In re the Detention of Thorell

72 P.3d 708, 149 Wash. 2d 724, 2003 Wash. LEXIS 463

• Court: Washington Supreme Court
• Decided: July 10, 2003
• Docket: Nos. 69574-1; 70488-1; 70967-0; 71059-7; 71315-4; 70405-8
• Status: Published

Opinions

Johnson, J.

This case involves three issues regarding commitment proceedings under chapter 71.09 RCW, the sexually violent predator act (SVPA). The first issue is whether the fact finder must determine that the person facing commitment as a sexually violent predator (SVP) has serious difficulty controlling behavior and, if so, whether this determination must be a separate finding based upon a jury instruction. The second issue is whether the fact finder must consider less restrictive alternatives (LRAs) to total confinement during the initial commitment hearing and the scope of the LRAs to be considered. The third and final issue is whether actuarial instruments may be admitted to aid in the prediction of future dangerousness and, if these instruments are admitted, whether Frye¹ or Evidence Rule (ER) 702 is the appropriate test of their reliability.

After careful analysis, we resolve these issues as follows. First, we hold the fact finder need not make a separate finding that a person committed under chapter 71.09 RCW as an SVP has serious difficulty controlling behavior. Second, we reexamine our prior cases and hold that LRAs need not be considered at initial commitment and may be considered for the first time during the annual LRA review without violating principles of equal protection. We further hold that the LRAs to be considered are those provided for under the SVPA. Finally, we hold actuarial instruments may be admitted if they satisfy the requirements of ER 702.

Analysis

This case involves the consolidated petitions of six persons who have been civilly committed under chapter 71.09 RCW. Each petitioner presents a unique claim, but the three issues identified above exist to varying degrees for each of them. We begin by analyzing controlling precedent before applying the law to the facts of each petitioner’s case.

I. Proof of Serious Difficulty Controlling Behavior

The first issue raised by the petitioners concerns whether the fact finder must determine that the person facing commitment as an SVP has serious difficulty controlling behavior under the United States Supreme Court’s recent decision in Kansas v. Crane, 534 U.S. 407, 414, 122 S. Ct. 867, 871, 151 L. Ed. 2d 856 (2002). If this determination is required, we must also decide whether the determination must be a separate finding based upon a jury instruction. To resolve this issue, we begin by examining the progression of the United States Supreme Court’s treatment of SVP commitment.

Freedom from bodily restraint has always been at the core of the liberty interest protected by the due process clause of the fourteenth amendment to the United States Constitution. Commitment for any reason constitutes a significant deprivation of liberty triggering due process protection. Foucha v. Louisiana, 504 U.S. 71, 80, 112 S. Ct. 1780, 118 L. Ed. 2d 437 (1992). However, the civil commitment of an SVP satisfies due process if the SVP statute couples proof of dangerousness with proof of an additional element, such as “mental illness,” because the additional element limits confinement to those who suffer from an impairment “rendering them dangerous beyond their control.” Kansas v. Hendricks, 521 U.S. 346, 358, 117 S. Ct. 2072, 138 L. Ed. 2d 501 (1997) (holding civil commitment of SVPs constitutional under the Kansas sexually violent predator act (Kansas SVPA)).

The United States Supreme Court recently clarified Hendricks’ mental illness element in SVP commitment proceedings as one requiring “proof of serious difficulty in controlling behavior.” Crane, 534 U.S. at 413. The Court advised this proof must be considered in the context of the features of the case, such as the nature of the psychiatric diagnosis and the severity of the mental abnormality. Crane, 534 U.S. at 413. Although the United States Supreme Court agreed with the state of Kansas that the State need not prove a total lack of control, the Court rejected Kansas’ position that it need not prove any lack of control. Put another way, the Court recognized that “lack of control” could not be demonstrated with precision, but required proof “sufficient to distinguish the dangerous sexual offender whose serious mental illness, abnormality, or disorder subjects him [or her] to civil commitment from the dangerous but typical recidivist convicted in an ordinary criminal case.” Crane, 534 U.S. at 413. Significantly, the United States Supreme Court specified neither the quantum of proof that it required nor the means of establishing the proof. Instead, the Court’s constitutional guidance in this area was given in the context of respondent Crane’s specific circumstances. Crane, 534 U.S. at 414.

Although the Crane opinion did not discuss the facts of the underlying SVP commitment in detail, the specific circumstances of the case illuminate the Court’s reasoning. Crane was convicted of lewd and lascivious behavior and pleaded guilty to aggravated sexual battery for two incidents. In re Care & Treatment of Crane, 269 Kan. 578, 586, 7 P.3d 285 (2000). Prior to his parole, Kansas entered a petition to have Crane declared an SVP under the Kansas SVPA. Crane, 269 Kan. at 579.

The Kansas SVPA definition of mental illness is substantially the same as Washington’s and permits the civil detention of a person convicted of any of several enumerated sexual offenses if it is proved beyond a reasonable doubt the individual suffers from a “mental abnormality”—a disorder affecting “emotional or volitional capacity which predisposes the person to commit sexually violent offenses”—or a “personality disorder,” either of which “makes the person likely to engage in repeat acts of sexual violence.” Kan. Stat. Ann. § 59-29a02(a), (b) (Supp. 2002).²

Several psychologists examined Crane to determine whether he had a mental abnormality or personality disorder under the Kansas SVPA. The psychologists determined Crane suffered from exhibitionism and antisocial personality disorder. Crane, 269 Kan. at 579. The opinion of one of the examining psychologists was that the two disorders in combination placed Crane’s condition within the range of disorders covered by the Kansas SVPA. However, the State’s experts concluded that Crane’s mental disorders did not impair his volitional control to the degree he could not control his dangerous behavior. Crane, 269 Kan. at 581.

Crane moved for summary judgment, arguing that Hendricks required the State to prove he was unable to control his violent behavior. Alternatively, Crane sought a jury instruction that would have required the jury to decide whether his diagnosed disorders had the effect of impairing his volitional control to such an extent that he was unable to control his behavior. Crane, 269 Kan. at 581. The trial court denied Crane’s summary judgment motion and did not instruct the jury that it must find Crane lacked volitional control.³ Crane, 269 Kan. at 581. The jury found, beyond a reasonable doubt, that Crane was an SVP, and he was civilly committed. Crane, 269 Kan. at 586.

The Kansas Supreme Court reversed Crane’s civil commitment, holding the trial court erred by refusing to instruct the jury it must find Crane’s mental disability so impaired his volitional control that he was unable to restrain his dangerous sexual behavior. Crane, 269 Kan. at 586. The court reasoned that because Crane had not been diagnosed with a mental abnormality, but a personality disorder, which “by definition does not include a volitional impairment,” his civil commitment violated due process. Crane, 269 Kan. at 586. The court held that, in the absence of a diagnosed volitional impairment, to satisfy its burden of proof the State must prove, not merely a likelihood that Crane would engage in repeat acts of sexual violence, but also an inability to control violent behavior. Crane, 269 Kan. at 585-86. Because the sufficiency of the State’s proof was a question for the jury, and the jury had not been instructed to make a finding that Crane could not control his behavior, the court reversed Crane’s civil commitment and remanded for a new trial. Crane, 269 Kan. at 593.

The Kansas Supreme Court based its holding solely on its reading of Hendricks.4 The court reasoned, “the plain language of the majority opinion in Hendricks” mandated a finding of volitional impairment. Crane, 269 Kan. at 586. The court supported its plain language argument by seizing on several Hendricks passages explicitly referring to volitional impairment, by relying on Hendricks’ pedophilia diagnosis, which arguably demonstrated Hendricks’ volitional impairment, and by referring to Hendricks’ concession that when stressed he could not control his urge to molest children. Crane, 269 Kan. at 581-82.

The United States Supreme Court rejected the Kansas Supreme Court’s bright-line analysis, but refused to abandon the requirement that states show the individuals they petition to commit in SVP proceedings have serious difficulty controlling sexually violent behavior. Crane, 534 U.S. at 414-15. The Court held that a person may not be committed as an SVP “without any lack-of-control determination.” Crane, 534 U.S. at 412. Thus, states must produce some proof that the individuals against whom petitions are brought have a serious lack of control over their behavior. Crane, 534 U.S. at 413. However, the Court refused to give the phrase “lack of control” a “particularly narrow or technical meaning,” rejecting the bright-line rules advocated by the parties in favor of a case specific analysis. Crane, 534 U.S. at 413.

