Filed 9/1/23 CERTIFIED FOR PUBLICATION
COURT OF APPEAL, FOURTH APPELLATE DISTRICT
DIVISION ONE
STATE OF CALIFORNIA
THE PEOPLE, D081246
Plaintiff and Respondent,
v. (Super. Ct. No. SCE202417)
RHONDA LYNN JENKINS,
Defendant and Appellant.
APPEAL from an order of the Superior Court of San Diego County, Kenneth K. So, Judge. Reversed with directions. Rebecca P. Jones, under appointment by the Court of Appeal, for Defendant and Appellant. Rob Bonta, Attorney General, Lance E. Winters, Chief Assistant Attorney General, Charles C. Ragland, Assistant Attorney General, Collette C. Cavalier, Kathryn Kirschbaum and Elana Miller Steele, Deputy Attorneys General, for Plaintiff and Respondent. Rhonda Lynn Jenkins appeals from an order following a bench trial recommitting her as a mentally disordered offender pursuant to Penal Code sections 2970 and 2972. 1 Jenkins contends substantial evidence does not support the trial court’s finding that she represented a substantial danger of physical harm to others because of her severe mental disorder. After careful review of the record, we agree. The mental health experts who evaluated Jenkins expressed concern that she is not ready to leave the hospital due to her mental illness and level of functioning, and that she has unrealistic expectations about the challenges she would face. They nevertheless failed to identify any history of dangerous behavior beyond her commitment offense in 1999 or explain how their concerns translate into difficulty controlling her dangerous behavior. Because there is insufficient evidence to support a finding beyond a reasonable doubt that Jenkins currently represents a substantial danger of physical harm to others, we reverse the court’s order recommitting her for an additional year. FACTS AND PROCEDURAL BACKGROUND A. Underlying Offenses In 1999, in response to paranoid ideation, Jenkins attacked her 82- year-old landlord with a hammer and caused three skull fractures. She also imprisoned him on her apartment floor for six hours without emergency medical care. She was convicted of attempted murder, with offense enhancements for using a deadly weapon and great bodily injury on a person 70 years or older (§§ 187, subd. (a), 664, 12022, subd. (b)(1), 1192.7(c)(8), & 12022.7, subd. (c)). She was also convicted of false imprisonment of an elder (§ 237, subd. (b), 368, subd. (f)) and willful cruelty to elder resulting in great bodily injury (§ 368, subd. (b)(l)). She was sentenced to 17 years in state prison.
1 Undesignated statutory references are to the Penal Code. 2 B. Mentally Disordered Offender Commitment In November 2014, Jenkins was transferred from prison to a state psychiatric hospital for treatment as a mentally disordered offender under section 2962. Her commitment was extended in 2017, 2018, 2019, 2020, and 2022. In July 2022, the most recent petition for recommitment was filed under section 2970. The petition alleged that Jenkins “is still suffering from a severe mental disorder which is not in remission or cannot be kept in remission without treatment, and by reason of such mental disorder represents a substantial danger of physical harm to others.” The petition requested a year extension of her involuntary treatment. Jenkins denied the allegations in the petition. At her request, two doctors were appointed to do an independent expert evaluation. In November 2022, the allegations were tried to a judge. The prosecution submitted three medical expert reports into evidence and did not present any live witnesses. The parties stipulated to the experts’ qualifications, to the receipt of their reports into evidence, and that the court “may consider the reports in their totality.” The defense presented the testimony of Jenkins and her treating psychologist. C. Expert Reports and Testimony The medical experts agree that Jenkins suffers from a severe mental disorder, schizoaffective disorder (bipolar type). Jenkins acknowledges and does not dispute this diagnosis. 1. Dr. Jason Rowden Dr. Rowden, a forensic psychologist, recommended that Jenkins’s commitment be extended. In his opinion, Jenkins lacks insight into the nature and severity of her illness and downplays it. Although Jenkins
3 acknowledges that the commitment offense was “violent” and claims that it “will never happen again,” she continues to struggle with depression and anxiety, “remains focused on somatic delusions, and she maintains paranoid beliefs about her landlord and this paranoia is evident on the unit as well with her interactions with her peers and unit staff.” Based on her uncontrolled symptoms, Dr. Rowden concluded Jenkins’s mental disorder was not in remission. Dr. Rowden noted that Jenkins wanted to decrease or eliminate her psychiatric medications, had limited insight into them, and had difficulty discussing them with her treating psychiatrist. She had several medication changes over the past year and disagreed with some of the changes. Dr. Rowden opined that Jenkins “remains a substantial danger of physical harm to others due to her lack of insight into her ongoing symptoms, difficulty refraining from engaging in violent behavior and poor insight into
her mental illness.” 2 Her “limited insight . . . could benefit from further refining her understanding.” In Dr. Rowden’s opinion, Jenkins continues to pose a substantial danger of physical harm to others if released. She committed her underlying offenses in response to paranoid ideation and does not understand the factors that led to her violent behaviors. When Jenkins was asked about her risk for future dangerous behavior, she stated, “Financial problems is a big problem.” According to Dr. Rowden, she has poor insight into her risk for dangerous behavior and does not appreciate the role her mental disorder played in increasing her risk of dangerous behavior.
