STEPHENS, J.

¶ 1 This case involves a constitutional challenge to the 2005 amendments to the annual review process under Washington's sexually violent predator (SVP) statute, chapter 71.09 RCW. David McCuistion was indefinitely committed as an SVP in 2004. In 2006, the Pierce County Superior Court held a show cause hearing for the consolidated periods of 2004-2006. It concluded that McCuistion failed to establish probable cause to believe his condition had "so changed" under RCW 71.09.090 as to require a trial on his continued confinement. McCuistion challenges this provision, arguing that, as amended, it offends both due process and the separation of powers. We hold that the 2005 amendments, which limit the facts that can be considered to establish probable cause, violate substantive due process. Accordingly, we reverse and remand for a show cause hearing under the prior version of RCW 71.09.090.

FACTS AND PROCEDURAL HISTORY

¶ 2 McCuistion has a history of sex crimes reaching back to 1980, including a 1993 conviction for third degree rape. At a trial in 2003, he was found to meet the criteria for an SVP under chapter 71.09 RCW and involuntarily committed. Pursuant to RCW 71.09.070 and .090, the State reevaluated McCuistion's mental health in 2004 and 2005 to determine whether he continued to meet the SVP definition and thus remained subject to commitment. Each evaluation concluded that McCuistion continued to meet the definition. During his commitment, McCuistion has refused sexual deviancy treatment.

¶ 3 An annual review hearing for 2004 through 2006 was held on October 27, 2006. At the hearing, the State submitted evaluations from two psychologists supporting its position that McCuistion should remain committed. McCuistion retained his own expert, Dr. Lee Coleman, who criticized the State's evaluations and reported his professional opinion that McCuistion did not meet the definition of an SVP. McCuistion also presented several declarations from employees of the facility in which he was housed, which attested to his good behavior. He maintained that he had matured and lost his impulsiveness during his years of incarceration and argued that the court was required to consider the effect of his age on his risk of recidivism. The Pierce County Superior Court found that while McCuistion had behaved well in the facility, there was no evidence that his underlying condition had changed. Concluding that Dr. Coleman's report was "contrary to the conclusions reached by previous examiners" and was "essentially a re-argument" of the original commitment decision, the court determined that Dr. Coleman's disagreement "with past examiners and fact-finders does not, itself, make his opinion the correct one." Clerk's Papers at 585. The court therefore held that McCuistion continued to meet the definition of an SVP and ordered that he remain in custody.

¶ 4 McCuistion unsuccessfully sought review in the Court of Appeals. We granted McCuistion's petition for discretionary review at 164 Wn.2d 1029, 196 P.3d 138 (2008).

ANALYSIS

¶ 5 McCuistion argues that the 2005 amendments to RCW 71.09.090, the statutory provision governing annual reviews, violate due process and the separation of powers. Whether a statute violates the constitution is an issue of law reviewed de novo. In re Parentage of C.A.M.A., 154 Wn.2d 52, 57, 109 P.3d 405 (2005).

¶ 6 The SVP statute provides for the civil commitment of an individual who has been convicted of a crime of sexual violence and who suffers from a mental abnormality making him likely to reoffend. The legislature intended the SVP law to increase public safety in two ways: by incapacitating dangerous offenders and by treating them to eliminate the danger. RCW 71.09.010; In re Pers. Restraint of Young, 122 Wn.2d 1, 10, 857 P.2d 989 (1993). Since 1990, when the civil commitment scheme was first created, we have consistently upheld the legislature's approach to the difficult problem of recidivism among SVPs. See Young, 122 Wash.2d at 26, 857 P.2d 989 ("[T]here are no substantive constitutional impediments to the sexually violent predator scheme.").

¶ 7 At the same time, we have recognized that, because the SVP statute contemplates indefinite civil commitment, it presents substantive due process concerns.1 Id. at 25-42, 857 P.2d 989 (exploring several aspects of due process). Civil commitment impairs an individual's fundamental right to liberty and so is subject to strict scrutiny. Id. at 26, 857 P.2d 989. Strict scrutiny requires that any deprivation of a fundamental right be narrowly tailored to the State's compelling interests. Id. The United States Supreme Court and this court have held that the State has a compelling interest in civilly committing only those who are both mentally ill and dangerous to themselves or others. Foucha v. Louisiana, 504 U.S. 71, 75-76, 112 S.Ct. 1780, 118 L.Ed.2d 437 (1992); Young, 122 Wash.2d at 27, 857 P.2d 989. At the initial commitment proceeding, the SVP statute satisfies strict scrutiny by requiring the State to prove beyond a reasonable doubt that the individual suffers from a mental disorder and is dangerous. Young, 122 Wash.2d at 27-33, 857 P.2d 989.

¶ 8 Because commitment for SVPs is indefinite in nature, the due process requirement that an SVP be mentally ill and dangerous is ongoing. Foucha, 504 U.S. at 77, 112 S.Ct. 1780 ("[T]he acquittee may be held as long as he is both mentally ill and dangerous, but no longer."); accord O'Connor v. Donaldson, 422 U.S. 563, 574-76, 95 S.Ct. 2486, 45 L.Ed.2d 396 (1975). This is true because a law allowing the detention of individuals who are no longer mentally ill or dangerous would not be narrowly tailored to the State's compelling interests. Id. We have therefore attached constitutional significance to the SVP statute's annual review process, whereby the State must show that the SVP continues to meet the standard for commitment. Young, 122 Wash.2d at 39, 857 P.2d 989; see also In re Det. of Ambers, 160 Wn.2d 543, 553 n. 4, 559 n. 7, 158 P.3d 1144 (2007) (noting the constitutional implications of annual reviews); In re Det. of Elmore, 162 Wn.2d 27, 36 n. 8, 168 P.3d 1285 (2007) (same); In re Det. of Petersen v. State, 145 Wn.2d 789, 795-96, 42 P.3d 952 (2002) (attaching constitutional significance to the burden of proof in annual reviews). In Young, we recognized that meaningful annual review is central to the SVP statute's constitutionality. We held that SVP commitment is narrowly tailored in part because it is "not subject to any rigid time limit," but rather is "tailored to the nature and duration of the mental illness," and because "the Statute's release provisions provide the opportunity for periodic review of the committed individual's current mental condition and continuing dangerousness to the community." Young, 122 Wash.2d at 39, 857 P.2d 989.