The United States Supreme Court advanced two considerations in support of its case specific approach. First, states retain considerable leeway in defining the mental abnormalities and disorders that make an individual eligible for SVP commitment. Second, the science of psychiatry is ever-advancing and its “distinctions do not seek precisely to mirror those of the law.” Crane, 534 U.S. at 413. Significantly, just as it rejected the Kansas Supreme Court’s distinction between personality disorders and mental abnormalities, the United States Supreme Court explicitly declined to distinguish between volitional, emotional, or other cognitive factors that might impair an SVP’s ability to control violent behavior. Crane, 534 U.S. at 414.

We conclude that Crane requires a determination that a potential SVP has serious difficulty controlling dangerous, sexually predatory behavior, but does not require a separate finding to that effect. The United States Supreme Court did not impose a new element in SVP commitment proceedings when explaining its case specific approach. Although Crane held that “there must be proof of serious difficulty in controlling behavior,” the Court made this holding in response to whether a fact finder must find a total lack of control, not whether the proof requires a specific finding. Crane, 534 U.S. at 413. We therefore read Crane as consistent with Hendricks, which held that a lack of control determination may be included in the finding of mental abnormality. Crane, 534 U.S. at 421 (Scalia, J., dissenting). What is critical to both Hendricks and Crane is the existence of “some proof’ that the diagnosed mental abnormality has an impact on offenders’ ability to control their behavior. Crane requires linking an SVP’s serious difficulty in controlling behavior to a mental abnormality, which together with a history of sexually predatory behavior, gives rise to a finding of future dangerousness, justifies civil commitment, and sufficiently distinguishes the SVP from the dangerous but typical criminal recidivist. It is the finding of this link, rather than an independent determination, that establishes the serious lack of control and thus meets the constitutional requirements for SVP commitment under Hendricks and Crane. Then, if the existence of this link is challenged on appeal, this case specific approach requires the reviewing court to analyze the evidence and determine whether sufficient evidence exists to establish a serious lack of control, as we do below.

We base our conclusion on the Supreme Court’s lengthy discussion of the impracticability of giving “lack of control” a narrow or technical meaning, and the Court’s recognition of the need to proceed contextually. Crane, 534 U.S. at 412. Moreover, the Supreme Court’s vacation of the Kansas Supreme Court’s holding that Crane was entitled to a jury instruction and the Crane majority’s explicit preference for case specific analysis rather than bright-line rules militate against the conclusion that a separate jury instruction must be given.

We find further support for our reading of Crane in a recent opinion of the Arizona Supreme Court, In re Leon G., 204 Ariz. 15, 59 P.3d 779 (2002). In an earlier decision, the Arizona Supreme Court had held that Hendricks did not “impose ‘volitional impairment’ as a separate requirement for civil commitment statutes.” In re Leon G., 200 Ariz. 298, 301, 26 P.3d 481, 484 (2001). Then, after the United States Supreme Court issued Crane, it vacated the Arizona court’s 2001 decision for reconsideration in light of Crane. Glick v. Arizona, 535 U.S. 982, 122 S. Ct. 1535, 152 L. Ed. 2d 461 (2002).

On remand, the Arizona court determined Hendricks and Crane merely required that an SVP statute “narrow the class of persons subject to commitment to only those who have ‘serious difficulty in controlling’ their behavior.” Leon G., 59 P.3d at 783. The court held that Hendricks and Crane do not require explicit references to “control.” Leon G., 59 P.3d at 785. The Arizona court concluded, as we have here, that

Crane’s statement that a state must prove “serious difficulty in controlling behavior” does not require express statutory language, but rather reiterates the requirement that an SVP statute substantially and adequately narrows the class of individuals subject to involuntary civil commitment. Crane does not alter the Court’s analysis in Hendricks that focused on the link between proof of dangerousness and proof of mental abnormality in upholding the Kansas Act.

Leon G., 59 P.3d at 786 (citation omitted).

The Arizona court found that if the State establishes the required nexus between a person’s mental disorder and the person’s dangerousness, and proves that the disorder, rather than a voluntary decision, makes the person act in a certain manner, the State has shown that the person has “serious difficulty in controlling” his or her behavior. Leon G., 59 P.3d at 787. As did the Arizona court, we conclude that Hendricks and Crane do not mandate a specific jury instruction as long as the State demonstrates the cause and effect relationship between the alleged SVP’s mental disorder and a high probability the individual will commit future acts of violence. See Leon G., 59 P.3d at 787-88.

We find additional support for a case specific approach in related cases involving the Minnesota Supreme Court’s analysis of Hendricks and the Eighth Circuit Court of Appeals’ review of that analysis in light of Crane. Linehan v. Milczark, 315 F.3d 920, 925 (8th Cir. 2003) (Linehan V); In re Linehan, 594 N.W.2d 867 (Minn. 1999) (Linehan IV). Both cases confirm our interpretation that Hendricks and Crane focus on the link between the mental disorder and difficulty controlling behavior, not on a separate finding of lack of control.

In Linehan IV, the Minnesota Supreme Court analyzed the lack of control aspect of Hendricks because it had previously ruled the State was required to prove SVPs had evidenced “ ‘an utter lack of power to control [their] sexual impulses.’ ” Linehan IV, 594 N.W.2d at 869 (quoting State ex rel. Pearson v. Probate Court, 205 Minn. 545, 555, 287 N.W. 297 (1939)). The Minnesota Supreme Court considered whether Hendricks required a complete or a partial lack of volitional control. The court concluded Hendricks required “some lack of volitional control” to narrow the scope of civil commitment statutes. Linehan IV, 594 N.W.2d at 873. Utilizing a case specific approach, the court examined the underlying record for proof of lack of control and found Linehan’s commitment did not violate due process or require a separate finding by the jury. Linehan IV, 594 N.W.2d at 877-78.

Then, after a denial of certiorari, Linehan filed a petition for habeas corpus. The federal district court denied his petition and Linehan appealed to the Eighth Circuit. Linehan V, 315 F.3d at 921. He argued, inter alia, that his commitment was unconstitutional because the State did not provide separate proof of his volitional impairment at his commitment proceedings. Linehan V, 315 F.3d at 921. The Eighth Circuit closely examined the Minnesota Supreme Court’s earlier decision in light of Crane and held that Crane was consistent with Hendricks because it merely clarified “the level of volitional impairment constitutionally required for commitment.” Linehan V, 315 F.3d . Although the Eighth Circuit characterized the Minnesota Supreme Court in Linehan IV as requiring a “finding” of lack of control, the Eighth Circuit held the state court correctly applied Hendricks when it made that “finding” in relation to findings of past sexual violence and a mental disorder. Linehan V, 315 F.3d at 927. The connection between past sexually violent behavior and a mental abnormality results in a “likelihood of future sexually dangerous behavior,” and thus a lack of control. Linehan V, 315 F.3d at 927. Thus, the Eighth Circuit also emphasized the link of these factors when upholding Linehan’s commitment as constitutional. Accord In re Dutil, 437 Mass. 9, 11, 768 N.E.2d 1055 (2002) (construing Massachusetts statute requiring “a general lack of power to control. . . sexual impulses” to satisfy Crane without requiring a separate finding). We agree with the supreme courts of Arizona, Minnesota, and Massachusetts that a separate finding that the individual facing SVP commitment has serious difficulty controlling behavior is not required under the United States Constitution.

Dissenting in Crane, Justice Scalia and Justice Thomas (Hendricks’ author) complained Crane might necessitate a separate jury instruction for lack of control and illustrated the inherent constitutional difficulties should states attempt such an instruction. Crane, 534 F.3d at 423-24 (Scalia, J., dissenting). The petitioners, as well as the opinion of Chief Justice Alexander here, rely on Scalia’s characterization of the Crane majority, using it to argue the majority in Crane requires proof of lack of control as a separate element of an SVP commitment. However, if this analysis were correct, the United States Supreme Court would have gone beyond vacating the Kansas Supreme Court’s prior case and would have declared Kansas’ sexually violent predator commitment statute unconstitutional because it did not include this separate element. Crane therefore cannot be read in the manner advanced by the dissent here.