2 Dr. Rowden’s report did not explain the factual basis for his conclusion that Jenkins had “difficulty refraining from engaging in violent behavior.” As discussed below, other than the original offense, the record does not describe any further violent behavior by Jenkins. 4 2. Dr. Nicole Friedman Dr. Friedman, a psychologist, was one of two experts designated to
conduct an independent evaluation of Jenkins. Her report 3 included a discussion of her review of Dr. Rowden’s report, the CONREP hospital liaison report, her interview with Jenkins, and her own recommendations. After summarizing other evaluators’ reports, Dr. Friedman discussed her 53-minute interview with Jenkins. Jenkins was in a wheelchair and seemed to be overweight. Dr. Friedman reported that Jenkins was oriented during the interview and correctly identified her diagnosis. Jenkins correctly noted that she was at Patton because she committed a crime in 1999 and that
she served 14 years in prison. 4 She also correctly identified her symptoms at the time of the offense. Jenkins noted that at Patton, she “had ‘learned to deal with things,’ such as ‘having a support system and needing to be on medication my whole life.’ ” When asked about discharge, Jenkins described her desire to live in a board and care, continue with her medications and therapy, and get support from a church group. Dr. Friedman noted that Jenkins “stated when she committed the crime she was scared and alone and didn’t have anyone supporting her and she doesn’t want to go through that again.”
3 Confusingly, the cover page for Dr. Friedman’s report states the reason for the referral to her was to “evaluate whether sanity has been recovered pursuant to Penal Code sections 2970 and 2972.” Neither code section refers to the recovery of sanity, nor was recovery of sanity a basis for the referral to her.
4 Dr. Friedman’s report seemed to suggest that Jenkins misstated her time in prison: “She said she went to prison for fourteen years, not seventeen.” Although Jenkins was sentenced to 17 years, in fact, she served 14 years, as Jenkins correctly noted to Dr. Friedman. 5 Nowhere in her report did Dr. Friedman offer an opinion as to whether Jenkins posed a substantial danger of physical harm to others. Her only discussion of any potential for violence was her statement that Jenkins’s “lack of insight into her mental illness” and her lack of “an appropriate level of psychiatric stability” were barriers for discharge. “Without appropriate psychiatric stability, this puts her at risk for violence given her history.” Dr. Friedman recommended that “Ms. Jenkins continue with her current level of treatment in a controlled setting to give her, and the community, greater security, and stability.” 3. Dr. Stacy Berardino Dr. Berardino, a clinical forensic psychologist, opined that Jenkins cannot be safely treated or released in the community and recommended that her commitment be extended. Dr. Berardino reported that Jenkins “was cooperative” and she acknowledged that she has schizophrenia and identified her symptoms, including “paranoid delusions, depression, bipolar depression, [and] being afraid.” Jenkins nevertheless continues to deny “aspects of her symptoms and the seriousness of such despite records clearly documenting differently.” Dr. Berardino stated that she had not reasonably followed her treatment plan. Jenkins had notable medication changes the year before trial and wanted to reduce or stop psychotropic medications entirely. She has continued to exhibit symptoms of her mental illness and has continued paranoid beliefs about her landlord. Her paranoia is evident on the unit and with staff. Dr. Berardino opined that Jenkins continued to represent a substantial danger to others based on her severe mental disorder. Dr. Berardino noted her history of failing to comply with prescribed medications and of aggressive
6 behavior, her desire to reduce her medications and her tendency to argue with her psychiatrist about necessary changes. Jenkins lacks insight into her symptoms and the need for medications, what might happen if she stops medications and the relationship between her mental illness and potential future dangerousness. She does not accept responsibility for interpersonal problems, is unable to identify triggers for her aggression and lacks an adequate relapse prevention plan and a reasonable discharge plan. 4. Dr. Kimberly Claggett Dr. Claggett testified on Jenkins’s behalf as her treating psychologist. At the time of trial, she had been seeing Jenkins for approximately 11 months in weekly individual therapy sessions. While initially more guarded and depressed, Jenkins has been “much more open” and “receptive to feedback” in the two to three months before the trial. Her mood improved, as has her behavior, and her acceptance of treatment and criticisms. She testified that she was not aware of any issues of Jenkins refusing to take her medications. Dr. Claggett recommended that Jenkins remain at the psychiatric hospital but believes that she could be ready for discharge within the next year. A conditional release program that provides additional support would be more appropriate than unconditional discharge, although Jenkins is opposed to participating in such a program. Dr. Claggett was not asked to opine as to whether Jenkins posed any risk of danger or violence to others, and she did not testify as to any such risk. D. Jenkins’s Testimony Appearing by Zoom from the state hospital, Jenkins acknowledged she has a mental illness and testified that she agrees with the doctors’ diagnosis and takes all her medications voluntarily. She felt ready to be discharged to
