¶22 (dissenting) — I agree with the majority that Alan Meirhofer has not proved his entitlement to a full evidentiary hearing under RCW 71.09.090 because subsection .090(4)(b) permits a hearing only upon proof of either of *652two facts: a physiological change has rendered the detainee permanently unable to commit a sexually violent act or there has been a “change in the person’s mental condition brought about through positive response to continuing participation in treatment which indicates that the person meets the standard for conditional release.” RCW 71.09-.090(4)(b)(ii). Meirhofer has failed to prove either type of change.
¶23 Rather than arguing his condition changed physiologically or as a positive response to treatment, Meirhofer argues that the basis for his commitment changed when the diagnoses describing his mental condition changed. The majority, having held that Meirhofer has no recourse under the statute, ironically holds that Meirhofer has not proved that “these statutory remedies are inadequate as applied to him.” Majority at 647, 651. Accordingly, the majority concludes that Meirhofer cannot bring a personal restraint petition to challenge the constitutionality of his continued detention as a sexually violent predator (SVP) because the statute provides Meirhofer with “ ‘[o]ther grounds ... to challenge the legality’ ” of his restraint. Majority at 649 (first alteration in original) (quoting RAP 16.4(c)(7)).
¶24 I disagree. A statutory remedy that does not allow a challenge to the constitutionality of restraint cannot be an adequate remedy. Moreover, the legislature cannot restrict the right of a person to challenge his or her restraint through a personal restraint petition. In a personal restraint petition, we provide relief for those restrained when the manner of the restraint is unconstitutional. RAP 16.4(c). It is unconstitutional to impose a civil commitment on someone who is not both mentally ill and dangerous. Foucha v. Louisiana, 504 U.S. 71, 75-76, 112 S. Ct. 1780, 118 L. Ed. 2d 437 (1992); In re Pers. Restraint of Young, 122 Wn.2d 1, 27, 857 P.2d 989 (1993).
¶25 Both Meirhofer and the State presented evidence at Meirhofer’s annual show cause hearing that his diagnoses have changed since his commitment hearing and that his *653risk of reoffending has declined substantially. These changes are significant: there is a reasonable possibility that a reasonable juror would decline to find, beyond a reasonable doubt, that Meirhofer continues to be mentally ill and dangerous. I would reverse the trial and appellate courts and remand for an evidentiary proceeding. For these reasons, I respectfully dissent.
DISCUSSION
¶26 In order to be committed, an SVP must be diagnosed with a mental illness that renders them dangerous. Civil confinement is subject to strict scrutiny because it is a massive curtailment of liberty, Young, 122 Wn.2d at 26. Liberty can be curtailed only by a narrowly drawn, compelling state interest. State v. McCuistion, 174 Wn.2d 369, 275 P.3d 1092 (2012), cert. denied, 133 S. Ct. 1460 (2013). A State’s interests are narrowly tailored to justify civil commitment only when the State can demonstrate a person continues to be mentally ill and dangerous. Foucha, 504 U.S. at 75-76; Young, 122 Wn.2d at 27. Without a finding of scientifically accepted mental illness and dangerousness, SVP commitment would constitute double jeopardy, continuing to punish those who have already served a sentence for their criminal behavior. See Young, 122 Wn.2d at 71 (Johnson, J., dissenting). Meirhofer’s diagnoses have changed since his commitment trial.
I. Relief under a Personal Restraint Petition
¶27 The SVP statute does not contemplate the situation before us. A State cannot continue to confine someone who is no longer mentally ill and dangerous. Id. The statute addresses this by providing periodic review and release mechanisms for individuals. RCW 71.09.090. If the detainee has changed through treatment or physical injury, he or she is entitled to a new hearing to determine if he or she continues to meet the definition of an SVP. RCW 71.09.090(4). However, even if Meirhofer’s condition has not changed as a *654result of treatment, the change in his diagnoses may mean that he is no longer mentally ill and dangerous.