¶ 9 Each year the State must evaluate committed individuals to determine if they continue to meet the definition of an SVP. RCW 71.09.070. If the State determines that an individual is no longer mentally ill or dangerous, the State authorizes him to petition for release or transfer to a less restrictive alternative confinement. RCW 71.09.090(1). The SVP may also petition the court without the State's authorization and obtain a show cause hearing. RCW 71.09.090(2). At this hearing, the State must present prima facie evidence that the petitioner still meets the SVP definition. RCW 71.09.090(2)(b). Then, if the committed person produces evidence establishing probable cause to believe that he has "so changed" as to no longer meet the definition of an SVP, he is entitled to a full hearing. RCW 71.09.090(2)(c)(ii). He makes this showing by presenting prima facie evidence to support the finding that he no longer meets the SVP definition, with the ultimate burden remaining on the State. Petersen, 145 Wash.2d at 796, 798-99, 42 P.3d 952. At the full hearing, which may be a jury trial, the State must once again prove beyond a reasonable doubt that the committed person is mentally ill and dangerous. RCW 71.09.090(3).

¶ 10 In Young, we approved of a prior version of the SVP statute's annual review provision, finding that it satisfied substantive due process minimums. Young, 122 Wash.2d at 39, 857 P.2d 989; former RCW 71.09.090 (1992). In 2005, the legislature amended the provision in response to a pair of Court of Appeals cases suggesting that demographic or scientific changes, such as a committed person's increase in age or new diagnostic procedures, could create a prima facie case that an individual had "so changed" that he no longer met the definition of an SVP.2 In re Det. of Young, 120 Wn.App. 753, 761-62, 86 P.3d 810 (2004) (age); In re Det. of Ward, 125 Wn.App. 381, 383, 104 P.3d 747 (2005) (diagnostic procedures); see also Ambers, 160 Wash.2d at 549-50, 158 P.3d 1144 (describing these cases as the impetus behind the 2005 amendments). The 2005 amendments clarified the legislature's intent that the phrase "so changed" refers specifically to physiological and treatment-based changes, not changes of the types recognized by the Court of Appeals:

RCW 71.09.090(4); Laws of 2005, ch. 344, §§ 1-2 (expressing the intent behind the addition of subsection .090(4) to the statute). Thus, the 2005 amendments altered the standard for obtaining a full hearing during the annual review process and effectively narrowed the universe of facts relevant to this standard.

¶ 11 In Ambers, we heard a constitutional challenge to the new amendments based on the State's position that the phrase "safe to be at large" in RCW 71.09.090(4)(b)(ii) heightened the SVP's burden for obtaining a new trial, and that changes must be treatment-based to entitle the SVP to a new trial. Ambers, 160 Wash.2d at 553, 558, 158 P.3d 1144. In dictum, we opined that, because of the State's continuing due process obligation to confine only dangerous persons, it "might be unconstitutional" to "require[] a more stringent standard at an annual review hearing than is required for initial commitment." Id. at 553 n. 4, 158 P.3d 1144 (citing O'Connor, 422 U.S. at 574-75, 95 S.Ct. 2486). However, we did not directly address the issue because we interpreted "safe to be at large" to be equivalent to the pre-2005 standard and determined that Ambers' change was in fact treatment-based. Id. at 557, 559 n. 7, 158 P.3d 1144.

¶ 12 In Elmore, we had to decide whether the 2005 amendments applied retroactively, which would have prevented Elmore's new commitment proceeding based solely on his change in age. Elmore, 162 Wash.2d at 32-33, 35, 168 P.3d 1285. We again avoided the issue of whether the 2005 amendments were constitutionally valid by interpreting them not to apply retroactively. Id. at 36 & n. 8, 168 P.3d 1285; see also In re Det. of Smith, 163 Wn.2d 699, 700-01, 184 P.3d 1261 (2008) (reversing per curiam on facts indistinguishable from Elmore).

¶ 13 Both of these cases foreshadowed that the 2005 amendments might overstep the bounds that substantive due process places upon the SVP civil commitment regime. The issue is squarely before us today, and we now elevate to a holding our dictum in Ambers. Because the 2005 amendments undermine meaningful annual review of SVP status consistent with minimum standards of substantive due process, the amendments are unconstitutional.3

¶ 14 Contrasting the initial commitment proceeding with the annual review proceeding reveals how the 2005 amendments have unconstitutionally eroded the substantive due process protections for civil commitment. At the initial commitment proceeding, the State must prove beyond a reasonable doubt that the person to be committed is an SVP, RCW 71.09.060(1). Specifically, the State must prove that the person, inter alia, suffers from a mental abnormality or personality disorder and is likely to engage in predatory acts of sexual violence if not confined in a secure facility. RCW 71.09.020(18). If the alleged SVP can show, for any reason, that there is reasonable doubt as to one of these criteria, the State fails to meet its burden and the person may not be involuntarily committed. RCW 71.09.060(1).

¶ 15 At the annual review, the State must make a prima facie showing that the committed individual still meets the definition of an SVP, which includes a showing that he is still dangerous. The pre-2005 statute allowed the SVP to counter this showing on any number of grounds. If the committed individual produced prima facie evidence showing, for any reason, that he was not an SVP, he was entitled to a jury trial at which the State would have to prove his continued SVP status beyond a reasonable doubt. Former RCW 71.09.090 (2001). (One such reason might have been that the committed individual no longer had a high risk of recidivism.) But under the 2005 amendments to RCW 71.09.090, the SVP may counter the State's showing on only two grounds: a permanent physiological change or a treatment-based mental change. Once the State offers prima facie evidence that the individual is still an SVP, the court has no authority to order a full hearing or trial absent evidence supporting one of these grounds. See RCW 71.09.090(4)(b). Furthermore, the SVP may not obtain a full hearing by showing a change in a single demographic factor, even if this change would alter the outcome of an SVP determination under a multifactor analysis that previously justified SVP commitment.4 RCW 71.09.090(4)(c).