The petitioners ask us to consider that the United States Supreme Court recently vacated and remanded an unpublished decision of Minnesota’s Court of Appeals, which relied upon Linehan IV for its holding. In re Martinelli, 2000 Minn. App. LEXIS 973, 2000 WL 1285430 (Minn. Ct. App. Sept. 12, 2000), cert. granted & judgment vacated sub nom. Martinelli v. Minnesota, 534 U.S. 1160, 122 S. Ct. 1171, 152 L. Ed. 2d 114 (2002). We conclude the petitioners’ reliance on Martinelli is misplaced. In that case, the Minnesota Court of Appeals concluded that the State needed only to prove Martinelli was “unable to adequately control his sexual impulses.” In re Martinelli, 2000 WL 1285430, at *2. The United States Supreme Court vacated Martinelli for reconsideration in light of Crane.5

We interpret the United States Supreme Court’s vacation of Martinelli to turn upon the degree of difficulty a person facing SVP commitment must have controlling behavior, not as a mandate to impose a separate lack of control finding never mentioned in the Crane majority. If a separate finding of this kind were required, Linehan IV, which held lack of control was not a separate element and instead analyzed the record for sufficient evidence of Linehan’s inability to control his behavior, would have been erroneous under Crane.

The recent reconsideration of Martinelli, however, underscores the importance of establishing a connection between the mental disorder and difficulty controlling behavior. In re Martinelli, 649 N.W.2d 886 (Minn. Ct. App. 2002). On remand, the court rejected any semantic distinctions between “serious difficulty,” “difficult, if not impossible,” and “lack of adequate control,” emphasizing that Crane declined to give any particular narrow or technical meaning to “lack of control.” Martinelli, 649 N.W.2d at 890-91. The court on remand focused on the factual context and held that what was essential was expert testimony tying a lack of control to a diagnosed mental abnormality or personality disorder before civil commitment may occur. Martinelli, 649 N.W.2d at 890-91. Although the court held that a finding of lack of control was essential, the court did not focus on whether such a finding must be made separately. Martinelli, 649 N.W.2d at 890-91. Nonetheless, this case reinforces that control must be considered within the factual context of the mental abnormality.

Finally, we acknowledge that our conclusion that Crane does not mandate a separate jury finding is arguably inconsistent with those of other state supreme courts, such as the Missouri Supreme Court. The Missouri court recently mandated its jury instruction defining “mental abnormality” must be updated to “meanQ a congenital or acquired condition affecting the emotional or volitional capacity that predisposes the person to commit sexually violent offenses in a degree that causes the individual serious difficulty in controlling his behavior.” Thomas v. State, 74 S.W.3d 789, 792 (Mo. 2002); accord In re Commitment of W.Z., 173 N.J. 109, 801 A.2d 205 (2002) (remanding to trial court for finding that individual has serious difficulty controlling behavior). However, as Chief Justice Limbaugh’s dissent in Thomas noted, the Missouri court’s decision to change the definition of mental abnormality in the jury instruction means that the Missouri statute governing SVP commitment lacks a significant constitutional element because the “lack of control” language added to the jury instruction is not in the SVP statute.⁶ Thomas, 74 S.W.3d at 793 (Limbaugh, C.J., dissenting).

In our view, if Missouri’s definitional change were necessary, the United States Supreme Court would not have upheld the Kansas SVPAin Hendricks because the statutes in those cases did not include “lack of control” as a separate element. Nor did they require separate jury findings as to the degree of control the SVPs required. Finally, Missouri’s approach does not recognize the United States Supreme Court’s refusal to give the phrase “lack of control” a “particularly narrow or technical meaning” in favor of a case specific analysis. Crane, 534 U.S. at 413.

We conclude that the United States Supreme Court’s holding in Crane clarifies Hendricks’ mental illness element. Applying the United States Supreme Court’s reasoning to the Washington SVPA, we hold that proof that a person facing commitment under chapter 71.09 RCW lacks behavioral control is not a new element of the SVP commitment and a jury need not make a separate finding regarding “lack of control.” However, the jury’s finding that an SVP suffers from a mental illness, defined under our statute as a “mental abnormality” or “personality disorder,” coupled with the person’s history of sexually predatory acts, must support the conclusion that the person has serious difficulty controlling behavior, although this evidence need not rise to the level of demonstrating the person is completely unable to control his or her behavior.

We now turn to whether the jury instructions in the present cases meet the constitutional standard of sufficiently narrowing the class of persons subject to SVP civil commitment by requiring a link between the mental abnormality and a serious lack of ability to control sexually violent behavior, applying our above holding. In each of these six cases, the juries were given the standard “to commit” instruction, which reads in relevant part:

To find that the respondent is a sexually violent predator, each of the following elements must be proved beyond a reasonable doubt:

(1) That the respondent has been convicted of or charged with a crime of sexual violence; and

(2) That the respondent suffers from a mental abnormality or personality disorder; and

(3) That such mental abnormality or personality disorder makes the respondent likely to engage in predatory acts of sexual violence if not confined in a secure facility.[⁷]

Because the standard “to commit” instruction requires the fact finder to find a link between a mental abnormality and the likelihood of future acts of sexual violence if not confined in a secure facility, the instruction requires a fact finder to determine the person seriously lacks control of sexually violent behavior.

Again, we look to the Arizona Supreme Court’s recent opinion, Leon G., for guidance. Leon G., 59 P.3d at 786-87. Like Washington’s SVPA, Arizona’s SVP statute defines a sexually violent predator as someone who has committed a sexually violent offense and has “ ‘a mental disorder that makes the person likely to engage in acts of sexual violence.’ ” Leon G., 59 P.3d at 786 (emphasis omitted) (quoting Ariz. Rev. Stat. § 36-3701.7). The Arizona court held the requirement that the person’s mental disorder or abnormality makes the person likely to commit acts of sexual violence sufficiently required the fact finder to link the abnormality with a serious lack of volitional control, meeting the constitutional standard as articulated in Crane and Hendricks. Leon G., 59 P.3d at 787. Because the language of the instructions in the cases before us is substantially the same as that analyzed by the Arizona court, we find that court’s analysis persuasive. We hold the jury instructions given in the petitioners’ cases here were constitutionally adequate.

We do not reach Ross’s and Johnson’s arguments under the Washington Constitution because these arguments are directed to the standard of review we should employ had we concluded lack of control was a separate element of SVP commitment.

We now turn to the standard of review for SVP commitment proceedings. We first note that we have not previously addressed the appropriate standard of review to establish the quantum of evidence required for SVP commitment. The only other state supreme court to do so is the Wisconsin Supreme Court, which held the criminal standard of review should be used. In re Commitment of Curiel, 227 Wis. 2d 389, 597 N.W.2d 697, 709 (1999). Most important to the Wisconsin court’s decision was the fact that, while its SVP commitment statute is a civil proceeding, it shares many of the same procedural and constitutional features present in criminal prosecutions. Curiel, 597 N.W.2d at 709. Although we agree with the Wisconsin Supreme Court that the criminal standard is the correct standard to apply in SVP commitment proceedings, we do so on different grounds.

While we have not previously established the standard of review for sufficiency of the evidence in SVP proceedings, we have assessed probable cause in SVP commitments under a criminal standard. In re Det. of Petersen, 145 Wn.2d 789, 42 P.3d 952 (2002). The Court of Appeals also concluded the criminal standard should be applied to SVP commitments, although it did so under a different approach. In re Det. of Ross, 102 Wn. App. 108, 119, 6 P.3d 625 (2000). The Court of Appeals reasoned that because the legislature has adopted the beyond a reasonable doubt standard for commitments under the SVPA, the sufficiency of the evidence should be tested against this standard. Ross, 102 Wn. App. at 119. We agree.

We hold that the quantum of the evidence in SVP commitment hearings should be examined under a criminal standard. Although an SVP commitment is a civil proceeding and a finding of serious difficulty controlling behavior is not a separate element of the commitment proceeding, the legislature’s adoption of the reasonable doubt standard for SVP commitment should be applied to the lack of control determination as well. Under this approach, the evidence is sufficient if, when viewed in the light most favorable to the State, a rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt. State v. Randhawa, 133 Wn.2d 67, 73, 941 P.2d 661 (1997). We hold, when viewed in the light most favorable to the State, there must be sufficient evidence in the finding of mental illness to allow a rational trier of fact to conclude the person facing commitment has serious difficulty controlling behavior. We will examine each of the petitioners’ cases for sufficient evidence to support their commitment once we have completed our analysis of current precedent.