¶28 Contrary to the State’s assertions, petitioners are not limited to challenging their confinement through statutes that do not contemplate their circumstances. Importantly, personal restraint petitions are not motions for reconsideration and they are not substitutes for direct review; the relief they seek is collateral. “The whole purpose of collateral review is to provide a forum for potentially meritorious prisoners’ claims.” In re Pers. Restraint of James, 96 Wn.2d 847, 855, 640 P.2d 18 (1982) (Utter, J., concurring). “Though it does tax the judicial system to provide such collateral review, its proven historical value looms much larger than the particular economic exigencies with which we as judges are now beset.” Id. (footnote omitted).
¶29 When alternative grounds of relief are inadequate and a petition is not barred by RCW 10.73.090, RAP 16.4 unconditionally allows persons under restraint to raise constitutional claims. Id. at 853; RAP 16.4(a). Here, the alternative grounds for relief are inadequate: Meirhofer argues that his diagnoses changed, altering the justification for his commitment. Pers. Restraint Pet. (PRP) at 8. The statute contemplates relief only in cases in which the condition of the person has changed, not changes in the diagnoses justifying commitment.
¶30 To challenge confinement through a personal restraint petition, a detainee must present a sufficiently significant change, such that a reasonable juror could conclude that the detainee is no longer mentally ill and dangerous. Meirhofer’s restraint may be unlawful because the basis for Meirhofer’s original commitment no longer exists.
II. Change in Meirhofer’s Diagnoses
¶31 The court will grant relief in a personal restraint petition if the petitioner is under restraint and the restraint *655is unlawful. RAP 16.4(a); In re Pers. Restraint of Grantham, 168 Wn.2d 204, 212-13, 227 P.3d 285 (2010). Meirhofer demonstrates two reasons why he is unlawfully restrained and entitled to relief. First, Meirhofer’s current diagnoses are different from the original diagnoses that the jury used to classify him as an SVP. Second, two of Meirhofer’s diagnoses are not accepted by the medical community.
A. Meirhofer’s diagnoses have changed since his commitment trial
¶32 Two licensed psychologists evaluated Meirhofer prior to his commitment trial; both diagnosed him with pedophilia and amphetamine dependence. In addition to those diagnoses, one of the psychologists diagnosed Meir-hofer with paraphilia, not otherwise specified (NOS)-non-consent, and the other diagnosed him with personality disorder, NOS, with antisocial traits. Meirhofer continued to be evaluated annually for nine years with similar diagnostic results: (A) pedophilia, (B) paraphilia, NOS, noncon-sent, and (C) personality disorder, NOS, antisocial.
¶33 In 2010, Meirhofer’s diagnoses changed. The State’s expert diagnosed him with: (not-A) “Rule Out Pedophilia,” (B) paraphilia, NOS, nonconsent, (C) personality disorder, NOS, antisocial, and (D) paraphilia, NOS, hebephilia. PRP, Apps. B, G. In addition, an expert retained by Meirhofer made two diagnoses, neither of which was a sexual disorder. After reviewing the commitment documents, interviewing Meirhofer, and using actuarial assessments, Meirhofer’s expert diagnosed him with (1) alcohol dependence and amphetamine dependence, remission in controlled environment and (2) personality disorder, NOS, with antisocial traits by history. PRP, App. D.
¶34 We can chart these diagnoses and their consequences, thus:
Original diagnoses: A + B + C = sexually violent predator
2010-11 diagnoses: not-A + B + C + D = ?
*656The State’s position is that (A + B + C) yields exactly the same result as (not-A + B + C + D) and that therefore, Meirhofer is still an SVP. The State’s position is fuzzy logic, or perhaps fuzzy psychology.