¶ 16 The flaw in the 2005 amendments is that they separate the annual review inquiry from the ultimate constitutional standard under Foucha, 504 U.S. at 77, 112 S.Ct. 1780. The SVP statute upheld in Young was narrowly tailored to allow the detention only of currently mentally ill and dangerous individuals. Young, 122 Wash.2d at 39, 857 P.2d 989. By altering the annual review standard, the 2005 amendments authorize the State to detain individuals who are no longer mentally ill and dangerous. There is a multitude of ways in which a person might potentially cease to meet the definition of an SVP and, thus, cease to be detainable under the due process standard. Yet, only two of those ways are cognizable under the 2005 amendments to the annual review provisions.5 By artificially limiting the type of information that is relevant to continued SVP commitment, the 2005 amendments allow the detention of someone who is no longer mentally ill or dangerous, and therefore disrupt the narrow tailoring present in the preamendment SVP law. Because the SVP law, as amended, is not narrowly tailored to the State's compelling interests, we strike down the 2005 amendments as unconstitutional.6

¶ 17 The State urges that due process can be satisfied through a committed person's other avenues for relief, such as a personal restraint petition (PRP). This argument proves too much. If the 2005 amendments can be saved because an unconstitutionally detained person may file a PRP seeking relief, then any annual review provision, no matter how alienated from the requirements of substantive due process, would be beyond scrutiny. After all, an individual can always file a PRP to seek to prove his restraint is unlawful. See RAP 16.4. The inquiry must focus on the annual review procedure itself. Due process requires that this procedure be narrowly tailored to meet the State's interests, which means confining only those individuals who continue to be both mentally ill and dangerous.

¶ 18 Because of the amendments' severability clause, Laws of 2005, ch. 344, § 3, on remand the lower court should consider McCuistion's petition under the pre-2005 show cause standard—that is, RCW 71.09.090 without subsection .090(4).7 The relevant provision is RCW 71.09.090(2)(c), which reads:

Without subsection .090(4)'s elaboration upon the phrase "so changed," this provision allows for the full range of relevant evidence to prove that a committed person no longer meets the definition of an SVP. The evidence need not pertain to a permanent physiological or treatment-based change.8

CONCLUSION

¶ 19 The 2005 amendments to RCW 71.09.090 violate substantive due process and are invalid. Accordingly, we reverse and remand for a new show cause hearing under the pre-2005 show cause standard.

WE CONCUR: GERRY L. ALEXANDER, RICHARD B. SANDERS, TOM CHAMBERS, and JAMES M. JOHNSON, Justices.

SANDERS, J. (concurring).

¶ 20 I concur with the majority that the 2005 amendments are unconstitutional. I write separately, however, to emphasize that a person can also challenge admissibility of an expert's opinion if it is not based upon generally accepted scientific theory and methodology. See Frye v. United States, 54 App.D.C. 46, 293 F. 1013, 1014 (1923). The trial court must evaluate the State's expert testimony under the Frye test to assure that opinion evidence is based upon both valid and reliable scientific theory and methodology that are generally accepted in the scientific community. See ER 702, 703; State v. Gregory, 158 Wn.2d 759, 829, 147 P.3d 1201 (2006) (citing State v. Gore, 143 Wn.2d 288, 302, 21 P.3d 262 (2001)); State v. Cauthron, 120 Wn.2d 879, 886, 846 P.2d 502 (1993) (adopting the standard from Frye, 293 F. at 1014).

¶ 21 David McCuistion presented the declaration of Dr. Lee Coleman,1 which purported to systematically tear apart the theory and methodology utilized by the State's experts. If Dr. Coleman is correct, the opinions of the State's experts are inadmissible under Frye for failing to employ valid and reliable scientific theory and methodology generally accepted in the scientific community. Coleman declared: (a) the "scientific analysis" in the state expert opinions is nothing more than a recitation of McCuistion's criminal record and is devoid of any actual scientific theory or methodology, Clerk's Papers (CP) at 617-18 (Coleman Decl. ¶¶ 5-6); CP at 619 (Coleman Decl. ¶ 9); CP at 620-21 (Coleman Decl. ¶¶ 13-14); (b) Dr. Ronald Page's diagnosis is based upon McCuistion's "history of alcohol dependence and immature personality traits," even though "only a tiny percentage of individuals like this commit criminal sexual acts," CP at 617 (Coleman Decl. ¶ 5); (c) the risk assessment instruments utilized by Dr. Richard Packard are strictly experimental and not generally accepted for the purpose here, CP at 620 (Coleman Decl. ¶ 11); and (d) these conclusions are based upon circular reasoning—individuals have a mental abnormality because they commit sex offenses and commit sex offenses because of that abnormality, CP at 620-21 (Coleman Decl. ¶ 13); CP at 622 (Coleman Decl. ¶ 16).

¶ 22 Like Dr. Coleman, the Washington State Psychiatric Association in its amicus brief in In re Personal Restraint of Young, 122 Wn.2d 1, 857 P.2d 989 (1993), also took the position that a mental disorder cannot be diagnosed by looking only at past deviant criminal behavior, including sexual predation. Br. of Amicus Curiae Wash. State Psychiatric Ass'n (Br. of Amicus WSPA) at 3; In re Pers. Restraint of Young, No. 57837-1 (Wash.Sup.Ct. Sept. 23, 1991), reprinted in 1 Briefs 122 Wn.2d (1993).

