CURRENT STATUS OF THE FIELD

Research Directions

The first edition of Evaluating Competencies reviewed three instru­ments in this area, all of them characterized as "prototypes" because their authors considered them "experimental" and not yet ready for clinical or forensic use.

Functional Component

Weithorn had turned to the analyses of Roth, Meisel, and Lidz (1977) for the ability constructs that that she used in her competence to consent to treatment instrument. These were also the starting point for Appelbaum and Grisso (1988) as they began their work on the UTD, POD, TRAT, and eventually the MacCAT-T, and they were employed also in Marson's CCTI. This may indicate that researchers and legal analysts in this area have reached a consensus about the fundamental abilities that need to be assessed in competency to consent instruments: the abilities to express a choice, to understand information that is disclosed, to appreciate the significance of its consequences for oneself, and to reason with (or logically process) the information when arriving at a decision. Moreover, the vari­ous instruments suggest a consensus, with adequate legal foundation, about what a person must be able to understand and cognitively process for informed consent (the disorder, treatment, its risks and benefits, and alternative treatments and their risks and benefits).

A note of caution is in order, however, for persons who pursue fur­ther research with these instruments or who seek to develop new ones. First, Roth et al.'s (1977) original scheme included a fifth ability: that the patient can make a "rational choice" (that is, can arrive at the "best" decision, the one that most "reasonable" people would make). Marson's CCTI includes this concept. In contrast, it was specifically set aside by Appelbaum and Grisso (1988) because legal and ethical views subsequent to the Roth et al. seminal analysis have recognized that as a legal standard this concept is antithetical to the fundamental notion of autonomy. It sug­gests that individuals are competent if they choose what others would choose, whereas modern notions of informed consent allow individuals to make whatever choice they wish as long as they have the capacities to understand, appreciate, reason, and express a choice.

Second, as demonstrated by the previous reviews, the definition of "appreciation" is not the same across the measures that claim to identify it. The POD and MacCAT-T focus on what has been called "lack of insight" into one's disorder and need for treatment (Appelbaum & Grisso, 1995). This definition focuses on cases in which people do not believe that the information they are being given about their disorder applies to their own situation. In contrast, the CCTI focuses on lack of awareness of potential consequences of one's treatment. Moreover, the actual stimuli and scoring or rating criteria used in the POD and MacCAT-T suggest somewhat different definitions of appreciation. The POD scores apprecia­tion as poor if individuals fail to acknowledge their illness (or the value of treatment) even when given hypothetical that challenge the premises behind their beliefs, suggesting a rigid discounting of reality. The MacCAT-T, however, scores failure to acknowledge one's illness based on a specific reason for non-acknowledgement: when the patient has a delu­sional premise that distorts reality and does not have a basis in the patient' s cultural or religious background.

The differences in definitions of appreciation have several implica­tions. Future researchers who seek to measure the concept must be clear about their definitions, and comparisons between instruments will have to take into account the fact that different measures, though similarly titled, might not be conceptualized or operationalized to measure the same thing. Unfortunately, the law does not currently provide a suffi­ciently fine-grained definition to allow us to judge which of the instru­ments' definitions of appreciation is more legally relevant. The debate to watch in this regard may be that of ethicists more than legal analysts.

Second, the differences will influence research that examines the rela­tion of appreciation to the other psycholegal constructs (understanding and reasoning). Both the CCTI and the MacArthur research measures have been used in this type of research. Studies that compare patients' performance on measures of the various psycholegal constructs address questions about the necessity or redundancy of using all three constructs in clinical practice. They seek to address whether states that employ different legal standards are likely to find different types or proportions of patients incompetent to consent (Dymek et al., 1999; Grisso & Appelbaum, 1995b; Berg et al., 1996). But the results are likely to be different for studies that employ different conceptual and operational definitions of appreciation.

Finally, the first edition discussed confusion at that time concerning whether the law defined "understanding" as actual understanding (the condition of a patient's knowledge of his or her own treatment situation) or the capacity to understand (one's ability to understand hypothetical, treatment-like information).

The difference is important for the test developer, because assessment of actual understanding requires an instrument that uses the patient's own symptoms and treatment options as the test stimulus (like the MacCAT-T, as well as the SSCI now being developed by Marson: see introduction to the instrument review section of this chapter). In contrast, assessment of capacity to understand allows for a measure that uses one or two stan­dardized symptom and treatment vignette disclosures (like the CCTI and the three research measures).