7 a nursing home or board and care facility, explaining that she needed physical help in addition to mental health treatment. Due to her concern about getting adequate treatment for her physical illnesses and other health concerns, she did not want to be released to a conditional release program. E. The Court’s Findings After “receiving evidence” and “considering the arguments of counsel,” the trial court found that Jenkins “is still suffering from a severe mental disorder that is not in remission or cannot be kept in remission without treatment, and by means of her severe mental disorder represents a substantial danger of physical harm to others.” The court did not offer any further explanation regarding those findings. The court extended Jenkins’s commitment date until November 27, 2023. DISCUSSION Jenkins contends there was insufficient evidence to support the extension of her commitment to the state hospital. After a careful review of the record, we agree. Under the applicable standard of review, there is not substantial evidence to support a finding beyond a reasonable doubt that Jenkins poses a substantial danger of physical harm to others, and therefore her commitment should not have been extended for an additional year. A. Relevant Legal Principles A mentally disordered offender proceeding is “civil, rather than criminal, in nature.” (People v. Fisher (2009) 172 Cal.App.4th 1006, 1013.) “The Mentally Disordered Offender Act . . . requires that offenders who have been convicted of violent crimes related to their mental disorders, and who continue to pose a danger to society, receive mental health treatment during and after the termination of their parole until their mental disorder can be
8 kept in remission. [Citation.] Although the nature of an offender’s past criminal conduct is one of the criteria for treatment as a mentally disordered offender . . . , [the] Act itself is not punitive or penal in nature. [Citation.] Rather, the purpose of the scheme is to provide [mentally disordered offenders] with treatment while at the same time protecting the general public from the danger to society posed by an offender with a mental disorder.” (In re Qawi (2004) 32 Cal.4th 1, 9.) If the individual’s severe mental health disorder is not in remission or cannot be kept in remission without treatment after the initial term, the district attorney may file a petition asking the superior court to continue involuntary treatment for an additional year. (§ 2970, subds. (a) & (b).) Each yearly extension requires the court or jury to find beyond a reasonable doubt that “the patient has a severe mental health disorder, that the patient’s severe mental health disorder is not in remission or cannot be kept in remission without treatment, and that by reason of the patient’s severe mental health disorder, the patient represents a substantial danger of physical harm to others.” (§ 2972, subds. (a)(2) & (c).) When reviewing a challenge to a civil commitment based on insufficient evidence, we consider the entire record in the light most favorable to the judgment to determine whether a reasonable trier of fact could have found beyond a reasonable doubt that the defendant met the requirements for the commitment. (In re Anthony C. (2006) 138 Cal.App.4th 1493, 1503.) While inferences may constitute substantial evidence in support of a judgment, they must be the probable outcome of logic applied to direct evidence; mere speculative possibilities or conjecture are infirm. (People v. Herrera (2006) 136 Cal.App.4th 1191, 1205.) “ ‘ “ ‘ “A legal inference cannot flow from the nonexistence of a fact; it can be drawn only from a fact actually
9 established.” ’ ” ’ ” (Ibid.) “[I]n determining whether the record is sufficient . . . the appellate court can give credit only to ‘substantial’ evidence, i.e., evidence that reasonably inspires confidence and is ‘of solid value.’ ” (People v. Bassett (1968) 69 Cal.2d 122, 139.) An involuntary civil commitment requires proof beyond a reasonable doubt that the person’s mental disorder causes serious difficulty in controlling dangerous behavior “in order to distinguish those persons who are subject to civil commitment from those persons more properly dealt with by the criminal law.” (In re Howard N. (2005) 35 Cal.4th 117, 122, 132). B. Analysis 1. Jenkins Only Challenges the Court’s Dangerousness Finding
With regard to the findings the trier of fact must make before the criteria for recommitment as a mentally disordered offender are satisfied, Jenkins does not challenge the court’s determinations that she has “a severe mental disorder that is not in remission or cannot be kept in remission without treatment.” Jenkins only challenges the court’s finding that, by reason of her mental disorder, she “represents a substantial danger of physical harm to others.” (§ 2972, subd. (c).) 2. There is No Evidence Jenkins Has Been Violent Since 1999
We agree with Jenkins that her commitment offense is the only evidence in the record that she has ever been violent or dangerous. Although the mental health experts report that Jenkins has a “significant history of violence” and “difficulty controlling her aggressive behavior,” there is no evidence she has been violent or physically aggressive since her commitment offense in 1999. The reports describe incidents of “interpersonal difficulties with staff and patients,” but there is no evidence Jenkins was violent or