¶ 23 The trial court must act as the gatekeeper of opinion testimony to assure such evidence is based upon scientific theory and methodology generally accepted in the scientific community. See, e.g., Gregory, 158 Wash.2d at 829, 147 P.3d 1201. But here the trial court failed to analyze testimony of the State's experts under the Frye test, even in light of Dr. Coleman's declaration. See CP at 585 (Order on Show Cause Hrg. ¶ 6).2

¶ 24 According to Coleman the proposition that a person can be diagnosed with a mental disorder that makes him likely to commit sex offenses, see RCW 71.09.020(18), ignores the fact that there is no such mental disorder currently recognized in the scientific community, nor is there any methodology by which a professional can causally link a disorder to an inability to control one's behavior. As Dr. Coleman testified: (a) the sexually violent predator classification has no connection to any condition recognized by the DSM-IV-TR (Am. Psychiatric Ass'n, Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR (4th rev. ed.2000)), which is generally accepted among mental health professionals in the United States, see CP at 622 (Coleman Decl. ¶¶ 15, 16); and (b) even where a mental disorder is identified, mental health professionals are unable to scientifically link that disorder to a deficit in volitional control—there is no generally accepted scientific method to do so, CP at 618-19 (Coleman Decl. ¶ 8).

¶ 25 Moreover, the Washington State Psychiatric Association's amicus brief in In re Detention of Young, supports Dr. Coleman's declaration: (a) "mental health professionals do not consider sex offenders to be suffering from a personality disorder which causes an individual to commit a sex offense," Br. of Amicus WSPA at 5; (b) "it is inappropriate to make broad generalizations as to a causal connection between sexual offenses in general and any particular psychiatric condition," id. at 4; and (c) many sex offenders do not commit crimes because of an illness and many individuals diagnosed with paraphilia— characterized by arousal to unconventional sexual objects, situations, or imagery—do not commit crimes, id.

¶ 26 To protect the public and address growing concern that sex offenders upon release were committing additional offenses, the legislature created a mechanism to involuntarily and indefinitely hold these persons in civil custody based upon an "expert" prediction they would be compelled to commit a sex crime due to a "mental abnormality or personality disorder." See RCW 71.09.020(18) (defining "sexually violent predator"),.060. However if the scientific community does not recognize such a condition, much less possess any methodology to identify individuals with such a condition, the statutory test cannot be met.

¶ 27 Without a scientifically recognized condition that compels a person to commit sex offenses, civil confinement also runs afoul of the constitution. Statistically, an individual who has previously committed a sex offense may be more likely to commit a future sex offense than an individual who has no record of sex offenses. So too an individual who has previously committed a theft or assault is more likely to do so again than a person who has not. However, individuals cannot be civilly confined for offenses they have not committed simply because they have a propensity to commit future offenses. Civil commitment is not permitted under the constitution unless the individual lacks control over his or her actions due to a mental disorder; it is not enough that a person has a propensity to reoffend based on his or her criminal record. See Kansas v. Crane, 534 U.S. 407, 412, 122 S.Ct. 867, 151 L.Ed.2d 856 (2002).

¶ 28 Where a person is deprived of his or her freedom based upon opinion testimony lacking scientific credibility, reliability, and accepted methodology, courts must step forward and announce with the courage of a small child that the Emperor wears no clothes.

APPENDIX A

DECLARATION OF LEE COLEMAN, MD

¶ 29 1. The following declaration is based on my review of records pertaining to David McCuistion and his status as a "sexually violent predator" (SVP) under the laws of the State of Washington.

¶ 30 2. I am familiar with the statutory requirements in Washington, having participated in several such cases in the past, and several dozen similar cases in California, where the laws are virtually identical.

¶ 31 3. The primary focus of this declaration concerns the question of whether there is current evidence that Mr. McCuistion continues to meet the statutory requirement for SVP status. Regardless of estimated risk for re-offending, such status requires that the individual be judged to suffer from a mental abnormality (defined as 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"). Furthermore, adjudication of this issue shall be based on the findings of a "professionally qualified person."

¶ 32 4. Given these statutory requirements, I have reviewed institutional records and professional evaluations of Mr. McCuistion and I have formed the opinion that his evaluators have not presented any evidence that such a mental abnormality exists, or has ever existed. Instead, they have relied on his past crimes: the required "mental abnormality" has been "determined" by simply summarizing his past behavior, and the "evidence" for the alleged disorder is a recitation of the details of his past behavior. As such, this information is not an expert finding worthy of credibility in the determination required by the Washington statute.

¶ 33 5. The above pattern may be seen from the very first SVP evaluation, performed in 1992 by Ronald Page. Nowhere does Dr. Page discuss any expert findings demonstrating that Mr. McCuistion has the required deficit of emotional or volitional capacity. Instead, he simply cites a history of alcohol dependence and immature personality traits. Such factors are clearly far too broad to fulfill the requirements of the SVP statute, since only a tiny percentage of individuals like this commit criminal sexual acts. Dr. Page offers nothing other than Mr. McCuistion's criminal record as justification for finding him to fit the SVP requirement. If this were sufficient, there would of course be no statutory requirement that a "professionally qualified person" offer opinions to the Court.

¶ 34 6. Dr. Savio Chan in 1995 likewise summarized Mr. McCuistion's criminal history, which now included a 1993 conviction for third degree rape and third degree assault. He described Mr. McCuistion's behavior during an interview, and reported that an MMPI was considered invalid. Based on this and nothing else, he wrote, "... the nature and pattern of his offenses, the number and age range of his victims, and his inability to control himself suggest the presence of sexual deviancy of paraphilia, NOS, and a personality disorder with antisocial features." In other words, Dr. Chan announces an alleged deficit of emotional or volitional capacity but demonstrates no methods or findings to support this conclusion. The crimes become the evidence, and this clearly violates the legal requirements for SVP status, which require a determination that a particular violent sexual offender has something more—the congenital or acquired condition—that is not always present in such offenders.

¶ 35 7. The next evaluation I have seen is from the End of Sentence Review Subcommittee in 1998. Once again, criminal history is reviewed, and the aforementioned conclusions of Drs. Page and Chan were cited. Dr. Chan was asked to do another evaluation and he concluded that "Paraphilia and Antisocial Personality Disorder constitute, congenital or acquired conditions affecting the emotional or volitional capacity which predisposes McCuistion to the commission of criminal sex acts in a degree constituting him a menace to the health and safety of others." This conclusion is completely without any basis because the label "paraphilia" does not represent an expert finding, but simply a restatement of the fact that Mr. McCuistion's crimes involved sexual behavior outside the norm: "para" meaning beyond the ordinary and "philia" meaning attraction to or preference for. Likewise for "antisocial personality disorder," obviously nothing more than a restatement of a pattern of criminal conduct. There is nothing expert here, as required by the statute.