Of course, there are substantial differences in what the researcher can confidently do with these two types of instruments from the point of view of standardized control that allows for inferences and comparisons involving groups of patients. Both methods allow for standardization of format, so that the same general types of information are provided to all patients and the same inquiries are made. But the use of hypotheticals allows the same content to be included in that format across patients. In contrast, instruments like the MacCAT-T and the SSCI (see the paragraph introducing the instrument review section) allow for the patient's own symptoms and treatment to be inserted in the format, thus producing nonstandardized content across patients. Examining patients' actual understanding of their own circumstances, therefore, requires a method­ological concession that weakens one's ability to make meaningful group comparisons. While this is a limitation, it need not deter researchers from using "actual understanding" instruments in research if they create condi­tions that will minimize those limitations. For example, the matter would be less critical in a study in which all of the patients at least had the same disorder (e.g., schizophrenia) but with the disclosure including the patient's own pattern of symptoms, compared to a study involving patients with a wide variety of psychiatric diagnoses.

Causal Component

The first edition urged future researchers to examine the relation between performance on competence to consent to treatment instruments and various psychological conditions that might impair abilities associ­ated with competence. This has been an integral part of the projects that spawned the instruments reviewed here. Performance on the various ele­ments of these competence to consent instruments have been examined for persons with various psychiatric and medical disorders (e.g., schizophre­nia, major depression, Alzheimer's Disease, and several non-psychiatric medical conditions), as well as persons with no disorders. Further studies with other patient populations (for example, persons with mental retarda­tion) would be especially helpful.

Most impressive has been the interest in examining the relation of scores on these instruments to performance on various neuropsychologi­cal measures that may account for deficits in the psycholegal abilities in question. Such studies are especially helpful in addressing the construct validity of the instruments, as well as providing information for clinicians who may use the same neuropsychological measures to explain the deci­sion making deficits of persons whom they are asked to evaluate in com­petence cases.

Research should also use these instruments to examine the decision­making capacities of children and adolescents. Early in the 1980s there was a brief surge of interest in the capacities of children and adolescents to make treatment decisions (e.g., Melton et al., 1983) among researchers who addressed psychological questions related to law and policy. This was stimulated in part by considerable interest among child advocates in pro­moting children's participation in decisions about their medical care. Policy issues that research in this area addressed ranged from youths' role in their own commitment to inpatient psychiatric facilities, to adolescent girls' capacities to make decisions about abortions.

Interactive Component

As noted in the introduction to this chapter, patients' competence to consent is judged not by their absolute abilities alone, but by the degree to which those abilities are challenged by the demands of their own treat­ment situation and the decisions they must make. Actual understanding instruments like the MacCAT-T and the anticipated SCCI (using patients' own symptoms and treatment options as the stimuli) will give more relevant data for judging patients' capacities in the context of the specific demands of their own situation. In the language of the interactive com­ponent of competence, patients' performance on instruments that use their own specific treatment situation is an index of person-situation congruency.

This does not mean that instruments using hypothetical situations in vignettes have no role in research on the interactive nature of competence to consent to treatment. The interactive, person-situation construction of legal competencies reminds us that people whose capacities are marginal may nevertheless be considered competent if we can manipulate the cir­cumstances of their consent process, such that the level of demand is decreased and/or their understanding and processing of information is increased. This strategy is called "assisted competence" (Grisso & Appelbaum, 1998). For example, people with marginal abilities to under­stand or reason about treatment information may be assisted by visual aids, simplifying the wording of the disclosure, having family members or companions assist them in thinking through the alternatives, and other strategies for manipulating the environment in ways that might incre­mentally compensate for their limited capacities. To date there have been few studies of the degree to which such strategies can be used to improve patients' understanding and processing of treatment information in informed consent situations. The instruments reviewed here that use hypo­thetical vignettes would be entirely suitable as pre-and-post measures to examine the effects of compensatory strategies on the improvement of patients' abilities to make treatment decisions.

Judgmental and Dispositional Components

The MacCAT-T and the CCTI have set a fortunate precedent by not summing patients' scores across the various psycholegal scales. To do otherwise, producing a total "competence" or "capacity" score, would not be logical. It would suggest, for example, that a high score in understand­ing can compensate for a low score in reasoning, despite the fact that understanding of information is useless if one cannot process the information rationally to arrive at a decision.

Research with the HCAT used a cut-off score to signify probable incompetence, setting the score on the basis of empirical comparisons of HCAT scores to independent clinical judgments of competence. The devel­opers of the MacArthur instruments and the CCTI chose not to develop cut­off scores, but rather to signify a higher level of risk of incompetence by using various points of comparison. In their studies with the UTD, POD and TRAT, the researchers used the range of scores below the minus-2 stan­dard deviation point for the total distribution of scores (combining patients and non-patients) as an index of impairment. The CCTI authors did the same but used their non-patient samples alone to establish the minus-2 standard deviation point. Such indexes are helpful in expressing degrees of serious impairment, as long as they are not translated automatically into statements that individuals below those scores are "incompetent."