10 physically aggressive in any of those incidents, and instead she was described as a victim who did not respond with any violence. Dr. Claggett explained that Jenkins had been in two altercations with other patients in the year before trial, and that Jenkins “was the victim both of those fights.” The other patients who attacked Jenkins are “pretty psychotic.” One of the other patients “was responding to internal stimuli, and [Jenkins] reported it to the psychiatrist. That patient then became overly paranoid that [Jenkins] was trying to sabotage her release, and . . . hit her because of that.” The other patient had been in “several fights since she’s fought” Jenkins, and “because of [the other patient’s] aggressiveness and for safety issues, . . . [she] was placed in a side room on her own,” and was placed in a “five-points restraints[, which] means that she’s restrained down by both of her wrists, her waist and her feet,” but nevertheless “continued to target” Jenkins. Dr. Rowden and Dr. Friedman discuss an incident where Jenkins explained that she “didn’t argue with” her roommate who was “yelling and calling” Jenkins names and “tried to kick” and “threaten[ed]” to beat [Jenkins] up.” Jenkins tried to tell a staff member at the hospital. In another incident, Jenkins became agitated due to chest pains she was experiencing. She banged on the unit door, stating, “[G]et me out of here! I need help!” She stated that she “was having chest pains, why weren’t they taking me to the hospital, they told me the ambulance is coming and it never came.” A couple weeks before trial, Jenkins had to change rooms because her roommates complained that she “was being mean, telling them what to do” and her behavior “trigger[ed] the other three patients in the room.” Jenkins explained that one roommate was “stealing things . . . off my bed” and one
11 roommate would “be quiet until I laid down to go to bed and she start[ed] talking to herself then.” The roommate also told her “not to talk to the [doctor].” Jenkins “told the staff” and also “asked [the roommate] to be quiet and the next day [the roommate] said she didn’t want me in the room anymore.” Dr. Claggett explained that “some of the patients felt like [Jenkins] was kind of bossing them around, telling them what to do, and in particular, one of the more psychotic patients started to get a little bit fixated on her, so she was moved out of the room for that reason.” Our review of the record shows that Jenkins did not act with physical aggression or violence in any of the described incidents. Although Dr. Friedman and Dr. Rowden note that Jenkins “failed to take any responsibility for her actions in the conflicts in the room,” that does not mean she is violent or has difficulty controlling physical aggression. “Such a complete absence of violent or aggressive behavior of any kind over a long period of time is necessarily an important, objective factor that must not be ignored when determining a [mentally disordered offender’s] dangerousness.” (People v. Johnson (2020) 55 Cal.App.5th 96, 110 (Johnson).) 3. Jenkins’s Behavior Has Continued to Improve and Her Health Has Deteriorated
Since these incidents, Jenkins’s interactions with other roommates “have improved” and any disagreements were “kind of normal disagreements.” She has shown improved ability to discuss disagreements without pointing blame and her behavior toward staff has also “greatly improved.” She has “definitely been much more open” and “receptive to feedback.” Her “mood has generally improved,” as has her behavior, and her acceptance of treatment and criticisms. “[S]he’s now in a quieter place, has higher functioning roommates and seems to be doing better there.” Dr. Claggett and Jenkins have discussed 12 discharge planning. “[H]er emotional kind of liability, which is the swings in emotions, and her interactions with other people . . . have hugely improved, . . . [but] will benefit from a little more work in that area.” “[I]n the past, [Jenkins] has presented as kind of abrasive towards others,” but there has been “a huge turnaround in that to the point that, . . . even the patients are responding differently. She’s the unit mom these days. Everybody calls her mom.” Additionally, at the time of trial, Jenkins was “almost 70 years old” and “her health is starting to go downhill.” “[W]ithin the past year, she’s started using a wheelchair.” “[S]he does talk of a lot of . . . chronic physical pain . . . . [S]he has issues with her knees.” “She’s discussed with [Dr. Claggett] some ongoing jaw pain, [temporomandibular joint and muscle disorders]” and “shoulder or clavicle pain as well.” The expert reports did not address whether Jenkins’s age, declining health, or physical issues played any role in their conclusion that she continued to pose a danger to others. “[C]onsidering the totality of the evidence presented at [Jenkins’s] commitment extension trial,” including her lack of violence since 1999, her improved behavior, her significant medical issues and her decreased mobility, “a rational trier of fact could not have found beyond a reasonable doubt that appellant ‘represents a substantial danger of physical harm to others.’ ” (Johnson, supra, 55 Cal.App.5th at pp. 111–112; see also People v. Redus (2020) 54 Cal.App.5th 998, 1011 (Redus) [noting that the trial court “had ‘a problem’ with the proof that appellant had serious difficulty controlling his dangerous behavior” where medical expert “described appellant as ‘a fragile old man’ ” who “had ‘gone downhill physically’ ”].)