¶ 36 8. Furthermore, the linking of any mental disorder, even if one had been demonstrated, to an alleged deficit in volitional control is something specifically addressed in the "American Psychiatric Association Statement on the Insanity Defense" (1982), which has concluded that any attempt by mental health professionals to make such a determination is beyond their special skills. What this means is that legislatures that have mandated that experts determine whether such a disorder is present in a particular offender have ignored consensus opinion from the relevant profession that such examinations do not exist.

¶ 37 9. Psychologist Richard Packard also performed an evaluation in 1988. He concluded that the correct diagnosis for Mr. McCuistion was "paraphilia, not otherwise specified," i.e., "sexual activity with non-consenting females." Can there be a more obvious example of "doublespeak"? Sexual activity with a non-consenting person is the crime of rape; this fact does not fulfill the legal requirement for SVP status, and the use of this meaningless "diagnosis" is obviously an attempt to give the appearance that the legal requirements have been met.

¶ 38 10. Dr. Packard also opined that it was "more likely" that outside of an institution Mr. McCuistion would re-offend, but based this opinion on the above "diagnosis" and "the absence of evidence to the contrary." Despite many well-intended efforts by countless professionals, the fact is that no inmate can produce such "evidence to the contrary." This is because no reliable association between participation in the sex-offender treatment programs and reduced recidivism has been shown. Kelley Blanchette of the Research Division of the Canadian Correctional Services has summarized current knowledge on this question: "Having recognized the potential shortcomings, a foregone response to the question `Does sex offender treatment work?' is this: We are still uncertain. There is disagreement even amongst the most prolific and knowledgeable researchers in the area." (Sex Offender Assessment, Treatment and Recidivism: A Literature Review Kelley Blanchette Research Division Correctional Services Canada August, 1996)

¶ 39 11. Dr. Packard also supported his conclusions with a "risk assessment" despite admitting that "Some of these risk assessment instruments are still considered as experimental and may have limited applicability specifically to Washington State sex offenders." In fact, all such instruments are strictly experimental, as has been repeatedly acknowledged by their most influential proponents, such as Karl Hansen of Public Safety and Emergency Preparedness Canada. Dr. Packard also argues that these instruments underestimate risk and mentions several factors that may do this. He fails to recognize other factors that may overestimate risk, such as inclusion of uncharged arrests in some studies, or unconvicted charges in others. He even includes an instrument called the "Sexual Violence Risk-20", which includes factors which have been mentioned in "clinical literature." In other words, it includes a catch-all of factors that may have been "excluded" in more controlled analyses, but are nonetheless to be included in Mr. McCuistion's risk analysis.

¶ 40 12. The most important factor, however, concerning these risk analyses is the fact that unless the individual has been found to have the requisite "congenital or acquired condition," risk analysis is moot.

¶ 41 13. The opinions of Dr. Gollogly, based on record review, are based on the same circular reasoning as the preceding. He summarizes Mr. McCuistion's criminal past, summarizes his sexual crimes with the label "Paraphilia NOS (Rape), and his other crimes with "Antisocial Personality Disorder," and then summarily states that "these diagnoses predisposes (sic) him to act out sexually deviant urges without regard to the rights of his victims ...' There is no mention of the legal requirement of a disorder of volitional control, undoubtedly because mental health professionals have no means to make such a determination. In his conclusion, Dr. Gollogly writes that "My opinion is based on my clinical judgment and the fact that instruments utilized in this assessment, commonly relied upon by experts in this field, indicate that Mr. McCuistion is at a high risk of sexually re-offending." These are flimsy supports indeed, given the consensus opinion in Psychiatry and Psychology that "clinical judgment" cannot determine "volitional control" and considering that SVP evaluators may rely on risk assessment procedures but SVP evaluators are but a tiny minority in the mental health community and the mainstream of specialists in the field of sex offender evaluation and treatment do not accept such procedures. (See Dangerous Sex Offenders. A Task Force Report of the American Psychiatric Association, American Psychiatric Press, Washington, D.C., 1999)

¶ 42 14. Finally, the October 31, 2004 Annual Review of Carole DeMarco demonstrates the identical problems as previous evaluations. The majority of her report is simply a repetition of previous records. In the section labeled "diagnosis and mental abnormalities, she writes that "Consistent diagnoses across time from a variety of doctoral-level clinicians are a strong indicator of diagnostic accuracy." This ignores, of course, the virtually universal practice in institutions whereby succeeding evaluators routinely apply whatever "diagnostic" labels were used by earlier evaluators. Consensus is not science and the fact that consensus is held out as "evidence" is revealing in itself.

¶ 43 15. Next, she writes that "Residents of the SCC usually suffer from a paraphilia." (!) Once again, citing the very fact that an institution that holds sex offenders would have a lot of sex offenders (the term "paraphilia" means nothing more) speaks volumes about the emptiness of this "diagnostic" process. Paraphilias, according to the DSM-IV-TR are said to be "chronic and lifelong" but there is nothing in the DSM that corresponds to the legal requirements for SVP status. The fact that evaluators almost universally try to boot-strap the DSM into these evaluations does nothing to alter this fact.

¶ 44 16. Then, in a classic example of circularity, Dr. DeMarco writes that "Individuals with mental disorders and/or personality disorders that impair the ability to inhibit impulses are often at a disadvantage to control his/her urges to engage in paraphilic behavior." This ignores the fact that the SVP law requires that a distinction be drawn, through the identification of a "congenital or acquired condition" between offenders who are simply repeat sex offenders and those who have an alleged disorder of "impaired volitional control. Because there is no such mental disorder, and the DSM includes nothing like this, and because mental health professionals cannot distinguish between those who commit such crimes as part of a "mental abnormality" and those who commit such crimes for other reasons, this type of circularity is offered instead.