In conclusion, we hold the fact finder need not make a separate finding that a person committed under chapter 71.09 RCW, the SVPA, has serious difficulty controlling his or her behavior.⁸ We hold that although Crane did not establish a new element in SVP commitments, Crane did require SVP commitments to be supported by proof beyond a reasonable doubt of serious difficulty controlling behavior.

II. Less Restrictive Alternatives to Complete Confinement

The second issue presented by this case is whether the fact finder must consider less restrictive alternatives (LRAs) to total confinement during the initial commitment hearing and the scope of the LRAs to be considered. We begin our analysis by summarizing the prior treatment of LRAs in SVP commitment proceedings.

Our precedents have established that the source of an SVP’s right to consideration of LRAs at commitment flows from the equal protection clause of the fourteenth amendment to the United States Constitution. “Equal protection does not require that all persons be dealt with identically, but it does require that a distinction made have some relevance to the purpose for which the classification is made.” Baxstrom v. Herold, 383 U.S. 107, 111, 86 S. Ct. 760, 15 L. Ed. 2d 620 (1966) (holding “dangerousness” a rational distinction for purposes of determining the type of custodial care to be given). We relied upon Baxstrom to analyze whether the State could consider LRAs to total confinement under chapter 71.05 RCW, the involuntary commitment act, while denying consideration of LRAs under chapter 71.09 RCW, the SVPA. In re Pers. Restraint of Young, 122 Wn.2d 1, 857 P.2d 989 (1993). We held “equal protection requires the State to comply with provisions of RCW 71.05 as related to the consideration of less restrictive alternatives.” Young, 122 Wn.2d at 47. We remanded Young’s case for consideration of alternatives to confinement and concluded “[b]ecause the sex predator determination has already been made, the finder of fact need only consider if the less restrictive alternatives are appropriate” Young, 122 Wn.2d at 47 (emphasis added).

Our Young holding and subsequent interpretations by the Court of Appeals led to a number of legislative amendments to the SVPA. As originally enacted, the SVPA did not include any LRAs to confinement. See Laws of 1990, ch. 3, § 1001. In 1995, after our decision in Young, the legislature amended the SVPA to include LRAs (1995 amendments). However, under the statute, LRAs could be considered only after initial commitment. RCW 71.09.090-.098. The 1995 amendments also altered the definition of a “sexually violent predator” to mean a “person likely to engage in predatory acts of sexual violence if not confined in a secure facility.” Former RCW 71.09.020(1) (1995) (emphasis added).

The 1995 amendments spawned further equal protection challenges. Division One of the Court of Appeals examined the preclusion of LRAs at trial and held it was “rational to impose total confinement before undertaking consideration of whether a less restrictive treatment program is a viable option.” In re Det. of Brooks, 94 Wn. App. 716, 722, 973 P.2d 486 (1999), aff’d in part, rev’d in part on other grounds by 145 Wn.2d 275, 36 P.3d 1034 (2001). Division One found the greater dangerousness and different treatment requirements posed by SVPs provided a rational basis for allowing consideration of LRAs only after the SVP had been confined in a secure facility. Division Two of the Court of Appeals disagreed, but did not do so on equal protection grounds. Instead, Division Two construed the 1995 definitional change to entitle SVP detainees to present proof of LRAs at their commitment hearings as a defense to commitment under the SVPA’s definition of a “sexually violent predator.” Ross, 102 Wn. App. at 114.

In response to Ross, the legislature sought to prevent consideration of hypothetical LRAs as a defense to commitment and amended the SVPA in four ways. First, the legislature added a new section to chapter 71.09 RCW precluding the consideration of LRAs during the commitment hearing. RCW 71.09.015; Laws of 2001, ch. 286, § 1. Second, the legislature restricted the definition of LRAs to “court-ordered treatment in a setting less restrictive than total confinement which satisfies the conditions set forth in RCW 71.09.092 [the annual LRA petition provision].” Laws of 2001, ch. 286, § 4(7). Third, the legislature amended the procedures at commitment to preclude consideration of placement and treatment options other than those available on unconditional release and to prevent courts from ordering LRAs prior to the annual LRA review petition. Laws of 2001, ch. 286, § 7(1). Finally, in a later special session, the legislature added several additional definitions regarding secure facilities, which are discussed in detail below. Laws of 2001, 2d Spec. Sess., ch. 12, § 102.

In 2001, we addressed the issue of LRAs to resolve the disagreement between decisions of the Court of Appeals. We held that there was no rational basis to differentiate in the timing of consideration of LRAs between persons committed under chapter 71.05 RCW and the SVPA. In re Det. of Brooks, 145 Wn.2d 275, 36 P.3d 1034 (2001). We also took the unusual step of vacating the petitioners’ commitments and remanding for new commitment hearings where LRAs would be considered. Brooks, 145 Wn.2d at 292-93. Because LRAs were legislatively barred from consideration at commitment, we held that SVPs were entitled to unconditional release if the court or the jury determined an LRA was appropriate. Brooks, 145 Wn.2d at 292. We concluded this holding entitled the State to a jury instruction to the effect that the only remedy other than total confinement was unconditional release, but we prohibited the State from arguing no LRA could exist “because the court ha[d] no power to order such placement and treatment.” Brooks, 145 Wn.2d at 293.

In the present case, the petitioners again compare themselves to those entitled to consideration of LRAs under chapter 71.05 RCW. The petitioners argue that the statutory prohibition against considering LRAs during their commitment hearings under the SVPA violates their right to equal protection under the law and entitles them at least to new commitment hearings and, at most, to complete release.⁹ Gordon also argues that the consideration of LRAs must include the equivalent of LRAs under chapter 71.05 RCW, whether available under the SVPA or not.¹⁰ The State counters by urging us to abandon portions of Brooks in light of subsequent legislative amendments. Alternatively, the State argues the scope of review on remand is purely dispositional in nature and should be restricted to LRAs approved under the SVPA. Finally, both parties assert that the jury instructions regarding LRAs have proved difficult to implement. We acknowledge that lower courts have experienced difficulty implementing the holdings of our decision in Brooks and take this opportunity to reexamine our decision.

We initially turn to the scope of the LRAs to be considered. Under the 2001 amendments, the SVPA restricts the court and jury from considering LRAs other than those appropriate to an SVP. RCW 71.09.015. We must decide whether this violates equal protection principles. We use rational basis review to resolve equal protection claims involving the consideration of LRAs in SVP commitment proceedings. See In re Det. of Turay, 139 Wn.2d 379, 409-10, 986 P.2d 790 (1999), cert. denied, 531 U.S. 1125 (2001). Rational basis review requires we look for a legitimate governmental objective and a rational means of achieving it. See Turay, 139 Wn.2d at 410. We emphasize that rational basis review is highly deferential to the legislature. We uphold a legislative classification “ ‘unless it rests on grounds wholly irrelevant to the achievement of legitimate state objectives.’ ” Turay, 139 Wn.2d at 410 (quoting State v. Thorne, 129 Wn.2d 736, 771, 921 P.2d 514 (1996)). As the United States Supreme Court has held, “[a]s long as [the State] ‘rationally advances a reasonable and identifiable governmental objective, we must disregard’ the existence of alternative methods of furthering the objective ‘that we, as individuals, perhaps would have preferred.’ ” Heller v. Doe, 509 U.S. 312, 330, 113 S. Ct. 2637, 125 L. Ed. 2d 257 (1993) (quoting Schweiker v. Wilson, 450 U.S. 221, 235, 101 S. Ct. 1074, 67 L. Ed. 2d 186 (1981)). Even “ ‘rational speculation unsupported by evidence or empirical data’ ” provides a basis for upholding the classification under this level of review. Heller, 509 U.S. at 320 (quoting Fed. Communications Comm’n v. Beach Communications, 508 U.S. 307, 315, 113 S. Ct. 2096, 124 L. Ed. 2d 211 (1993)). The burden rests with the party challenging the classification to show it is purely arbitrary. Gossett v. Farmers Ins. Co., 133 Wn.2d 954, 979, 948 P.2d 1264 (1997).