13 4. The Expert Reports Are Insufficient to Establish That Jenkins Currently Poses a Substantial Danger of Physical Harm to Others
Even though it is not necessary to have “proof of a recent overt act,” (People v. McKee (2010) 47 Cal.4th 1172, 1203) the court cannot overlook “the statutory requirement of proof beyond a reasonable doubt that the person currently poses a substantial danger of physical harm to others.” ( Johnson, supra, 55 Cal.App.5th at pp. 106–107.) Expert testimony can assist in making this determination. (People v. Zapisek (2007) 147 Cal.App.4th 1151, 1165.) “But that does not mean an unsupported psychiatric opinion will suffice.” (People v. Cheatham (2022) 82 Cal.App.5th 782, 791 (Cheatham).) “[E]xpert medical opinion evidence that is based upon a ‘ “guess, surmise or conjecture, rather than relevant, probative facts, cannot constitute substantial evidence.” ’ ” (In re Anthony C. (2006) 138 Cal.App.4th 1493, 1504.) In Johnson, supra, 55 Cal.App.5th 96, a medical expert opined that the defendant would be dangerous if released because he did not participate fully in treatment, did not have a relapse prevention plan, did not have insight into his illness and need for medication, was unlikely to take his medication if released, and was likely to decompensate if he stopped taking his medication. (Id. at p. 108.) That same expert, however, conceded on cross- examination that the defendant had not demonstrated any violent behavior in the past 30 years. (Ibid.) In reversing the recommitment order, the Court of Appeal found that the record did not contain “any evidence that [the defendant’s mental illness] would lead him to endanger others. Indeed, the evidence shows that when he did stop taking his medication for two months, although his symptoms of schizophrenia increased, he did not engage in any violent behavior whatsoever.” (Id. at pp. 109–110, emphasis in original.)
14 In Redus, supra, 54 Cal.App.5th 998, one expert recommended that the defendant be recommitted because he lacked insight into his disease, he would have a high risk of violence if released into the community and did not believe he needed medications. (Id. at pp. 1002–1004.) Another expert believed the defendant was dangerous because he did not accept his need for treatment or medications and was quietly angry inside the hospital. (Id. at pp. 1004–1006.) Nonetheless, the Court of Appeal found there was insufficient evidence that the defendant was dangerous as he “had not committed a violent act since his commitment offense some 45 years earlier” and had “controlled his dangerous behavior for decades, despite his ongoing delusions and paranoia.” (Id. at pp. 1011–1014.) In Cheatham, supra, 82 Cal.App.5th 782, the medical experts testified that the defendant “could return to substance abuse if released,” which could lead him to “stop taking his medications” and increase “his mental health symptoms.” (Id. at pp. 787–790.) The experts expressed concern that this sequence of events would lead the defendant to “have serious difficulty controlling his dangerous behavior.” (Id. at p. 790.) The defendant, however, had never “committed a single violent, aggressive, or threatening act that was attributable to his mental disorder.” (Id. at p. 794.) The Court of Appeal reversed the commitment order after concluding that a “serious mental disorder in and of itself cannot justify an extension of [the defendant’s] commitment. To find otherwise would justify indefinite involuntary commitments for all those who have a serious mental disorder without regard to the actual risk of harm they pose to others because of their disorder.” (Ibid.) The record here similarly lacks evidence that Jenkins has committed any act of violence or physical aggression after her commitment offense in
15 1999. Dr. Friedman, for example, cited Jenkins’s “lack of insight into her mental illness” as “a barrier for discharge” and concluded her “lack of psychiatric stability . . . puts her at risk for violence given her history.” But Dr. Friedman does not identify any violence since the commitment offense, and her conclusion that Jenkins remains “at risk for violence” does not meet the statutory threshold that the defendant “represents a substantial danger of physical harm to others.” (§ 2972, subds. (a)(2) & (c) [emphasis added].) Certainly Jenkins suffers from serious and ongoing mental health issues. Certainly she would benefit from continued medication and treatment. Certainly she will face serious challenges when she is discharged from Patton State Hospital. Although serious and concerning, those realities, without more, are insufficient to support a finding of substantial danger to others beyond a reasonable doubt. As the Johnson court noted, “[t]he court was understandably concerned about appellant’s ability to function and keep himself safe if he were to stop taking his medication and decompensate after being released from the hospital. However, appellant’s risk of danger to others, not his own welfare, is what was at issue at his MDO recommitment trial.” (Johnson, supra, 55 Cal.App.5th at p. 110.) We give due weight to Dr. Rowden’s statement that “the most accurate predictor of future violence is one’s past history of violence.” However, “speculation is not evidence, less still substantial evidence.” (People v. Waidla (2000) 22 Cal.4th 690, 735.) Section 2972 requires more than a conclusory speculation that a person who committed a violent offense in the past might pose a substantial danger of violence two decades later. Similar to Redus, Johnson, and Cheatham, the record here shows the medical experts believe that Jenkins should not be released because of her lack of insight as to her mental illness and the violence of her offense of