¶ 45 17. Dr. DeMarco claims that "Paraphilia Not Otherwise Specified (Nonconsent) is an accepted diagnosis among practitioners knowledgeable about sexual offenders." I believe it would be more accurate to say that the only practitioners who use this label are those who perform SVP evaluations. But regardless of how many use it, the so-called "diagnosis" is obviously nothing more than doublespeak for the crime of rape. If this is the best the evaluators are capable of doing, when seeking the "congenital or acquired condition," surely it means that the entire evaluation process is a sham created to fulfill legal and legislative agendas.

¶ 46 18. Dr. DeMarco comes to the heart of the matter, writing that "While many, if not most, cases of rape are not considered a mental abnormality per se, there is a small proportion of rapists who repetitively engage in rape behavior and display significant arousal to nonconsenting sexual activity. Such individuals report having recurrent, repetitive, and compulsive urges and fantasies to commit rapes, and at times they find it difficult to resist such urges." If there were established procedures (the penile plethysmograph notwithstanding) for distinguishing between those who rape by choice and those driven to it by a mental disorder that impairs volitional control, wouldn't we expect this to be part of the DSM? The only mental health professionals who insist they can identify such a disorder are those who perform SVP evaluations.

¶ 47 19. Dr. DeMarco calls as well on actuarial risk assessment instruments, and I have commented on the unproven nature of these attempts, as well as the fact that they are moot if the legally required "congenital or acquired condition" cannot first be demonstrated. She also repeats the claim that such instruments underestimate risk, and I have already discussed this false idea as well.

¶ 48 20. As a "dynamic risk factor" (characteristics that could change over time, as opposed to the static nature of one's criminal past), she wrote that Mr. McCuistion "continues to associate with individuals who have an antisocial attitude and engage in a high level of fault finding with SCC rules and policies indicating an antisocial lifestyle." Given the universal recognition by SVP inmates that the evaluation and treatment program is based on a law that has no recognized basis in science or psychology, it is totally unacceptable to equate "fault finding with SCC rules and policies" with risk of sexual reoffending.

¶ 49 21. Finally, I believe it is important to recognize that it is impossible for any person, whether mental health professional or lay person, to become familiar with past behavior such as exhibited by Mr. McCuistion without feeling revulsion. This is precisely why it is crucial to recognize how easy it is for mental health evaluators, expected as they are to have something "special," to repackage this typical and perfectly expectable reaction as a "diagnosis," "risk assessment," etc. In reality, there is no recognized "congenital or acquired condition" that fits the desires of the legislature of Washington, California, or any of the other states with these laws. The findings of each of Mr. McCuistion's evaluators amount to no evidence whatever.

¶ 50 22. The above has been written by me in Berkeley, California, on September 21, 2005, and is offered under penalty of perjury.

OWENS, J. (dissenting).

¶ 51 A "`[s]exually violent predator'" (SVP) is a person who has already been found, beyond a reasonable doubt, to be "likely to engage in predatory acts of sexual violence if not confined in a secure facility." RCW 71.09.020(18). Rejecting the careful determination of the Washington legislature, the majority essentially creates a situation in which a single doctor, without ever examining the SVP in question, can put an SVP one step closer to release. The majority even allows this when there is no evidence that an SVP has changed in the years since his underlying commitment. This greatly undermines the legislature's careful decision that the 2005 amendments to section 71.09.090 of the sexually violent predators statute (the SVP statute), chapter 71.09 RCW, were necessary to promote the "`very long-term' needs of the sexually violent predator population for treatment and the equally long-term needs of the community for protection from these offenders." Laws of 2005, ch. 344, § 1.

¶ 52 The majority makes two glaring legal errors. First, it mischaracterizes a procedural due process issue as a substantive due process issue, inappropriately subjecting the law to heightened scrutiny. When we properly apply the procedural due process standard to the 2005 amendments to the SVP statute, due process is not violated. Second, even if substantive due process were warranted, the majority also fails to use the proper "strict scrutiny" test to determine the constitutionality of the 2005 amendments. Because the 2005 amendments to the SVP statute readily pass constitutional muster, I must respectfully dissent.

I. David McCuistion's Challenge to the 2005 Amendments to RCW 71.09.090 is Governed by the Standards of Procedural Due Process

¶ 53 McCuistion argues that the 2005 amendments to RCW 71.09.090 violate due process. The majority holds that these amendments violate substantive due process because they divorce the annual review inquiry from the standard set out in Foucha v. Louisiana, 504 U.S. 71, 112 S.Ct. 1780, 118 L.Ed.2d 437 (1992). The majority especially highlights the constitutional significance that we have attached to the SVP statute's annual review process. Majority at 1149-50.

¶ 54 An examination of substantive due process indicates that it is not the appropriate standard here. "The substantive component of the due process clause protects against certain government actions `regardless of the fairness of the procedures used to implement them.'" In re Pers. Restraint of Bush, 164 Wn.2d 697, 706, 193 P.3d 103 (2008) (quoting Daniels v. Williams, 474 U.S. 327, 331, 106 S.Ct. 662, 88 L.Ed.2d 662 (1986)); see Zinermon v. Burch, 494 U.S. 113, 125, 110 S.Ct. 975, 108 L.Ed.2d 100 (1990). Substantive due process, in essence, prevents the government from performing certain actions no matter how many procedural protections there are or how fair those procedural protections may be, unless the governmental action meets strict scrutiny. Examples of activities protected by substantive due process include consensual sex within one's home, Lawrence v. Texas, 539 U.S. 558, 578, 123 S.Ct. 2472, 156 L.Ed.2d 508 (2003), and the use of contraceptives, Griswold v. Connecticut, 381 U.S. 479, 485, 85 S.Ct. 1678, 14 L.Ed.2d 510 (1965). Foucha establishes that McCuistion has a substantive due process right to not be held unless he is currently mentally ill and dangerous. Foucha, 504 U.S. at 80, 112 S.Ct. 1780.