B. Meirhofer is currently confined for “mental illnesses” that are not recognized by the medical community
¶35 The majority justifies Meirhofer’s continued confinement by relying on mental illnesses that are not accepted by the majority of the scientific community. However, the mental illness must be a legitimate illness widely recognized by the psychiatric community. This is made clear in Kansas v. Hendricks, in which the United States Supreme Court ruling upheld the Kansas SVP commitment scheme in a five-to-four decision. 521 U.S. 346, 358, 117 S. Ct. 2072, 138 L. Ed. 2d 501 (1997). Although divided on the holding, all nine justices agreed that civil commitment is constitutional only if we can distinguish individuals who are mentally disordered from those who are mere criminals. Id. at 360 (the disorder must be one that “the psychiatric profession itself classifies as a serious mental disorder”), 373 (Kennedy, J., concurring) (confinement cannot be based on a mental abnormality that is “too imprecise”), 375 (Breyer, J., dissenting) (the statute is constitutional, inter alia, because pedophilia is a “serious mental disorder”).
¶36 Science must have “gained general acceptance in the particular field in which it belongs” before it is admitted in trial. Frye v. United States, 54 App. D.C. 46, 293 F. 1013, 1014 (1923). One way to determine “general acceptance” with respect to psychiatric disorders is to refer to the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM). See, e.g., Hall v. Florida, _ U.S. _, 134 S. Ct. 1986, 1991, 1994, 2000, 188 L. Ed. 2d 1007 (2014) (determining the meaning of “intellectual disability” using the DSM); State v. Greene, 139 Wn.2d 64, 71, 984 P.2d 1024 (1999) (the DSM is an authoritative source that “ ‘ “reflect [s] a consensus of current *657formulations of evolving knowledge” in the mental health field’ ” (quoting State v. Greene, 92 Wn. App. 80, 98, 960 P.2d 980 (1998) (quoting Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders, at xxvii (4th ed. 1994) (DSM-IV))).13 In 2010, the State’s expert diagnosed Meirhofer with two disorders that are not in previous or current versions of the DSM: (1) hebephilia and (2) paraphilia, NOS, nonconsent.14 See DSM-IV at xxvii; Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 566-67 (4th rev. ed. 2000) (DSM-IV-TR); Am. Psychiatric Ass’n, Diagnostic and Statistical Manual of Mental Disorders 697 (5th ed. 2013) (DSM-5). Likewise, the international standard diagnostic tool does not include a code for hebephilia or paraphilia, NOS, nonconsent. World Health Org., International Statistical Classification of Diseases and Related Health Problems § F65.4 (10th rev. ed. 2013) (ICD-10), http://www.cdc.gov/nchs/data/dvs/2e_volume l_2013.pdf.15 When neither the American psychiatric community nor the international medical community recognizes a disorder, we should not do so either.
¶37 Paraphilia, NOS, nonconsent, and hebephilia are not listed in the DSM-IV or DSM-5. The essential features of paraphilia are “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors generally involving 1) nonhuman objects, 2) the suffering or humiliation of oneself or one’s partner, or 3) children or other nonconsenting per*658sons . . . DSM-IV-TR, supra, at 566.16 The explicit diagnosis of “Paraphilia Not Otherwise Specified” applies to paraphilias that are less frequently encountered, such as necrophilia, sexual attraction to corpses, and klismaphilia— receiving sexual pleasure from introducing liquids into the rectum or colon. Id. at 576 (boldface omitted). The diagnosis “paraphilia not otherwise specified nonconsent” is not included in the list of paraphilias nor was it adopted in the DSM-5. In fact, the DSM-5 explicitly rejected “ ‘coercive paraphilia’” as a diagnosis. Allen Frances, Should Having Antisocial Personality Qualify a Rapist for SVP Commit-mentPsychiatric Times, July 15, 2011, http://www.psychiat rictimes.com/couch-crisis/should-having-antisocial-personal ity-qualify-rapist-svp-commitment. Similarly, “hebephilia,” defined as attraction to pubescent children, is not listed in the DSM-5 or previous versions. The DSM-5 sexual disorders workgroup considered adding hebephilia to the pedophilia category but explicitly rejected the addition because it was based on imprecise and incomplete research. Allen Frances & Michael B. First, Hebephilia Is Not a Mental Disorder in DSM-IV-TR and Should Not Become One in DSM-5, 39 J. Am. Acad. Psychiatry & L. 78, 82-84 (2011).17
III. Meirhofer Is Entitled to an Evidentiary Hearing
¶38 The majority dismisses Meirhofer’s personal restraint petition on the basis of Meirhofer’s 2000 restraint order, this court’s decision in McCuistion, and an inaccessible avenue for statutory relief. I address these arguments in turn.
