The Quandary of Unrestorability

Forensic psychiatrists have long been interested in the topic of incompetence to stand trial (ICST), but research in this area did not really begin until the early 1970s, when McGarry operationalized the Dusky decision,¹ and the U.S. Supreme Court handed down its decision in Jackson v. Indiana.² Since then, the difficulties associated with identifying ICST defendants and restoring them to competence have produced a small, intermittent stream of research articles. Initially, most of the publications tried to identify which characteristics of a defendant might predict a finding of ICST, but more recently, the focus has shifted to identifying which factors might predict the outcome of restoration to competence (RTC). Research on the outcome of attempted restoration has generally found that 80 to 90 percent of ICST defendants are restored to competence within six months of treatment. The remaining 10 to 20 percent, though, pose a more difficult problem, not only because prosecutors are loath to drop the charges against a defendant, but also because unrestored defendants can consume a disproportionate amount of state mental health resources, particularly if they are state hospital patients. As a low-base-rate phenomenon, unsuccessful restoration is difficult to predict in any given ICST defendant. Nonetheless, I believe it is time for our field to study more carefully the area of prediction of unsuccessful restoration, for two reasons: first, psychiatrists and psychologists are called on by statute in all but seven states to predict the probability of RTC of individual defendants, either at the initial evaluation or after referral for RTC; second, as a result, we must develop an evidence base that will allow us, as evaluating and treating clinicians, to make reasonably accurate predictions about who will not be restored for the lawyers, judges, and state mental health authorities who are involved in these cases.

As is well known to forensic psychiatrists, the Jackson decision held that, “At the least, due process requires that the nature and duration of commitment bear some reasonable relation to the purpose for which the individual is committed.”² The Court, however, left it to the states to determine what to do with a defendant who was not restored to competence at the end of a “reasonable” effort at restoration. A logical place to start in considering an approach to the problem of unrestorable defendants is to review the current state of affairs in this rather specialized part of criminal and mental health law, particularly since previous commentators have concluded that the response of the states to the Jackson decision was less than overwhelming. For example, in 2003, Miller³ concluded that more than half the states had no effective limit on the duration of RTC.

A review of the state statutes on the evaluation, restoration, and disposition of defendants found incompetent to stand trial is rather illuminating (Tables 1, 2). Based on my reading of these statutes, four states have no statutory limit on the length of time a defendant can be held for restoration to competence: Delaware, Maine, Mississippi, and Montana. Vermont has no statutory limit on the time for RTC but requires that the defendant meet civil commitment criteria. Another five states provide for a limited initial period for restoration but have no statutory limit on the length of time an incompetent defendant can be held after that period has elapsed: Hawaii (felonies only), Missouri, Nebraska, Wisconsin, and Wyoming. Finally, nine states approach the matter slightly differently, though with similar effect, as they allow for an initial period of restoration, but then by statute allow indefinite hospitalization afterward, as long as the defendant meets civil commitment criteria: Alabama, Arkansas, California, Connecticut, Idaho, Indiana, Kansas, Kentucky, and Tennessee. Thus, 19 states appear to have no statutory limit on the length of time a defendant can be held after a finding of ICST. Of course, actual practice in these states may be different from what the statutes hold, due to policy decisions by either or both the criminal justice system and the state mental health authority. Nonetheless, it is still surprising to find so many states that appear to have ignored for more than four decades the clear intent of a decision of the United States Supreme Court.

Indiana is an example of how statutory appearances can be somewhat deceptive with regard to the time allowed for RTC in reality. Indiana promptly changed its statute regarding competence to stand trial after the Jackson decision, limiting the initial period of RTC to 90 days, with an additional 90 days if requested, after which the state mental health authority was required to request regular civil commitment. However, when Miller³ surveyed the state forensic program directors, Indiana was listed in the category of having no limit on inpatient treatment for RTC. How did this come to pass? From 1972 until 2010, the policy of the Indiana Division of Mental Health and Addiction was not to discharge an ICST defendant unless he was restored to competence or the charges were dropped. As a consequence, Indiana state hospitals always sought the renewal of the civil commitment of incompetent defendants, and it was always granted by the courts. In effect, unless a defendant was RTC, the discharge decision rested with the prosecutor; curiously, ICST defendants often had their charges dropped after a length of stay that was close to the time the defendant would have served if found guilty on the underlying charges. However, some ICST defendants were held longer than the maximum sentence when the prosecutor declined to drop the charges, which generally occurred when the charges were politically sensitive (e.g., sex offenses such as child molestation). Indiana's approach did not change until 2010, when the Indiana Supreme Court, in Davis v. Indiana, held that a judge could, in certain circumstances, dismiss the charges if an ICST defendant had been held longer as incompetent than the maximum sentence for the underlying charge.⁴

