusion. It is not sufficient that counsel were able to argue to the jury their respective understandings of the term based on expert testimony; lawyers have a hard enough time convincing jurors of facts without also having to convince them what the applicable law is. See Supp’l Br. of Resp’t at 19 (“ A jury should not have to obtain its instruction on the law from arguments of counsel.’ ” (quoting State v. Aumick, 126 Wn.2d 422, 431, 894 P.2d 1325 (1995))). We cannot say the failure to instruct on the definition of “personality disorder” in no way affected the final outcome of the case; accordingly, it was not harmless. A new trial is required.

II. Impeachment Evidence

¶15 While the failure to instruct the jury on the meaning of “personality disorder” is dispositive here, the propriety of the State’s impeachment evidence is an issue that is likely to arise on retrial. We therefore address it.8

¶16 The evidence admitted here was irrelevant, unduly prejudicial under ER 403, and impinged upon the jury’s role as the sole determiner of credibility. On this point, federal case law is persuasive because our ER 403 rule mirrors the Fed. R. Evid. 403 rule.9 One federal court aptly observed:

When a judge attacks a witness there is no effective defense. Peer review of such witnesses is different; if an expert does not *393act properly that expert ought to be attacked in the normal course of scientific debate — or in the case of a trial, with the opportunity for rehabilitation and explanation. To appropriately meet the evaluations of another judge would require the jury to delve deeply into the case that judge was trying. This enterprise is not appropriate under Rule 403.

In this case the cross examination of the witness was blistering and lasted more than a full trial day. An extensive deposition and exchange of reports provided ample fuel for the broad-based attack on the expert’s credibility. Defendants’ experts will continue the attack. They do not need another judge as an ally.

Blue Cross & Blue Shield of N.J., Inc. v. Philip Morris, Inc., 141 F. Supp. 2d 320, 325 (E.D.N.Y. 2001). Here, the trial court allowed as evidence another judge’s findings as to whether Pouncy’s expert’s methodologies were sufficient under the Frye test.10 However, the State here did not ask for a Frye hearing, so the judge in this case had no opportunity to determine whether Dr. Wollert’s methodologies enjoyed general acceptance among mental health professionals. In this context, admitting the evidence amounted to nothing more than allowing a judge to impeach a witness. The trial court correctly noted such evidence is irrelevant. We further note that is it highly prejudicial.

¶17 In addition, the evidence constituted inadmissible hearsay. There is no question that the Yakima judge’s findings were out-of-court statements used to prove the truth of the matter asserted — that Dr. Wollert’s methodologies lacked acceptance by his peers. Although the judge’s findings of fact were contained in a certified public document, the document is not included under the public records exception to the hearsay rule. In order to qualify for the exception, the proffered document “ ‘must contain facts and not conclusions involving the exercise of judgment or *394discretion or the expression of opinion.’ ” State v. Monson, 113 Wn.2d 833, 839, 784 P.2d 485 (1989) (quoting Steel v. Johnson, 9 Wn.2d 347, 358, 115 P.2d 145 (1941)). Despite their designation as findings of fact, the Yakima judge’s credibility determinations about Dr. Wollert involved a discretionary exercise not contemplated by the public records exception to the hearsay rule. Because the Yakima judge’s findings are hearsay, the trial court’s ruling to admit the evidence cannot be sustained. “[Allowing courts discretion to admit hearsay evidence not in compliance with the rule creates a ‘catch-all’ exception to the hearsay rule” akin to the federal catchall provision, which we have expressly rejected. State v. Neal, 144 Wn.2d 600, 610, 30 P.3d 1255 (2001).11

¶18 In sum, the impeachment evidence the State introduced here was inadmissible on several grounds. The trial court’s ruling admitting it was untenable and an abuse of discretion. Wash. State Physicians Ins. Exch. & Ass’n v. Fisons Corp., 122 Wn.2d 299, 339, 858 P.2d 1054 (1993) (noting that a trial court necessarily abuses its discretion when it bases a ruling on an erroneous view of the law). On retrial, the impeachment evidence should not be admitted.

CONCLUSION

¶19 The trial court erred when it failed to instruct the jury on the definition of “personality disorder.” The error was not harmless, and a new trial is required. On retrial, *395evidence of the findings of a judge in an unrelated trial should not be admitted, as such impeachment evidence is irrelevant and unduly prejudicial, and constitutes inadmissible hearsay. We affirm the Court of Appeals, reverse Pouncy’s SVP determination, and remand for a new trial.

¶20 (concurring/dissenting) — I agree with the majority that the trial court erred when it admitted as evidence another judge’s findings as to whether Curtis Pouncy’s expert’s methodologies were generally accepted in the scientific community under the Frye standard. Frye v. United States, 54 App. D.C. 46, 293 F. 1013 (1923). As a result, I agree that this case must be reversed and remanded for a new trial.