*659¶39 Relief is available under RAP 16.4 because Meir-hofer is currently restrained. The majority states that RAP 16.4(c)(2) and (3) likely do not apply because Meirhofer’s 2000 restraint order is “not before us.” Majority at 649. However, Meirhofer is challenging his current confinement; he is not challenging the 2000 commitment order. A State’s interests are narrowly tailored to justify civil commitment only when the State can demonstrate a person continues to be mentally ill and dangerous. Foucha, 504 U.S. at 75-76; Young, 122 Wn.2d at 27. Annual review is one way of ensuring that the “ ‘nature and duration of commitment bear some reasonable relation to the purpose for which the individual is committed.’ ” Jones v. United States, 463 U.S. 354, 368, 103 S. Ct. 3043, 77 L. Ed. 2d 694 (1983) (quoting Jackson v. Indiana, 406 U.S. 715, 738, 92 S. Ct. 1845, 32 L. Ed. 2d 435 (1972)). But it is not the only way. Meirhofer is lawfully subject to civil restraint only so long as he continues to be both mentally ill and dangerous; his current restraint cannot rest solely on the 2000 order. Foucha, 504 U.S. at 75-76; Young, 122 Wn.2d at 27. As such, he may seek relief under RAP 16.4(c).
¶40 The majority incorrectly relies on this court’s decision in McCuistion to deny Meirhofer a constitutional right to an evidentiary trial. Majority at 648-49. While the majority correctly reads paragraph 6 of that decision in noting that McCuistion brought “many challenges,” majority at 649 n.10, it overlooks the conclusions the trial court reached after considering those arguments, recited in paragraph 7:
The trial court dismissed Dr. Coleman’s report as “essentially a re-argument of the original finding that Mr. McCuistion is a sexually violent predator” and reasoned that Dr. Coleman’s disagreement “with past examiners and fact-finders does not, itself, make his opinion the correct one.” The court further explained that “[t]he change in his behavior within the confines of a secure facility does not demonstrate that his mental disorder has been changed in any way.”
*660McCuistion, 174 Wn.2d at 377-78 (alteration in original) (citations omitted). Furthermore, the majority ignores the grounds for our decision — -we specifically concluded that McCuistion’s argument consisted of little more than a collateral attack on the original finding that McCuistion was an SVP and that his other arguments were merely a series of “vague assertions,” “bare assertion,” and declarations that, even if we were to consider, would have been insufficient to make a prima facie showing of change. Id. at 382-84. McCuistion is distinguishable from Meirhofer; Meirhofer does not challenge his initial commitment, and Meirhofer’s change in diagnosis could lead a reasonable juror to conclude he is no longer mentally ill and dangerous.
¶41 Finally, we disagree with the majority’s dismissal of Meirhofer’s personal restraint petition under RAP 16.4(d) because, according to the majority, the secretary of the Washington State Department of Social and Health Services (DSHS) can authorize the detainee to petition for conditional release without first having to show cause. Majority at 650-51; see also In re Det. of Petersen, 138 Wn.2d 70, 81, 980 P.2d 1204 (1999). The secretary of DSHS does not provide a meaningful avenue of relief that is sufficient to preclude Meirhofer from seeking relief under RAP 16.4(d). It would violate due process to allow DSHS to preclude a detainee from accessing the courts. The jailor cannot also be the independent judiciary.