We first note that SVPs have treatment issues distinct from those amenable to treatment under chapter 71.05 RCW. The legislature specifically found that SVPs do not have a mental disease or defect rendering them appropriate for treatment under chapter 71.05 RCW. RCW 71.09.010; Turay, 139 Wn.2d at 410-11. Second, the legislature found that, unlike persons eligible for civil commitment under chapter 71.05 RCW, SVPs are highly likely to engage in repeat acts of predatory sexual violence. As a result, SVP treatment needs are long term, and the treatment modalities for the SVP population are very different from the traditional treatment modalities for people better treated under chapter 71.05 RCW. Turay, 139 Wn.2d at 411. We conclude, as we have previously, that providing treatment specific to SVPs and protecting society from the heightened risk of sexual violence they present are legitimate state objectives. Turay, 139 Wn.2d at 410 (“ ‘it is irrefutable that the State has a compelling interest both in treating sex predators and protecting society from their actions’ ” (emphasis added) (quoting Young, 122 Wn.2d at 26)).

We now conclude that differentiating between LRAs for those involuntarily committed under chapter 71.05 RCW and the SVPA is a rational means to achieve these legitimate objectives. The LRAs under chapter 71.05 RCW do not consider the treatment issues unique to SVPs or the particular danger SVPs may present to the public. We conclude this constitutes a rational basis to restrict LRAs in SVP proceedings to those the legislature has found appropriate for the treatment of SVPs. We hold that this provision does not violate equal protection principles.

Our conclusion is bolstered by our determination that the current definition of a secure facility under the SVPA allows the functional equivalent of LRAs to be considered under the SVPA without reliance upon those provided for by chapter 71.05 RCW. A “sexually violent predator” is defined as “any person 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(16). “Secure facilities” are residential facilities that include security measures sufficient to protect the community. RCW 71.09.020(13). Secure facilities include “total confinement facilities, secure community transition facilities, and any residence used as a court-ordered placement under RCW 71.09.096 [the annual LRA petition provision].” RCW 71.09.020(13) (emphasis added). A “total confinement facility” is a “facility that provides supervision and sex offender treatment services in a total confinement setting.” RCW 71.09.020(17). A “secure community transition facility” is a “residential facility for persons civilly committed and conditionally released to a less restrictive alternative under [the SVPA].” RCW 71.09.020(14). Therefore, the current definition of a secure facility under the SVPA allows the functional equivalent of LRAs to be considered independently of chapter 71.05 RCW.

We next reexamine whether the SVPA, as amended in 2001 in response to Ross, violates equal protection because it prohibits consideration of LRAs at the initial commitment trial when chapter 71.05 RCW does allow consideration of LRAs at initial commitment.

Unlike someone committed under chapter 71.05 RCW, a sexually violent predator is someone who, by definition, is “likely to engage in predatory acts of sexual violence if not confined in a secure facility” and is not amenable to voluntary treatment on unconditional release. RCW 71.09.020(16) (emphasis added). The SVPA limits the fact finder at the initial hearing to the consideration of “placement conditions and voluntary treatment options that would exist for the person if unconditionally released from detention.” RCW 71.09.060(1). This means a fact finder may consider evidence that voluntary treatment on unconditional release is appropriate. Because this goes to whether the definition of SVP is met, the individual may bring this evidence in defense of commitment. The SVPA restricts the court, however, from ordering an LRA prior to a hearing under the annual LRA review provision, RCW 71.09.090, following initial commitment. RCW 71.09.060(4). Because of this restriction on the trial court, those who meet the statutory definition and are committed as SVPs are not entitled to consideration of LRAs until their first annual review. In contrast, those subject to commitment under chapter 71.05 RCW are entitled to consideration of whether they “should be detained for a period of either evaluation or treatment, or both, in an inpatient or a less restrictive setting” at their initial commitment hearing. RCW 71.05.020(4) (emphasis added).

Again, we engage in rational basis review. Taray, 139 Wn.2d at 410. As we explained above, providing appropriate treatment for SVPs and protecting society from the heightened risk of violence that they create when in the community are legitimate state objectives. At issue here is whether delaying the consideration of appropriate LRAs for SVPs until their first annual review rationally furthers these objectives.

The State asserts, as it did in Brooks, that treatment specific to an individual SVP requires an individual to spend a period of time in intensive inpatient treatment. In Brooks, we held that this rationale was arbitrary because there is a 45-day period of examination between a potential SVP’s probable cause hearing and commitment trial during which a person could be evaluated. Furthermore, the trial can be delayed upon a showing of good cause, giving the State more time for its evaluation. Brooks, 145 Wn.2d at 291-92.

The State emphasizes here, however, that the time for LRA evaluation must be spent in intensive inpatient treatment, which occurs only after commitment. Before commitment, the individuals are preoccupied with their legal challenges. Defense lawyers often direct their clients awaiting trial to limit their participation in treatment by not making any admissions or acknowledgments of past violent sexual acts or desires to commit such acts. Similarly, inmates in prison-based treatment programs while incarcerated are motivated not to discuss their offense cycle in order to avoid SVP commitment upon release. Successful treatment and evaluation for LRAs, however, depends on openly discussing and understanding one’s past violent sexual behavior and the desire to commit acts of sexual violence in the future. The State argues that it is only after commitment that SVPs tend to participate in treatment fully and the appropriateness of LRAs can accurately be evaluated.

Given the deference we must afford the legislative classification in this case, we conclude that the State’s reasons for delaying consideration of LRAs are rationally related to the treatment of sexually violent predators. The petitioners here have not met their burden to show delaying consideration of LRAs until the first annual review is purely arbitrary. We therefore do not reach issues of severability. We conclude there is no further need for explanatory jury instructions regarding the treatment of LRAs under the SVPA. To the extent our holding here conflicts with Brooks, that case is overruled.

III. Actuarial Instruments

The final issue presented by two of the petitioners is whether actuarial instruments may be admitted to aid in the prediction of future dangerousness and, if these instruments are admitted, the appropriate test of their reliability. In greatly simplified terms, there are two broad approaches to conducting risk assessments: clinical judgment or actuarial assessment. See generally Dennis M. Doren, Using Risk Assessment Instrumentation, in Evaluating Sex Offenders: A Manual for Civil Commitments and Beyond ch. 5, at 103 (2002). The clinical approach requires evaluators to consider a wide range of risk factors and then form an overall opinion concerning future dangerousness. The actuarial approach evaluates a limited set of predictors and then combines these variables using a predetermined, numerical weighting system to determine future risk of reoffense which may be adjusted (or not) by expert evaluators considering potentially important factors not included in the actuarial measure.

Actuarial approaches use statistical analysis to identify a number of risk factors that assist in the prediction of future dangerousness. Because actuarial models are based on statistical analysis of small sample sizes, they have a variety of potential predictive shortcomings. See generally Harry M. Hoberman, Dangerousness and Sex Offenders— Assessing Risk for Future Sex Offenses, in 2 The Sexual Predator (Anita Schlank ed., 2001). However, despite their potential statistical limitations, some experts have called for the complete rejection of clinical assessment in favor of purely actuarial assessment. See Vernon L. Quinsey et al., Fifteen Arguments Against Actuarial Risk Appraisal, in Violent Offenders: Appraising and Managing Risk 171 (1998).

The crux of the parties’ arguments is whether actuarial instruments should be viewed as novel scientific evidence. Strauss argues that actuarial instruments are novel scientific evidence.¹¹ If so, the standard we apply in Washington to determine whether evidence based on novel scientific procedures is admissible is that set forth in Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923). The Frye standard requires a trial court to determine whether a scientific theory or principle “ ‘has achieved general acceptance in the relevant scientific community’ ” before admitting it into evidence. Young, 122 Wn.2d at 56 (quoting State v. Martin, 101 Wn.2d 713, 719, 684 P.2d 651 (1984)). “ ‘[T]he core concern... is only whether the evidence being offered is based on established scientific methodology.’ ” Young, 122 Wn.2d at 56 (quoting State v. Cauthron, 120 Wn.2d 879, 889, 846 P.2d 502 (1993)).