16 conviction. But after more than 23 years since that offense, the record is devoid of sufficient evidence of additional violent conduct that would support a finding beyond a reasonable doubt that her mental illness presents a substantial danger of physical harm to others. We reverse the court’s order. DISPOSITION Reversed with instructions to vacate the Order entered on November 16, 2022, and to enter an order denying the petition filed on July 26, 2022.
KELETY, J.
I CONCUR:
McCONNELL, P. J.
17 Buchanan, J., Concurring. I join the majority opinion without reservation. I write separately only to comment on another troubling feature of the People’s evidence: their experts failed to use any of the standard violence risk assessment tools in formulating their opinions about Jenkins. Instead, they relied solely on their
own unstructured clinical judgment. 1 For decades, we have known that this
is a notoriously unreliable way of predicting future violence. In my view, the time has come for courts to banish this demonstrably unsound practice in civil commitment proceedings. “Psychiatric and psychological education and training does not typically include courses in the prediction of dangerousness, and the professions have themselves disclaimed expertise of the prediction of dangerousness.” (Shuman, Psychiatric and Psychological Evidence, § 16:2 (Dec. 2022 update) (Shuman).) “Studies of predictions by psychiatrists and psychologists in the 1960s and 1970s showed poor accuracy in judging whether persons with mental disorders and sex offenders would be likely to be violent at some point after release. Indeed, the most frequently cited conclusion was [Professor John] Monahan’s statement that when mental health professionals predicted that a person would be violent, they were twice as likely to be wrong as
1 In this opinion, I use the term “unstructured” to refer to risk assessments “based solely on clinical experience and judgment of assessors using informal and subjective methods, which are predominantly justified by their training, expertise, and professional designations. [Citation.] This approach is referred to as ‘unstructured’ because of its lack of explicit rules for assessors, which increases its vulnerability to biases and as a consequence its limited reliability and validity [citation].” (Wertz, et al., A Comparison of the Predictive Accuracy of Structured and Unstructured Risk Assessment Methods for the Prediction of Recidivism in Individuals Convicted of Sexual and Violent Offense (2023) 35 Psychological Assessment No. 2, 152 (Wertz).) right.” (Appelbaum, Reference Guide on Mental Health Evidence, in Reference Manual on Scientific Evidence (3d ed. 2011) p. 849 & fns. 204, 205 (Reference Guide), citing Monahan, The Clinical Prediction of Violent Behavior (1981) p. 60.) Nearly 50 years ago, our Supreme Court acknowledged this reality. After reviewing relevant empirical studies and scientific literature, the court declared the state of the evidence to be “ ‘unequivocal’ ” that “[n]either psychiatrists nor anyone else have reliably demonstrated an ability to predict future violence or ‘dangerousness.’ ” (People v. Burnick (1975) 14 Cal.3d 306, 327 (Burnick).) As the high court noted, “the same studies which proved the inaccuracy of psychiatric predictions have demonstrated beyond dispute the no less disturbing manner in which such prophecies consistently err: they predict acts of violence which will not in fact take place (‘false positives’) thus branding as ‘dangerous’ many persons who are in reality totally harmless.” (Ibid.; see also People v. Murtishaw (1981) 29 Cal.3d 733, 768 (Murtishaw) [“Numerous studies have demonstrated the inaccuracy of attempts to forecast future violent behavior.”].) In the intervening half century, the use of unstructured clinical judgments to predict a person’s risk of violence has not proven to be any more accurate. But much work has been done to develop more reliable methods of prediction. Specifically, researchers have identified known risk factors that are empirically linked to violent behavior and incorporated them into structured violence risk assessment instruments. (Reference Guide, supra, at pp. 848–849.) “Among the best known of these are the HCR-20, the Violence Risk Assessment Guide (VRAG), and the computerized Classification of Violence Risk (COVR). A set of instruments also exists for the prediction of the risk of future sexual offenses.” (Id. at p. 848, fns. omitted.)