IV. Meirhofer Is Entitled to Relief
¶42 An SVP may challenge confinement through a personal restraint petition when changes occur that fall outside the statute. It is a question of fact whether a change in the diagnoses leads to the conclusion that the SVP is no longer mentally ill and dangerous. I would hold that if a detainee presents a sufficiently significant change, such that there is a reasonable possibility that a trier of fact could find the person is no longer suffering from a mental abnormality that renders him or her dangerous, then the *661detainee is entitled to a new evidentiary hearing to determine if he or she continues to meet the constitutional criteria for confinement.
¶43 Meirhofer meets the test: it is reasonably possible that his change in diagnoses could lead a reasonable trier of fact to believe that he is no longer suffering from a mental abnormality that renders him dangerous. In considering Meirhofer’s case, we examine the evidence that would be presented to the jury to make the case that Meirhofer is no longer mentally ill and dangerous.
¶44 The jury would consider if Meirhofer continues to be mentally ill. The jury would be told that Meirhofer was diagnosed with paraphilia, NOS, nonconsent, and para-philia, NOS, hebephilia. They would learn that neither diagnosis is found in the pages of the DSM. They would hear of the controversial nature of these two diagnoses and the rejection of both by the national and international psychiatric communities. The jury would then consider Meirhofer’s third diagnosis, antisocial disorder, and would hear that the DSM assigns antisocial personality disorder to individuals who habitually violate the rights of others without remorse. The definition includes four diagnostic criteria and seven subfeatures, including failure to obey the laws. The jury would be told that under this assessment, Meirhofer will likely never lose the diagnosis of antisocial personality disorder, regardless of his participation in treatment. They would probably be told that the United States Supreme Court explicitly rejected the assertion that a person with previous criminal history plus a diagnosis of “antisocial personality that sometimes leads to aggressive conduct . . . may be held indefinitely.” Foucha, 504 U.S. at 82. Notably, the jury would not hear that Meirhofer is a pedophile, a label widely recognized and associated with egregious criminal behavior.
¶45 The jury would also consider whether Meirhofer is still dangerous. They would hear that he was convicted in 1988 and served a 9-year prison sentence, after which he *662was confined at the Special Commitment Center on McNeil Island for 14 years. They would be shown actuarial instruments18 that predict his likelihood of reoffending. They would be told that the State’s expert predicted a likelihood of reoffending between 20 percent and 30 percent, down from 92 percent during his commitment trial. Mot. for Discr. Review (June 15, 2012) at 12-13, App. B at 13, App. G at 12. Meir-hofer’s own expert would confirm this score. Id. App. D at 23, 29. The jury would be told that the State’s psychologist, despite his actuarial assessment, believes that Meirhofer is still dangerous because of his criminal actions nearly 27 years prior. They would be told that his risk assessment will never drop to zero because of his previous criminal behaviors.
¶46 Given all the evidence, the jury would weigh and consider if Meirhofer continues to have a mental abnormality that renders him dangerous, beyond a reasonable doubt. Beyond a reasonable doubt is the highest standard of proof, meaning that the jury must decide that there is no doubt that Meirhofer continues to be mentally ill and dangerous. A jury might find that the State has carried this heavy burden of proof, but there is a reasonable possibility that the jury would hold to the contrary and that Meirhofer would be released. Regardless, due process requires a hearing to determine whether the significant change in Meirhofer’s diagnoses means that he is no longer an SVP.