The State contends actuarial instruments are not novel scientific evidence, so the trial court need not conduct a Frye hearing. The State asserts the methods and procedures used to construct actuarial instruments are well accepted in the scientific community and that Strauss’s arguments go to weight rather than admissibility. Finally, the State contends Strauss did not preserve this error through objection at trial. The Washington Association for the Treatment of Sexual Abusers (WATSA) joins the State, arguing that actuarial instruments “anchor” its risk assessments.¹² Moreover, WATSA argues the proof of the scientific community’s acceptance of actuarial instruments is that the failure to use such instruments constitutes an ethical violation for its members. If the State and WATSA’s contentions are correct, the use of these instruments as an aid to expert opinion testimony should be assessed under ER 702 and ER 703. State v. Baity, 140 Wn.2d 1, 10, 991 P.2d 1151 (2000).

We agree with the State and WATSA. On two prior occasions, we have accepted evidence of predictions of future dangerousness in SVP commitment hearings as based on established scientific methodology and declined to require a separate hearing under Frye. First, in Young, we examined clinical predictions of future dangerousness. We held an independent Frye determination unnecessary because predictions of dangerousness do not violate due process. Young, 122 Wn.2d at 56. We accepted the uncertainty surrounding psychiatric predictions and found them amenable to due process with procedural safeguards and a heavy burden of proof. In re Harris, 98 Wn.2d 276, 280-81, 654 P.2d 109 (1982). We also relied upon the fact that such predictions had previously been used to impose exceptional sentencing in sex offender cases. See State v. Pryor, 115 Wn.2d 445, 454, 799 P.2d 244 (1990). We viewed this as an analogous proposition already deemed admissible under Frye. Consequently, we held an independent evidentiary analysis unnecessary. Young, 122 Wn.2d at 56.

The second occasion in which we examined the scientific basis for predicting future dangerousness was In re Detention of Campbell, 139 Wn.2d 341, 355, 986 P.2d 771 (1999). Campbell challenged the prediction of future dangerousness both as inadmissible profile evidence and as inadmissible under FryeíER 702.¹³ The essence of Campbell’s argument was that the State’s expert testimony on dangerousness, which was based on clinical assessment, should have been excluded due to the superiority of actuarial assessment.¹⁴ We disagreed and rejected Campbell’s arguments under the precedents of our Young decision and the United States Supreme Court’s decision in Barefoot v. Estelle, 463 U.S. 880, 896-903,103 S. Ct. 3383, 77 L. Ed. 2d (1983) (holding predictions of future dangerousness should be admitted and evaluated by the fact finder). We concluded differences in opinion, which Campbell argued demonstrated the unreliability of the expert’s testimony, went to the weight of the evidence rather than to its admissibility. Campbell, 139 Wn.2d at 358. Based on our established precedent, we reiterate that the Frye standard has been satisfied by both clinical and actuarial determinations of future dangerousness.

Our conclusion is similar to that of other jurisdictions, which have upheld the use of actuarial assessments. Relying upon the thorough analysis of the New Jersey Court of Appeals, the New Jersey Supreme Court concluded “actuarial risk assessment instruments may be admissible in evidence in a civil commitment proceeding under the SVPA when such tools are used in the formation of the basis for a testifying expert’s opinion concerning the future dangerousness of a sex offender.” In re Commitment of R.S., 173 N.J. 134, 801 A.2d 219, 221 (2002); In re Registrant C.A., 146 N.J. 71, 679 A.2d 1153 (1996) (holding Frye inapplicable to actuarial device unique to New Jersey SVPA); accord State ex rel. Romley v. Fields, 201 Ariz. 321, 35 P.3d 82 (Ct. App. 2001) (holding Frye inapplicable to predictions of future dangerousness based upon actuarial instruments). We agree with the supreme courts of New Jersey and Arizona. We hold Strauss’s arguments go to the weight of the evidence rather than its admissibility and are to be assessed under ER 702 and ER 703.

The second challenge to the use of actuarial assessments is advanced by Johnson, who argues that actuarial instruments are inadmissible profiling tools because they do nothing more than assign values to characteristics in an effort to fit an individual within a profile.¹⁵ Johnson argues we have defined and excluded inadmissible profile testimony as evidence that merely identifies a person as a member of a group likely to commit a crime. State v. Petrich, 101 Wn.2d 566, 576, 683 P.2d 173 (1984); In re Marriage of Luckey, 73 Wn. App. 201, 208, 868 P.2d 189 (1994); State v. Braham, 67 Wn. App. 930, 936, 841 P.2d 785 (1992); State v. Maule, 35 Wn. App. 287, 293, 667 P.2d 96 (1983) ; State v. Claflin, 38 Wn. App. 847, 690 P.2d 1186 (1984) . Johnson contends actuarial evidence “unquestionably fits squarely within the definition of profile testimony”¹⁶ and is thus inadmissible.

The State refutes this argument with the admissible nature of expert testimony on dangerousness. The State argues if such testimony is admissible, then actuarial models, which it asserts are more reliable than clinical judgment, should be equally admissible. We are persuaded by the State’s arguments. The circumstances of the testimony at issue under our “profiling” jurisprudence are distinguishable from those here.

We have clearly rooted our rejection of profile testimony in ER 403, ER 702, and ER 703. Under ER 403, we have determined that profile testimony should be excluded because its “potential for prejudice is significant compared to its minimal probative value.” Petrich, 101 Wn.2d at 576; Braham, 67 Wn. App. at 939 (testimony on SVP “grooming” behaviors similar to conduct of defendant inadmissible under ER 403); Claflin, 38 Wn. App. at 852 (testimony that 43 percent of child molestation cases reported to have been committed by “father figures” inadmissible under ER 403).

Other profile testimony has been rejected under ER 702 and ER 703. In Marriage of Luckey, an expert testified that he had administered the Minnesota Multiphasic Personality Inventory to the father in a custody dispute and concluded that the father’s scaled scores matched the profiles of known child molesters. Marriage of Luckey, 73 Wn. App. at 204. The Court of Appeals affirmed the trial court’s exclusion of this evidence under ER 702 and ER 703. Marriage of Luckey, 73 Wn. App. at 204; Maule, 35 Wn. App. at 293 (holding ER 702 and ER 703 excluded testimony by expert that “the majority” of child sexual abuse cases involve “a male parent-figure”).

Testimony regarding the future dangerousness of SVPs, by its nature, is prejudicial. The purpose of the testimony, after all, is to assist the fact finder in determining whether the SVP is likely to commit future violent acts. The correct standard balances the potential for unfair prejudice against the evidence’s probative value. The probative value of this testimony is high and directly relevant to whether an individual should be committed as a sexually violent predator. Consequently, we have already rejected challenges to predictions of future dangerousness under ER 403. Young, 122 Wn.2d at 53. We concluded it was not reversible error for the court to admit evidence of prior criminal activity to predict future dangerousness because the likelihood of continued violence on the part of these individuals was central to the determination of whether they should be committed under the SVPA. Young, 122 Wn.2d at 53. Similarly, we have already accepted expert predictions of future violence “central to the ultimate question here: whether petitioners suffer from a mental abnormality or personality disorder.” Young, 122 Wn.2d at 58. We hold that actuarial assessments, which satisfy the requirements of ER 403, ER 702, and ER 703 are admissible and not profile evidence.

IV. Application to the Petitioners’ Cases

Having determined that Crane did not create a new element in SVP commitment proceedings, that the legislature may restrict consideration of LRAs to an SVP’s first annual review, and that actuarial models may be used to predict future dangerousness, we now address the sufficiency of the evidence and the consideration of LRAs under the facts of each petitioner’s case. In order to uphold the commitment of the individuals below, we must find that the juries at their respective commitment trials had sufficient evidence to find the following elements beyond a reasonable doubt:

(1) That the respondent has been convicted of or charged with a crime of sexual violence; and

(2) That the respondent suffers from a mental abnormality or personality disorder; and

(3) That such mental abnormality or personality disorder makes the respondent likely to engage in predatory acts of sexual violence if not confined in a secure facility.

As we explain above, as part of this review, we must determine that the mental abnormality or personality disorder, coupled with the person’s sexual offense history, supports the finding that the person has serious difficulty controlling his behavior beyond a reasonable doubt.

A. Thorell

We conclude, after reviewing the record of the commitment proceedings, that sufficient evidence existed at trial for the jury to find, beyond a reasonable doubt, that Thorell was unable to control his urge to molest children at the time of his commitment. Thorell has a lengthy history of child molestation, conceding at his commitment hearing that he suffers from pedophilia.