It is intriguing to see how the states have tried to solve the problem of what to do with defendants who are not restored in a fairly short time. In Indiana, the Davis decision did not put an end to the discussion about these defendants, as subsequent cases sought to clarify when ICST defendants could have their charges dropped. The most interesting iteration involved a defendant whose attorneys sought to have charges dismissed after he was found both ICST and unrestorable by the initial, pretrial evaluators. The trial judge in this case always included in her order to evaluate competence a requirement that the evaluator offer a prediction of the likelihood of RTC if the defendant was felt to be ICST. However, in Curtis v. State, the Indiana Supreme Court held that restoration must be attempted for defendants found ICST, as Indiana statute requires an assessment of the probability of attaining competence “within the foreseeable future,” or within 90 days of the start of RTC.⁵ Indiana statute does not require an assessment of the probability of restoration of an ICST defendant by the initial examiner, and, of the 43 state statutes that do require assessment of the probability of RTC, 24 defer the evaluation of restorability until after restoration has started (Table 1). Thus, in 19 states, it is possible for an ICST defendant to be declared unrestorable after the initial evaluation of competence to stand trial. Since all 19 of these states allow outpatient evaluation of competence and six require the initial evaluation to be performed on an outpatient basis, in each of these states, a defendant could be found unrestorable on the basis of a single, time-limited evaluation.

These various statutes lead us back to the question of how to determine when an ICST defendant is not restorable. One way is to declare him unrestorable when the statutory period of time allowed for RTC has passed. This approach is simple and clear and is one that has been adopted by several states, including eight that allow a total time for RTC of three years or less (Table 2). However, it ignores some important questions. For example, if the defendant truly has essentially no chance of successful RTC, is it fair to keep him under court jurisdiction for an extended period, either on conditional release or in a state hospital? Further, if a defendant who was declared unrestorable did not commit the offense, he will never have the chance to be acquitted, unless his case is in Illinois, where a hearing on acquittal is required if a defendant is still ICST after one year of RTC; New Mexico, where a hearing on acquittal is required if the ICST defendant is facing serious charges; or Delaware, South Carolina, and Wisconsin, where a defendant may request a hearing for acquittal after a finding of ICST. Conversely, if the defendant did commit the alleged offense, is it reasonable to dismiss the charges, no matter the severity, after one year, as is the case in Alaska, Georgia, New Hampshire, and Washington? In the end, the time-limit approach, although it is used by many states, simply dodges the question of whether the defendant truly is not restorable or just has not been restored after a certain length of time.

The other way to determine if a defendant is unrestorable is to rely on the opinion of a psychiatrist or psychologist. As mentioned, many states require competence evaluators to assess the likelihood of RTC either at the initial evaluation or after a period of attempted restoration. State statutes typically use the phrase substantial probability when referring to assessing restorability (i.e., whether there is a substantial probability that the defendant will not be restored in the foreseeable future or within the statutory limits). However, substantial probability is not defined in any of the state competence statutes, which gives the evaluator wide latitude when offering an opinion regarding likelihood of restoration. As far as I can tell from the literature on RTC, there is no agreed-upon definition of substantial probability, nor is there any consensus in the literature as to when a defendant can be called unrestorable or which disorders might qualify a defendant for this finding. In addition, even when a forensic evaluator writes that it is unlikely or very unlikely that a defendant will be restored to competence, court participants may well conclude that it is still possible the defendant can be restored and thus deny designation as unrestorable.

Since the states have not defined substantial probability for the purposes of predicting restorability, then perhaps forensic clinicians should try to do so. The closest our field has come to developing a consensus on RTC has been the 2007 AAPL Practice Guideline on evaluation of competence to stand trial,⁶ which recommended that evaluators consider whether a defendant's incompetence was due to a treatable condition (e.g., a knowledge deficit or an untreated but treatable mental disorder) or to an untreatable condition (e.g., a developmental disorder). The authors also recommended review of the defendant's history of response to treatment and current knowledge about treatment of the defendant's disorder(s), but offered no recommendations as to what degree of confidence would be needed to make a prediction of unrestorability, other than to observe that courts may view any probability greater than zero to be substantial enough to justify attempted RTC. The research literature on prediction of RTC was rather modest when the Practice Guideline was published, and it is only recently that consistent findings have begun to emerge in this area. In brief, there is reasonably strong evidence that defendants who have chronic psychotic disorders and a history of poor response to treatment, as well as defendants with mental retardation, have a significantly decreased chance of successful restoration.^7,8 However, there does not appear to be any consensus on how to use this information in a reliable way to determine whether an individual is likely or not to be RTC.