2 These assessment tools now provide a structured framework for analyzing some or all of the following four steps in predicting a person’s risk of violence: (1) identifying the presence or absence of empirically valid risk factors for violence, (2) establishing a method for measuring or scoring these individual risk factors, (3) establishing a procedure for combining scores on the individual risk factors into a total score, and (4) producing a final estimate of violence risk. (Faigman, et al., 2 Modern Scientific Evidence: The Law and Science of Expert Testimony (2022-2023 ed.) § 9:11 (Faigman); see also id., § 9:17, at Table 2.) Though far from perfect, the predictive value of various risk assessment tools has been validated in peer-reviewed studies. (See, e.g., Cartwright, et al., Predictive Value of HCR-20, START, and Static- 99R Assessments in Predicting Institutional Aggression Among Sexual Offenders, 42 Law & Hum. Behav. 13, 14 (2018) [“Meta-analytic research shows that many violence risk assessment instruments can have good validity in predicting violence”].) Modern methods for predicting violent behavior now vary according to how many of these four steps they structure. At one end of the spectrum is the “completely unstructured (‘clinical’) assessment,” which “structures none of these four components.” (Faigman, supra, at § 9:11.) At the other extreme is a “completely structured (‘actuarial’) assessment” tool such as the VRAG, which structures all four steps. (Ibid.) Occupying a middle ground are violence risk assessment tools such as the HCR-20 (which structures the first two steps) and the COVR (which structures the first three steps). (Ibid.) Risk assessment methods that combine a structured use of empirically validated risk factors with professional judgment are often referred to as “structured professional judgment.” (See Douglas, et al., Historical-Clinical- Risk Management-20, Version 3 (HCR-20v3): Development and Overview, 13
3 International Journal of Forensic Mental Health (2014) 93, 94 [describing structured professional judgment approach].) Of these varying approaches, “there is widespread consensus among researchers that the unstructured (‘clinical’) approach is the least accurate and has only tenuous empirical support. Empirical studies find that ‘clinicians are able to distinguish violent from nonviolent patients with a modest, better-than-chance level of accuracy,’ but that overall ‘clinicians are relatively inaccurate predictors of violence.’ ” (Faigman, supra, at § 9:13, fns. omitted; see also id., § 10:30 [stating as a “scientific certainty” that predictions of sexual violence using actuarial instruments “are superior to those based on unaided clinical judgment,” which “have never been shown to exceed the accuracy exhibited by laypeople”].) “Clinical judgment alone . . . has been criticized for being subjective and impressionistic, lacking transparency, reliability and validity, and leading to idiosyncratic decisions based on the experience of the assessor.” (Roychowdhury & Adshead, Violence Risk Assessment as a Medical Intervention: Ethical Tensions (2014) 38 Psychiatric Bulletin 75, 80.) Earlier this year, a study published in a journal of the American Psychological Association confirmed once again that structured risk assessment tools are more reliable than unstructured clinical judgment. The authors concluded: “In accordance with previously published results, the results indicated a higher predictive accuracy for structured compared to unstructured risk assessment approaches for the prediction of general, violent, and sexual recidivism. Taken together, the findings underline the limited accuracy of [unstructured clinical judgments] and provided further support for the use of structured and standardized risk assessment procedures in the area of crime and delinquency.” (Wertz, supra, at p. 152.)