¶47 Meirhofer’s case might be unusual. Not all changes in diagnosis will require relief under a personal restraint petition. For example, I agree with the majority that the detainee in Klein would not be afforded a new proceeding under the test we propose here. State v. Klein, 156 Wn.2d 102, 124 P.3d 644 (2005). Klein, an insanity acquittee, was *663originally diagnosed with psychoactive substance induced organic mental disorder, a mental disorder induced by using one drug. Her diagnosis was later changed to polysubstance dependence, a dependence on many drugs. Id. at 120. The court concluded that Klein’s diagnoses are based on “the very same symptoms [ and] differ only in the name attached to it.” Klein, 156 Wn.2d at 120-21. The change from dependence on one drug to many drugs does not give rise to a reasonable possibility that a reasonable trier of fact would determine that Klein no longer suffered from a mental illness that made her dangerous.
¶48 Unlike Klein’s circumstances, Meirhofer’s diagnostic change removed one of his original diagnoses (not-A) and added a nebulous new diagnosis (D). In contrast, Klein’s diagnosis could be represented as a change in diagnosis from (A) to (A+). Therefore, I would remand for an eviden-tiary hearing to determine if Meirhofer continues to meet the criteria for confinement. We cannot live in the land of make-believe, where fictitious roads for relief exist and we pretend that a change in diagnosis makes no difference. The cornerstone of our system is due process — we must continue to believe in the power of the jury to make a determination for the community. A jury should determine if, beyond a reasonable doubt, Meirhofer continues to meet the criteria for commitment.
¶49 Given the reasonable possibility that a trier of fact could determine Meirhofer’s restraint is unlawful, the evi-dentiary hearing should be granted. This test would not empty the Special Commitment Center at McNeil Island of its detainees, but it would provide an evidentiary trial for a subset of detainees whose situations fall outside of the SVP statute.
CONCLUSION
¶50 Meirhofer’s diagnoses have sufficiently changed in a manner other than what is detailed in the SVP statute. A *664jury has not found that these conditions alone support a finding that civil commitment is the least restrictive means to adequately protect the community. Under the personal restraint petition analysis, I would hold that Meirhofer is entitled to an evidentiary hearing to determine if the change in diagnoses justifies Meirhofer’s continued indefinite confinement under the statute.
¶51 Therefore, I dissent.
Stephens and Gordon McCloud, JJ., concur with Wiggins, J.It would be an overstatement to say that only disorders listed in the DSM constitute a mental illness under which a person can be committed as an SVP.
Hebephilia and paraphilia, NOS, nonconsent describe sexual urges that, by themselves, are not dangerous. Only if these urges are acted on do they become a serious crime. Allen Frances & Michael B. First, Hebephilia Is Not a Mental Disorder in DSM-IV-TR and Should Not Become One in DSM-5, 39 J. Am. Acad. Psychiatry & L. 78, 84 (2011).
The ICD-10 does include “pedophilia,” defined as “[a] sexual preference for children, boys or girls or both, usually of prepubertal or early pubertal age.” ICD-10, § F65.4.
Paraphilia has eight categories: exhibitionism, fetishism, frotteurism, pedophilia, sexual masochism, sexual sadism, transvestic fetishism, and voyeurism. DSM-IV-TR, supra, at 566-67; see also DSM-5, supra, at 685-705.
The proposed distinction between pedophilia and hebephilia was narrow. Pedophilia describes those individuals who are attracted to prepubescent children (those with downy pubic hair or no pubic hair), while hebephilia would have described those who are attracted to pubescent children (or those with coarse pubic hair). Frances & First, supra, at 83.
An actuarial instrument is an empirical tool that estimates recidivism rates of sex offenders by weighing a number of risk factors. The offender receives a score of one or zero based on the presence or absence of the risk factor, the total is then added up, and the total score represents the offender’s risk of future offense. The weightiest risk factor is the number of prior sex offenses. See generally Dennis M. Doren, Using Risk Assessment Instrumentation, In Evaluating Sex Offenders: A Manual For Civil Commitments and Beyond 103 (2002); In re Del. of Thorell, 149 Wn.2d 724, 753, 72 P.3d 708 (2003).