The use of "normal" comparison groups is especially helpful and consistent with the law's way of thinking about legal competencies. The "average adult" is considered competent to decide most things. We know, however, that even "average adults" manifest limitations in their under­standing and reasoning abilities. Using their imperfect abilities as a stan­dard guards against penalizing patient samples simply because they, too, manifest some deficits in those abilities. Their deficits should be substan­tially greater than the "average" person's abilities in order to raise the question of their competence.

Clinical Application

Description

The assessment of patients' competence to consent to treatment is a routine obligation for every health and mental health professional. Typically it is assessed by default; nothing occurs in the normal course of the doctor-patient interaction to even raise the question. In this sense, the clinician is obligated simply to be vigilant for cues that might reasonably raise the question of competence (Grisso & Appelbaum, 1998a).

Once a cue does raise the question, typically a full "forensic" evalua­tion of the patient's competence to consent still will not be required. A more informal process will evolve in which the clinician will explore the possibility that the initial cues need to be taken more seriously. If they do, it is at this point that some cases may call for the use of a standardized

process represented by the instruments in this chapter. The instruments may be used by the attending clinician or by a specialist with whom the clinician may consult (e.g., a forensic psychologist, or a psychiatrist pro­viding consultation-liaison to various medical departments in a hospital).

The value of these instruments is especially great when it may be important to document the data and reasons for one's judgments about a patient's competence and to clearly describe those matters to others, be they colleagues or courts. This may be necessary if the clinician intends to pro­ceed with a treatment, perhaps on the family's agreement, after deciding that the patient is incompetent. In many states, doing this requires the deci­sion of a court, and the instruments can be especially useful for describing one's data and logic in those cases because of the instruments' relationship to the legal and ethical standards for competence to consent to treatment.

Concerning the selection of instruments, it is difficult to imagine many clinicians employing the UTD, POD, and TRAT together in their assessments of patients' treatment decision making capacities. They require considerable time, and the TRAT's use of card-sorts is not particu­larly amenable to bedside use. The HCAT cannot be recommended for descriptive objectives of these evaluations, because the data that it obtains simply are not relevant in most cases. The HCAT allows one to describe what a person can understand about the concept of informed consent, but it does not provide evidence for patients' abilities to understand their dis­order or the nature of the treatment that is being proposed.

In contrast, the CCTI and the MacCAT-T require no materials other than the examiner's clipboard and recording forms, and they require only about 20 minutes on average. They provide information relevant for the full range of abilities with which legal and ethical analyses of competence to consent to treatment are likely to be concerned. The CCTI does this for the patient's capacities to respond to hypothetical rather than the patient's own disorder, but it offers normative points of comparison for expressing the patient's abilities in relation to others. The MacCAT-T allows one to describe patients' actual understanding and processing of their own disor­der and treatment situation, which is an advantage for presentation to courts. However, opportunities for comparing the patient's performance to that of others is not as good as with the CCTI, because the use of the patient's own disorder limits standardization and the opportunity for true norms to have been developed.

As noted earlier in this chapter, not all states use all four of the psyc- holegal constructs represented in these two instruments. This need not deter their use by clinicians who must often make their own decisions, with­out legal intervention, about patients' competence to consent. Butexaminers who are performing competence evaluations for courts' determinations of patients' competence to consent should be aware of the standards that apply in their own state. While they may administer all of the instrument in question, only some of the information might be considered legally relevant in their jurisdiction.

Explanation

Clinicians will need to explain the reasons for patients' deficits on the instruments. These may occur for a wide variety of reasons, such as symp­toms of psychotic disorders that can affect cognitive and emotional func­tioning, neuropsychological disorders, mental retardation, dementias, and temporary but extreme symptomatic conditions sometimes associated with medical conditions (e.g., trauma-related anxiety and compensatory defense mechanisms to cope with stress). Often these conditions are obvious and within the boundaries of discovery in ordinary clinical observation. At other times they may require confirmation with additional psychological and diagnostic tests.

Prediction

The measures described in this chapter do not have a predictive objective. They seek to describe what people can do now, at the time they need to make a treatment decision. However, sometimes courts will want to address whether a person may be able to make future treatment situa­tions. In such cases, the instruments themselves provide no basis for making that prediction. The clinician's attempts to do so would be based primarily on an assessment of the cause of the patient's deficits in per­formance on the instruments, some of those causes logically being more enduring or remedial than others.

Conclusions and Opinions

The instruments reviewed here obviously cannot define legal compe­tence or incompetence. This is true whether the clinician uses the data from the instruments in the course of making a clinical decision or offers the data in a court where a legal decision about competence or incompe­tence will be made. The data from some of these instruments certainly provide relevant information for making the competence decision. But additional clinical data as well as the application of moral and legal considerations will be required to make the final judgment concerning whether the patient's incapacities, and the consequent risks, are sufficient to curtail his or her right to make the treatment decision.