¶21 There is no need for this court to additionally decide whether the trial court erred when it declined to define the term “personality disorder” in a jury instruction. I would not reach this issue. Moreover, I disagree with the majority’s analysis. I would hold that the trial court committed no error by declining to define this term in a jury instruction.

Discussion

¶22 Because the trial court’s error in admitting another judge’s ruling provides an adequate basis for reversal, there is no need to reach the jury instruction issue. There is also no need to reach the issue in this case because the legislature has amended RCW 71.09.020 to include a definition of “personality disorder.” Compare RCW 71.09.020(9), with former RCW 71.09.020 (2006). On remand, the trial court will have to provide new jury instructions to a new jury and will necessarily provide the new statutory definition of this term in a jury instruction.

¶23 Even if reaching the jury instruction issue were necessary to resolve this case, the technical term rule does *396not provide an additional basis for reversal. Prior to the statutory amendment, there was no need to include a definition of “personality disorder” in the jury instructions. First, Pouncy’s trial was a mental commitment proceeding, not a criminal trial. The jury in this case was not evaluating the elements of a crime, and so did not require instruction on definitions from the criminal code to satisfy constitutional requirements. Cf. State v. Allen, 101 Wn.2d 355, 358, 678 P.2d 798 (1984) (“trial court must instruct the jury on every element of the crime”). Second, at the time of Pouncy’s trial, the legislature had not yet selected a legal meaning for the term “personality disorder.” Former RCW 71.09.020 (2006). Given this context, it would not have been appropriate to require the trial court to provide a legal definition for “personality disorder” in a jury instruction.

¶24 Third, under our precedent, the technical term rule does not apply in this case. Only where there is a technical legal meaning, such as one provided by a statute, case law, or a pattern jury instruction must the trial court provide the legal meaning in a jury instruction, per the technical term rule. State v. Brown, 132 Wn.2d 529, 611, 940 P.2d 546 (1997). This understanding of the technical term rule has been reiterated by this court in multiple cases. For example, in Allen, 101 Wn.2d at 361, we held the technical term rule required the trial court to provide an instructional definition where a statutory definition provides a “specific legal definition” distinct from common usage. Id. (“intent” and “knowledge,” “because they have been statutorily defined, have specific legal definitions aside from any common understanding or dictionary definitions which might be ascribed to them”). In State v. Scott, 110 Wn.2d 682, 691-92, 757 P.2d 492 (1988), we restated the technical term rule in reverse; where there is no conflict between the statutory definition and the plain meaning of a term, an instructional definition is not required.

¶25 In Brown, we reaffirmed the principle that the technical term rule requires a trial court to provide a definitional instruction only where a term has a legal meaning distinct from common usage.

*397A term is “technical” when it has a meaning that differs from common usage. The phrases here are not defined by statute. No appellate court has defined them and no pattern jury instructions address them. We conclude the phrases are expressions of common understanding to be given meaning from their common usage.

Brown, 132 Wn.2d at 611 (footnote omitted) (citing Scott, 110 Wn.2d at 694 (Utter, J., concurring)).12

¶26 Thus, the technical term rule requires a trial court to provide a definitional instruction where an authoritative legal source provides a technical definition, i.e., one that conflicts with common usage. However, where there is an accepted definition within the scientific community, but no authoritative legal definition, the trial court is not required to define the term in a jury instruction.

¶27 Nonetheless, the majority understandably and correctly expresses concern that the term “personality disorder” is outside the experience of the average juror. Knowledge of this term was integral to the jury’s determination. At some point in the proceedings, then, the jury needed to be advised of the definition of “personality disorder” to avoid deliberating in ignorance.

¶28 But this concern does not mean that the trial court needed to inform the jury of the legal meaning because there was no legal meaning. There was no authoritative legal source for the trial court to draw a definition from: no statute, no pattern jury instruction, and no appellate court case. There was, however, a well-accepted definition of long standing from the American Psychiatric Association, as we mentioned in In re Personal Restraint of Young, 122 Wn.2d 1, 49-50, 857 P.2d 989 (1993) (citing Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 335-58 (3d ed. rev. 1987)).13 This being the case, the trial *398court appropriately allowed expert testimony from this professional medical community regarding the medical definition. Particularly where there is no legal definition to be had, it is a matter for the jury to select what evidence it will accept, and what it will not, from the testimony presented.

¶29 The trial court was correct to follow the approach of In re Detention of Twining, 77 Wn. App. 882, 895-96, 894 P.2d 1331 (1995), where the court allowed expert testimony on the definition of “personality disorder” and declined to define this term in the jury instructions. The trial court did not err by allowing expert testimony and refusing to give Pouncy’s requested jury instruction.

Conclusion

¶30 I would reverse because the trial court erred when it admitted another judge’s ruling related to the reliability of Mr. Pouncy’s only expert, and I would decline to reach the jury instruction issue. In any event, the technical term rule does not provide an independent basis for reversal where the term at issue had no specific legal definition and the trial court properly allowed expert testimony regarding the term’s well-accepted definition within the professional medical community.

Reconsideration denied May 5, 2010.