The State presented extensive evidence demonstrating that Thorell had serious difficulty controlling his behavior. For example, during his confinement, and continuing up to the time of his SVP hearing, Thorell continued to promote his sexual fantasies involving children by modifying children’s pictures to make pornography, writing pornographic stories featuring children, and concealing store advertisements featuring children (prohibited to him under the SVP treatment program). The State also presented evidence that during the same period, Thorell attempted to under-report his deviant fantasies to his treatment group “so that if I went to trial, why there would be a positive recommendation.” 4 Clerk’s Papers (CP) (Thorell) at 801.

The State’s evidence allowed the jury to conclude, beyond a reasonable doubt, that Thorell’s previous violent offenses, his mental disorder, and the resulting serious lack of control, led to the likelihood Thorell would engage in future sexually predatory acts if not confined to a secure facility. We therefore affirm Thorell’s commitment.

B. Ross

We conclude the evidence in Ross’s case, when viewed in the light most favorable to the State, is sufficient for the jury to find beyond a reasonable doubt that Ross had serious difficulty controlling his behavior and was thus likely to commit acts of predatory sexual violence if not confined to a secure facility.

Ross has a history of committing acts of sexual violence. In addition, the jury heard the testimony of one of the State’s experts, Dr. Eusanio, who diagnosed Ross with borderline personality disorder, antisocial personality disorder, narcissistic personality disorder, and alcohol dependence. CP (Ross) at 232-33. Dr. Eusanio testified that these affected Ross’s ability to control his behavior. During direct examination, Dr. Eusanio explained the behavioral results of having this cluster of personality disorders:

Q. Now, were these various diagnoses affecting Mr. Ross’s emotional or volitional capacity?

A. Yes.

Q. And would you just tell the jury now what I just asked you in terms of emotional and volitional capacity?

A. Well, when you have this stance to the world, this set of attitudes, you have difficulty looking at the law as applying to you. You don’t see how it protects other people. If you want something, you want it for yourself, and nothing will stand in your way. If you are told that something is illegal and you still have a strong compulsion to do it, for example, drugs or sexual offenses, you will continue to do it and use whatever resources you have to hide your behavior and to con, manipulate, get around a lot.

Q. These diagnoses and your impression therefrom, you believe they predispose him to the commission of criminal acts?

A. Yes, I do.

2 Report of Proceedings (RP) (Ross) at 234-35. Dr. Eusanio’s testimony provided the link between Ross’s history and personality disorder, which supports the conclusion that Ross had a serious lack of control.

Dr. Packard, another expert witness for the State, diagnosed Ross with antisocial personality disorder and paraphilia with an orientation toward rape. CP (Ross) at 809. Dr. Packard explained that a personality disorder “is an enduring pattern of thinking, feeling, and behaving that is relatively stable over time.” CP (Ross) at 810. Ross’s cluster of personality disorders gives rise to “general callousness to the rights of others, . . . disregard to the rights of others,.. . engaging in criminal activities, [and] . . . self-centeredness.” CP (Ross) at 817. Based on Ross’s history of violent sexual acts, his treatment file while incarcerated, and his personality disorder diagnosis, Dr. Packard testified that Ross is likely to commit violent and predatory sexual acts in the future if not confined to a secure facility. CP (Ross) at 824.

Ross’s own expert, Dr. Brown, raised the possibility Ross would not commit sexually violent acts in the future if he avoided drugs and alcohol and engaged in outpatient therapy. CP (Ross) at 1221. However, Dr. Brown also testified that his preference would be for state supervision of Ross upon release. CP (Ross) at 1221. Court ordered supervision is not available at the initial commitment trial. This means the jury had to decide whether to commit Ross or release him without any supervision. As we explain above, a person may defend against SVP commitment by bringing evidence that unsupervised release is appropriate because that goes to the definition of whether someone is an SVP. Someone subject to SVP commitment may not defend against commitment, however, by bringing evidence that LRAs are appropriate. Dr. Brown’s testimony thus provides no support for Ross’s argument that he does not meet the statutory definition of an SVP.

Finally, a diagnosis of a mental abnormality or personality disorder is not, in itself, sufficient evidence for a jury to find a serious lack of control. Such a diagnosis, however, when coupled with evidence of prior sexually violent behavior and testimony from mental health experts, which links these to a serious lack of control, is sufficient for a jury to find that the person presents a serious risk of future sexual violence and therefore meets the requirements of an SVP. We conclude the testimony of the State’s experts gave the jury sufficient evidence to commit Ross as an SVP, and affirm his commitment.

C. Gordon

Gordon’s primary objection to his commitment under the SVPA is that his schizophrenia could also qualify him for commitment under chapter 71.05 RCW.¹⁷ However, there is no talismanic significance to a particular diagnosis of mental illness. No technical diagnosis of a particular “mental abnormality” definitively renders an individual either an SVP or not. As explained above, it is a diagnosis of a mental abnormality, coupled with a history of sexual violence, which gives rise to a serious lack of control and creates the risk a person will likely commit acts of predatory sexual violence in the future.

The evidence before the jury at Gordon’s commitment trial allowed the jury to find that he meets the statutory criteria of an SVP. Gordon was convicted of rape in the second degree. At trial, defense and State experts generally agreed that Gordon’s diagnoses are schizophrenia, antisocial personality disorder, and polysubstance abuse disorder. We hold that merely because an individual suffers from a mental illness that may be treatable under chapter 71.05 RCW, the State is not barred from seeking to commit the individual under the SVPA in appropriate circumstances.

In Gordon’s case there was sufficient evidence at trial for the jury to find, beyond a reasonable doubt, that Gordon currently had serious difficulty controlling his dangerous sexual behavior. The State’s expert testified that Gordon heard voices and, at times, experienced irrational beliefs that were related to sexual issues. The State’s expert also testified mania was present because of Gordon’s schizophrenia and that this impaired Gordon’s ability to control impulsive behavior. In the expert’s opinion, Gordon’s mania was unresponsive to medication. The State presented evidence that Gordon “[a]bsolutely” had poor behavioral control at the time of his commitment hearing. 17 RP (Gordon) at 34. Finally, during the course of his confinement, Gordon told a Department of Corrections counselor, “I’m Mr. Rapo, I have raped 15 people, women, in the past. . . I’m going to rape again.” 9 RP (Gordon) at 138.

We hold the evidence was sufficient for the jury to find, beyond a reasonable doubt, that Gordon had serious difficulty controlling his behavior. We affirm Gordon’s commitment.

D. Bishop

The evidence of lack of control over behavior in Bishop’s case was sufficient for the jury to find, beyond a reasonable doubt, that Bishop had serious difficulty controlling his urge to molest children at the time of his commitment. Bishop is a diagnosed pedophile with a lengthy history of child molestation. Several days before the State sought to have him committed as an SVP, Bishop estimated he was likely to reoffend within nine months. The State also introduced evidence that during his confinement Bishop had continuously attempted to solicit sex from other inmates who fit his preferred molestation profile. One year before his scheduled release date, Bishop was caught soliciting a young, slim, mentally retarded inmate with childlike features. When questioned about the incident, Bishop admitted he had targeted the inmate because he was unable to acquire children while incarcerated. Finally, the State’s expert believed that Bishop demonstrated signs that his deviancy was escalating even while he was in therapy. We hold there was sufficient evidence for the jury to find, beyond a reasonable doubt, that Bishop had serious difficulty controlling his urge to molest children at the time of commitment.

LRAs have already been considered for Bishop during his annual review under RCW 71.09.070.¹⁸ In any event, Bishop’s argument is that Brooks requires the complete reversal of his commitment. In re Det. of Brooks, 145 Wn.2d 275, 36 P.3d 1034 (2001).Because we hold that LRAs may be considered for the first time at the first annual review and overrule Brooks on that issue, we reject this argument and affirm Bishop’s commitment.