¶ 55 McCuistion does not contest the State's power to detain him under any circumstances; he argues only that the procedures used were inadequate. He does not argue that the State does not have the power to continue to detain him. In contrast, he merely asks for procedural safeguards at a particular hearing. This type of challenge is governed by principles of procedural due process, which require that the State must not deprive an individual of a protected liberty interest without appropriate procedural safeguards. Bush, 164 Wash.2d at 704, 193 P.3d 103. The majority implicitly acknowledges that this case concerns only what procedures are necessary to deprive an SVP of liberty, given that the remedy it provides is additional procedures. The majority mandates that a trial court consider additional types of evidence at the annual show cause hearing beyond merely (1) evidence of an identified physiological change and (2) evidence of a change in an SVP's mental condition brought about through positive response to continuing participation in treatment. See majority at 1148-49. Substantive due process analysis requires that there be no possible procedures sufficient to excuse a governmental action, yet the majority requires additional procedures to ensure that liberty is protected. For this reason, it is self-evident that this is not an issue of substantive due process. Since McCuistion asks for additional procedural safeguards at the annual review hearing, this is an issue of procedural due process.

II. The 2005 Amendments to RCW 71.09.090 Do Not Violate the Flexible Standard of Procedural Due Process

¶ 56 When reviewing claims of violations of due process, we must remember that "due process is a flexible concept." In re Det. of Stout, 159 Wn.2d 357, 370, 150 P.3d 86 (2007). Procedural due process requires, at minimum, notice and an opportunity to be heard, but its requirements vary depending on the context. Id. (citing Mathews v. Eldridge, 424 U.S. 319, 334, 96 S.Ct. 893, 47 L.Ed.2d 18 (1976)). "In determining what procedural due process requires in a given context, we [have] employ[ed] the Mathews test, which balances: (1) the private interest affected; (2) the risk of erroneous deprivation of that interest through existing procedures and the probable value, if any, of additional procedural safeguards; and (3) the governmental interest, including costs and administrative burdens of additional procedures." Id. (citing Mathews, 424 U.S. at 335, 96 S.Ct. 893).

¶ 57 The first factor, "the private interest affected," weighs in McCuistion's favor. The continuation of civil commitment without a release trial is certainly a significant deprivation of liberty and also is likely to create a stigma in the broader society. Addington v. Texas, 441 U.S. 418, 425-26, 99 S.Ct. 1804, 60 L.Ed.2d 323 (1979).

¶ 58 The second factor, "the risk of erroneous deprivation of that interest through existing procedures and the probable value, if any, of additional procedural safeguards," weighs strongly in the State's favor. Stout, 159 Wash.2d at 370, 150 P.3d 86. By the time of the review hearing, an SVP has already gone through a full civil commitment trial, where the State has to prove beyond a reasonable doubt that an "`individual suffers from a mental abnormality which renders him a danger to the community.'" In re Det. of Petersen, 138 Wn.2d 70, 78, 980 P.2d 1204 (1999) (quoting In re Personal Restraint of Young, 122 Wn.2d 1, 39, 857 P.2d 989 (1993)). Notwithstanding the 2005 amendments to the SVP statute, additional procedural safeguards also arise automatically every year. The secretary of the Department of Social and Health Services (DSHS) must "provide the committed person with an annual written notice of the person's right to petition the court for conditional release to a less restrictive alternative or unconditional discharge over the secretary's objection." RCW 71.09.090(2)(a). DSHS is required to allow a qualified professional to perform an annual current examination of the mental condition of each defendant in order to determine whether the defendant still meets the definition of an SVP. RCW 71.09.070. Each defendant also has the right to have an expert of his choosing examine him. Id. Furthermore, each defendant has the right to a show cause hearing each year, at which if the State fails to prove a prima facie case that he has so changed that he no longer meets the definition of an SVP or if the defendant presents a prima facie case that he has so changed that he no longer meets the definition of an SVP, a release trial will be ordered. RCW 71.09.090(2)(a)-(c). A defendant has the right to counsel at each of these show cause hearings. RCW 71.09.090(2)(b).

¶ 59 Additionally, defendants are often able to challenge their commitments through other means, such as CR 60(b), personal restraint petitions, and writs of habeas corpus. These numerous protections already minimize the risk of erroneous deprivation of liberty and provide adequate assurances that McCuistion would not be unfairly held in civil commitment if he truly no longer met the definition of an SVP. This factor weighs heavily in the State's favor.

¶ 60 The third factor, "the governmental interest, including costs and administrative burdens of additional procedure," also weighs very strongly in the State's favor. It is well established that the State has an interest in detaining "mentally unstable individuals who present a danger to the public." United States v. Salerno, 481 U.S. 739, 748-49, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987). Indeed, we have noted that SVPs are even more dangerous to others than other mentally ill individuals. Young, 122 Wash.2d at 45, 857 P.2d 989. We have also stated that "it is irrefutable that the State has a compelling interest both in treating sex predators and protecting society from their actions." Id. at 26, 857 P.2d 989 (emphasis added).

¶ 61 If McCuistion prevailed in his position, the costs and administrative burdens that would arise would also be unacceptably high. Here, McCuistion obtained an out-of-state expert who did not even interview him, but merely perused his records and then wrote a declaration stating that McCuistion never had a mental illness to begin with. Clerk's Papers (CP) at 616-24. Under the standard that McCuistion advocates, any expert anywhere could force a new release trial for every SVP every single year by declaring that the defendant never met the commitment definition in the first place. This is contrary to the statute's intention that SVP commitment be "for an indefinite period, until that person's condition has changed sufficiently that he or she is safe to be either at large or in a less restrictive setting." Petersen, 138 Wash.2d at 82, 980 P.2d 1204. Dr. Lee Coleman did not even attempt to show that anything about McCuistion had changed in the years since the underlying commitment. CP at 616-24. In another Court of Appeals case, Dr. Coleman unsuccessfully made the exact same argument. In re Det. of Reimer, 146 Wn.App. 179, 184, 190 P.3d 74 (2008). Under this standard, one doctor could easily create a problematic backlog in the courts with innumerable unmerited release trials every year. Where the State has a compelling interest in treating SVPs and protecting society from their crimes, the third factor weighs strongly in the State's favor.