I believe that research is now needed on how evaluators decide whether and when a defendant is unrestorable, which would require a shift in the emphasis of research in this area. Researchers studying competence to stand trial and RTC, including myself, have largely relied on retrospective studies containing databases of the demographics and clinical characteristics of defendants who have been evaluated for competence, have been referred for RTC, or have not been restored after a period of RTC. Such studies have become more sophisticated over time, have been based on larger databases and analyzed with better statistical tools, and have yielded a better understanding of the demographic and clinical factors associated with successful and unsuccessful RTC. They have not answered the question of when and according to what criteria an individual defendant can be deemed unrestorable. It may not be possible to answer this question definitively, but if we have a better understanding of how forensic clinicians make this decision, our field will then have practical evidence on which to base our predictions. Such studies will necessitate either prospective examinations of the decision-making process of evaluators asked to determine the restorability of ICST defendants or the development of a database with a sufficient number of competence reports in which the reasoning for an opinion regarding RTC was laid out in enough detail to discern the rationale for the opinion. In addition to this, as Mossman observed, to do a truly thorough study on RTC, “it [is] desirable to have forensic examiners systematically document defendants' symptoms using structured instruments, to have treating clinicians use structured interviews when arriving at diagnoses, and to have degrees of improvement in competence quantified by using standardized assessment instruments” (Ref. 7, p 41). In the meantime, we will have to struggle individually with how best to answer questions posed by the courts and the law about the restorability of ICST defendants. Until forensic clinicians put forward a consensus professional judgment as to what substantial probability means and develop a reliable means of determining when a defendant is not restorable, forensic clinicians will be unable to answer satisfactorily the questions posed by the criminal justice system.

This dilemma reflects one of the important tensions underlying the field of forensic psychiatry. As we all have appreciated at one time or another, the interaction between the mental health and criminal justice systems is an uneasy one, fraught with tension, because forensic clinicians cannot always answer the questions the criminal justice system apparently needs to have answered. Mental health clinicians practice in a world of gray, while the courts require clear outcomes. The uncertainty and inconsistency of psychiatric diagnosis, which is often complicated by changing criteria and the frequent presence of comorbid conditions, has led to judicial criticism of the utility of psychiatric assessment.⁹ In addition, the treatments our field has to offer for these diagnoses are imperfect at best. Finally, individuals do not always fit neatly into diagnostic criteria and only sometimes respond to treatments as the research literature says they should. The criminal justice system, in contrast, functions in adversarial proceedings that demand definitive answers; a defendant is guilty or not guilty, competent or incompetent, and restorable or unrestorable. We offer written opinions and testify with reasonable medical certainty, which is understood to be a preponderance of the evidence, to try to answer the courts' questions in a relevant way despite the uncertainty of our database, and the legal system tolerates the ambiguity. Evidently, the legal system needs mental health clinicians to answer important questions, despite the imperfect match of philosophies, and so the two systems have, over time, worked out ways of collaborating to accomplish apparently mutual goals. One example of this is civil commitment. As summarized by Appelbaum,¹⁰ research on commitment hearings has shown that the participants in an adversarial proceeding in which important civil rights are at stake generally collaborate to achieve what they consider to be the appropriate outcome for the person with serious mental illness who is the subject of the hearing. Similarly, at least in Indiana, clinicians collaborated with judges for decades to ensure the continued civil commitment of ICST defendants to the state hospital for as long as the criminal charges were not dismissed.

If, on the other hand, we develop an evidence base that provides clear direction as to how to conduct a thorough assessment of the probability of RTC, as well as research that tells us which clinical conditions are associated with a substantial probability (also defined) of unsuccessful RTC, then we will be in a position where our relationship with the criminal justice system will be less tense and less ambiguous. Forensic psychiatrists have long tried to serve courts, attorneys, and other clients by providing clinical assessments for legal purposes, without a research base to show that what we offer is valid and reliable. Prediction of unrestorability is but one example of an area in which our field needs good research to support our practices. We have a long history of eminence-based practice¹¹; we must move into an era of evidence-based practice, difficult though that may be.

• © 2012 American Academy of Psychiatry and the Law