4 The authors further noted: “A number of previously published studies consistently highlighted that unstructured assessments were significantly more susceptible to biases [citations].” (Id. at p. 153.) In recognition of the validity of structured violence risk assessment tools, the California Legislature and Judicial Council have explicitly endorsed their use in a variety of contexts. (See, e.g., Pen. Code, § 290.5, subd. (a)(3) [court may consider “the person’s risk levels on SARATSO static, dynamic, and violence risk assessment instruments” in deciding whether to order continued sex offender registration]; Pen. Code, § 1170.05, subd. (d)(4) [making ineligible for alternative custody program those who are screened “using a validated risk assessment tool and determined to pose a high risk to commit a violent offense”]; Pen. Code, § 1170.06, subd. (d)(1) [same]; Pen. Code, § 1320.35 [allowing use of “pretrial risk assessment tool” that has been validated as accurate, reliable, and unbiased using scientifically accepted methods]; Cal. Stds. Jud. Admin., § 4.35 [allowing courts at sentencing to use risk assessment instruments that have been validated as accurate and reliable].) For parole hearings, a regulation requires state psychologists to prepare reports that “shall incorporate structured risk assessment instruments like the HCR-20-V3 and STATIC-99R that are commonly used by mental health professionals who assess risk of violence of incarcerated individuals.” (Cal. Code Regs., tit. 15, § 2240, subd. (a).) Yet courts have been slow to respond to these developments. Although our Supreme Court long ago limited the use of unreliable opinion testimony to predict a risk of future violence in capital cases (Murtishaw, supra, 29 Cal.3d at pp. 767–775), it has imposed no similar constraints in other contexts. As this case demonstrates, unstructured clinical opinion evidence continues to be deployed to deprive people of their liberty in civil commitment
5 proceedings. “Unfortunately, civil commitment hearings have provided some of the worst examples of unhelpful, conclusory psychiatric and psychological testimony.” (Shuman, supra, at § 16:5.) In any other context, the use of such inherently unreliable and speculative expert opinions would be barred. (See Sargon Enterprises, Inc. v. University of Southern California (2012) 55 Cal.4th 747, 771–772 [discussing trial court’s responsibility to act as gatekeeper to exclude unreliable expert testimony].) Now that more reliable tools are available, I see no good reason to continue allowing the use of such a manifestly inferior method of predicting future violence—one that “brand[s] as ‘dangerous’ many persons who are in reality totally harmless.” (Burnick, supra, 14 Cal.3d at p. 327.) And even if admissible, unstructured clinical predictions of violence about someone like Jenkins who has committed no violent act for decades should not suffice to support a civil commitment and satisfy the requirement of substantial evidence, i.e., evidence that is “ ‘reasonable in nature, credible, and of solid value.’ ” (Conservatorship of O.B. (2020) 9 Cal.5th 989, 1006.) If we know anything about such predictions, it is that they lack solid value. In 1975, when our Supreme Court first acknowledged the unreliability of this type of evidence, it nevertheless declined to “go so far as to join in the conclusion of certain well-known writers that in civil commitment proceedings no psychiatrists should be permitted to give their opinions as to future dangerousness and that any commitment based on such an opinion constitutes a deprivation of liberty without due process of law.” (Burnick, supra, 14 Cal.3d at pp. 327–328.) Yet even then, the court conceded that these were “not the views of a radical fringe of either the psychiatric or legal professions.” (Id. at p. 328, fn. 19.) The court quoted two of these commentators as follows: “ ‘Justifying the deprivation of a person’s liberty on
6 the basis of judgments and opinions that have not been shown to be reliable and valid should be considered a violation of both substantive and procedural due process. Certainly a procedure by which judges flipped coins to determine who would be committed would offend our sense of fundamental fairness. It is our contention that psychiatric judgments have not been shown to be substantially more reliable and valid.’ ” (Ibid., quoting Ennis & Litwack, Psychiatry and the Presumption of Expertise: Flipping Coins in the Courtroom (1974) 62 Cal. L.Rev. 693, 743.) What has changed in the last half century is that we now have evidence-based instruments at our disposal. No doubt, we will never be able to predict future dangerousness with any precision or certainty. But we are no longer compelled to rely on unstructured clinical judgments as a matter of necessity. (See Murtishaw, supra, 29 Cal.3d at p. 772 [noting in 1981 that “expert prediction, unreliable though it may be, is often the only evidence available to assist the trier of fact” in determining whether someone is dangerous].) We can instead insist on the use of structured risk assessment tools that yield more dependable and less subjective results. We can also require that these tools be properly validated and correctly implemented by adequately trained clinicians. (See Faigman, supra, at § 9:14 [noting that “[u]n-validated and poorly validated risk assessment instruments abound” and “appropriate processes for implementing risk assessment instruments are often violated in practice, especially in adversary contexts”].) And we must be mindful of the limitations of these tools as well. (See Cal. Stds. Jud. Admin., § 4.35(d)(4)(C) [requiring courts using a risk assessment tool at sentencing to consider “any limitations of the instrument” including whether “any scientific research has raised questions that the instrument unfairly classifies offenders by gender, race, or ethnicity”].)
7 What we should not indulge anymore is the pretense that unstructured clinical judgments are a defensible way of predicting future violence in civil commitment proceedings.
BUCHANAN, J.