¶ 62 When applying the Mathews test, we look at all three factors and balance the interests against each other. It is clear that in similar circumstances a compelling interest in promoting public safety can outweigh an individual's private liberty interest. The Ninth Circuit Court of Appeals, for example, in determining that it was acceptable to shift the burden of proof onto the defendant at an insanity acquiteee's release hearing, held that "[t]he state's interest in preventing the premature release of individuals who have already demonstrated their dangerousness to society by committing a criminal act outweighed the acquittee's interest in avoiding continued confinement." United States v. Phelps, 955 F.2d 1258, 1267 (9th Cir.1992). This theory applies even more strongly to SVPs.

¶ 63 Taken together, the Mathews factors weigh strongly in favor of the State. While McCuistion does have a strong private liberty interest at stake, existing procedural safeguards are already adequate to prevent erroneous deprivation of his liberty interest. Furthermore, the State has compelling interests in protecting the public and treating SVPs, and heavy costs and administrative burdens would arise in the absence of the amendments. As such, I would hold that procedural due process is not violated. The 2005 amendments to the SVP statute limit the creation of a new release trial to when there is significant reliable evidence that an SVP has truly changed. These procedures do not offend procedural due process.

III. The 2005 Amendments to RCW 71.09.090 Do not Violate Substantive Due Process

¶ 64 Even accepting the majority's formulation of the issue as one of substantive due process, due process is not violated. This court has never before held that any element of the SVP statute violates due process. Here, the majority fails to properly apply strict scrutiny when it refuses to look at both of the State's compelling interests in protecting public safety and providing treatment to SVPs.

¶ 65 When a liberty interest is protected by substantive due process, state interference with this liberty interest is subject to strict scrutiny. In re Parentage of C.A.M.A., 154 Wn.2d 52, 60-61, 109 P.3d 405 (2005). Strict scrutiny means that governmental interference with the liberty interest "`is justified only if the state can show that it has a compelling interest and such interference is narrowly drawn to meet only the compelling state interest involved.'" Id. at 61, 109 P.3d 405 (quoting In re Custody of Smith, 137 Wn.2d 1, 15, 969 P.2d 21 (1998)). The State undoubtedly has compelling interests here. "[S]exually violent predators generally have personality disorders and/or mental abnormalities which ... render them likely to engage in sexually violent behavior.... [Their] likelihood of engaging in repeat acts of predatory sexual violence is high." RCW 71.09.010. "[I]t is irrefutable that the State has a compelling interest both in treating sex predators and protecting society from their actions." Young, 122 Wash.2d at 26, 857 P.2d 989. The only question then is whether the 2005 amendments are narrowly drawn to meet only the compelling state interests involved. I would hold that they are.

¶ 66 The legislature, in enacting the 2005 amendments to RCW 71.09.090, intended to "address the `very long-term' needs of the sexually violent predator population for treatment and the equally long-term needs of the community for protection from these offenders." Laws of 2005, ch. 344, § 1. The legislature specifically found "that the mental abnormalities and personality disorders that make a person subject to commitment under chapter 71.09 RCW are severe and chronic and do not remit due solely to advancing age or changes in other demographic factors." Id. The legislature wanted to ensure that the statutory focus remains on treatment and did not want to remove the incentive for successful treatment participation. Id. It was fearful that the Young and Ward1 decisions would "subvert[ ] the statutory focus on treatment and reduce[ ] community safety by removing all incentive for successful treatment participation in favor of passive aging and distract[ ] committed persons from fully engaging in sex offender treatment." Id. The legislature also stated that persons committed as SVPs "generally require prolonged treatment in a secure facility followed by intensive community supervision in the cases where positive treatment gains are sufficient for community safety." Id.

¶ 67 Where the State has a compelling interest in providing treatment and protecting public safety, the State's actions must be narrowly tailored to meet these interests. This "`requires that the nature and duration of commitment bear some reasonable relation to the purpose for which the individual is committed.'" Young, 122 Wash.2d at 33, 857 P.2d 989 (internal quotation marks omitted) (quoting Jones v. United States, 463 U.S. 354, 368, 103 S.Ct. 3043, 77 L.Ed.2d 694 (1983)). The legislature has found that allowing evidence of actuarial models of decreased dangerousness would inhibit the incentive for committed SVPs to undergo treatment. It is vital for the legislature to ensure that committed SVPs are truly treated before their release into the community, as there is a great deal of literature discussing how recidivism rates for SVPs are considerably lower when they have completed treatment programs. See Resp't's Statement of Additional Authority (Grant Duwe et al., The Impact of Prison-Based Treatment on Sex Offender Recidivism, 21 Sexual Abuse: A Journal of Research & Treatment 1, 12, 18 (June 2009); Friedrich Lösel et al., The Effectiveness of Treatment for Sexual Offenders: A Comprehensive Meta-Analysis, J. of Experimental Criminology 135, 138 (2005)). While some doctors, such as Dr. Coleman, clearly do not believe in the SVP statute, it is the legislature that must write laws that meet strict scrutiny, not Dr. Coleman. The legislature has determined, after extensive investigation, that committed SVPs only truly lose their dangerousness when they have gone through intensive treatment and that the best way to promote both treatment and public safety is to strongly incentivize treatment. We should not second-guess the legislature's judgment. Where an SVP is committed both to provide treatment and to protect the public, the 2005 amendments that incentivize treatment bear a reasonable relationship to the purpose for the commitment. Young, 122 Wash.2d at 33, 857 P.2d 989. Substantive due process is not violated.

¶ 68 When properly characterized as an issue of procedural due process, the procedures at the annual review hearing do not violate due process. Even framing the issue as one of substantive due process, the 2005 amendments to the SVP statute are narrowly tailored to meet the State's compelling interests. For both of these reasons, I must respectfully dissent.

WE CONCUR: BARBARA A. MADSEN, Chief Justice, CHARLES W. JOHNSON, and MARY E. FAIRHURST, Justices.