E. Strauss

The evidence of lack of control over behavior in Strauss’s case was sufficient for the jury to find, beyond a reasonable doubt, that Strauss had serious difficulty controlling his urge to rape at the time of commitment. Strauss was diagnosed with paraphilia not otherwise specified, sexual sadism, and antisocial personality disorder. The State’s expert testified that the combination of these disorders was a “very, very lethal cocktail that will enable somebody to go out and carry on raping.” 11 RP (Strauss) at 196. The evidence proved that Strauss committed his last rape 39 days after his release, while still on parole from an earlier rape conviction. In Strauss’s deposition, taken a month before the commitment hearing, he testified that he did not remember the details of the rapes he had committed, was not certain why he had committed them, or what had drawn him to particular victims. Strauss testified the women he “victimized were just, happened to be in the wrong place at the wrong time,” which he further elaborated meant “[a]n unpopulated area, nobody around, they’re there by themselves.” 2 CP (Strauss) at 255. Although Strauss conceded he had a behavioral problem, he testified that he had not sought counseling to help him gain behavioral control. Finally, when asked if he had ever been able to effectively control his impulse to rape, Strauss answered that he was unable to respond “[bjasically because I don’t recall ever being in a position of [sic] on the verge of attacking someone and not attacking somebody.” 2 CP (Strauss) at 255-56. We hold the evidence was sufficient for the jury to find, beyond a reasonable doubt, that Strauss had serious difficulty controlling his urge to rape at the time of commitment.

F. Johnson

The only issue is whether sufficient evidence existed for the jury to conclude beyond a reasonable doubt that Johnson had serious difficulty controlling behavior at the time of his commitment.¹⁹ Johnson has an extensive history of sexual crimes, including two separate convictions for indecent liberties and a conviction for communication with a minor for immoral purposes. Based on a review of Johnson’s record, the State’s expert diagnosed Johnson as suffering from pedophilia, opposite sex, nonexclusive type, and antisocial personality disorder. The State’s expert testified the combination of pedophilia and antisocial personality disorder increased the likelihood that Johnson would be unable to control his behavior because he would act impulsively and would be less likely to learn from negative experience. Johnson testified at trial and by deposition. As a part of his deposition testimony, Johnson stated he believed the seven-year-old girl he had molested in 1984 had cried after the molestation, not because of his actions, but because she had not obeyed her mother. Johnson testified his practice when not incarcerated had been to wait near parks and playgrounds so that he would have the opportunity to prey on young girls and that if the opportunity arose, he would prey on “who [m] ever was there.” 4 RP (Johnson) at 90-92. Johnson also admitted during his redirect testimony that he had violated the conditions of his release for the 1984 molestation by engaging in contact with a two-year-old girl in 1992. The State’s expert concluded Johnson’s deposition “was remarkable for the evidence of a lack of insight, a lack of empathy and a lack of remorse for his conduct.” 5 RP (Johnson) at 260. The State’s expert considered Johnson’s lack of empathy particularly important because it reduced Johnson’s interest in controlling his impulses. We hold there was sufficient evidence for the jury to find, beyond a reasonable doubt, that Johnson had serious difficulty controlling his behavior, when committed.

Because we hold lack of control is not a separate element of the SVP commitment, we do not reach Johnson’s arguments under the Washington Constitution. We deny the State’s motions, passed to the merits, to strike Johnson’s harmless error argument and to be granted leave to file a supplemental brief addressing this argument. We affirm Johnson’s commitment.

Conclusion

To summarize our conclusions in these consolidated cases, we make three holdings. First, Crane does not require a separate jury finding that the SVP lacks control over behavior. Kansas v. Crane, 534 U.S. 407, 122 S. Ct. 867, 151 L. Ed. 2d 856 (2002). Instead, the jury’s finding of mental illness, coupled with a history of sexual violence, should be supported by sufficient evidence of serious difficulty controlling behavior. This evidence should be assessed under the criminal standard of review. Second, LRAs need not be considered at the initial hearing. Finally, actuarial assessments may be admitted to show the likelihood of reoffense. The appropriate tests of reliability under which such instruments should be admitted are ER 702 and ER 703.

We affirm the commitments of Thorell, Ross, Gordon, Bishop, Strauss, and Johnson because they were supported by sufficient evidence of serious difficulty controlling behavior, and there are no remaining issues regarding the consideration of LRAs in their cases.

Madsen, Ireland, Bridge, Chambers, and Owens, JJ., and Smith, J. Pro Tem., concur.

1 Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923).

2 The Washington statute, RCW 71.09.020, provides in relevant part: “ ‘Sexually violent predator’ means any person 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(16).

“ ‘Mental abnormality’ means a congenital or acquired condition affecting the emotional or volitional capacity which predisposes the person to the commission of criminal sexual acts in a degree constituting such person a menace to the health and safety of others.” RCW 71.09.020(8).

3 The trial court instructed the jury in part that, in order to establish that Crane is a sexually violent predator, the State must prove “ ‘[t]hat the respondent suffers from a mental abnormality or personality disorder which makes the respondent likely to engage in future predatory acts of sexual violence, if not confined in a secure facility.’ ” Br. of Pet’r Kansas to the United States Supreme Court, 2001WL 674238, at *7 (quoting Jury Instruction 7). The court further instructed the jury that “ ‘ “[mjental abnormality” means a congenital or acquired condition affecting the emotional or volitional capacity which predisposes a person to commit sexually violent offenses in a degree constituting such person a menace to the health and safety of others.’ ” Id. at *7 (quoting Jury Instruction 9). Finally, the jury was instructed that “ ‘ “Personality disorder” means a condition recognized by the American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (4th ed. 1994), and includes antisocial personality disorder.’ ” Id. at *8-9 (quoting Jury Instruction 9).

4 Hendricks, 521 U.S. 346.

5 Following the release of Linehan IV, the Minnesota Court of Appeals had concluded the State need prove only Martinelli was unable to “adequately control his sexual impulses,” while Crane requires the person committed have “serious difficulty controlling” behavior. In re Martinelli, 2000 Minn. App. LEXIS 973, 2000 WL 1285430, at *2 (Minn. Ct. App. Sept. 12, 2000).

6 See Mo. Rev. Stat. §§ 632.480-.495 (1999).

7 Clerk’s Papers (CP) at 1727 (Thorell Jury Instruction 7); 1 CP at 12 (Ross Jury Instruction 6); 3 CP at 305 (Gordon Jury Instruction 9); 3 CP at 867 (Bishop Jury Instruction 13); 7 CP at 1261 (Strauss Jury Instruction 8); 1 CP at 21 (Johnson Jury Instruction 7). We note that this language is consistent with the statutory definition of a sexually violent predator. RCW 71.09.020(16).

8 We note that in four of the cases before us, additional definitions on volitional control were given. “ “Volition’ is defined as an act of willing, choosing, or deciding; the power or capacity [or capability] of choosing.” The definition in Petitioners Thorell and Bishop’s commitments included the “capacity to choose. The definition in Petitioners Gordon and Strauss’s included “capability.” 7 CP at 1727 (Thorell Jury Instruction 7); 3 CP at 305 (Gordon Jury Instruction 9); 3 CP at 867 (Bishop Jury Instruction 13); 7 CP at 1261 (Strauss Jury Instruction 8). In the two cases where volition was not defined for the jury, “mental abnormality was further defined for the jury to mean a condition affecting “the emotional or volitional capacity.” 1 CP at 14 (Ross Jury Instruction 8); 1 CP at 22 (Johnson Jury Instruction 8). Although not constitutionally or statutorily required, including an instruction on the serious lack of volitional control in future SVP commitments will assist in appellate review and is therefore the better practice.

9 Petitioner Johnson concedes that he received a full consideration of LRAs at trial and does not assert this issue on appeal. Suppl. Br. of Pet’r Johnson in Supp. of Pet. for Review at 8.

10 Suppl. Br. of Pet’r Gordon at 12.

11 Br. of Pet’r Strauss at 1. Thorell also argued this issue to the Court of Appeals. Br. of Appellant Thorell at 2.

12 Br. of Amicus Curiae WATSA at 16.

13 Br. of Appellant Campbell at 69.

14 Reply Br. of Appellant Campbell at 26.

15 Suppl. Br. of Pet’r Johnson at 7.

16 Suppl. Br. of Pet’r Johnson at 8.

17 Suppl. Br. of Pet’r Gordon at 18.

18 Suppl. Br. of Resp’t King County Prosecutor (Thorell, Gordon, Strauss, and Bishop) app. C.

19 Suppl. Br. of Pet’r Johnson at 3.