An evaluation of functional mental capacity in forensic mental health practice: the Dundrum capacity ladders validation study

Functional Mental Capacity evolved first as a legal concept, arising from the interplay of expert evidence by clinicians and legal judgements and precedents, followed in some jurisdictions by legislation. The law identifies different competencies e.g. testamentary capacity, competency to marry, fitness to plead and stand trial. Decision-making capacity is but one element in the legal classification of competence. Decisional Capacity relates to a person’s ability to perceive, retain and understand information pertaining to the choices at hand, and to use this information to reason about and appreciate the consequences of the choice they may make. Capacity relates to a decision about a person’s capability to carry out a specific act or set of acts. Legal Competency relates to a decision made by a judge about whether a person, under the law, has or does not have the capability to carry out a specific act or set of acts. In this instance the clinician will provide the court with information summarising the person’s decision-making skills and offer an opinion on how these findings may affect the person’s abilities in a specific area. The judge then uses this information, together with legal factors to arrive at a finding of competence or incompetence.

A review of case law literature by Grisso and Appelbaum (1998) [3] resulted in the so-called ‘four abilities’ model in the assessment of functional mental capacity, namely; the ability to understand the information pertinent to the decision at hand, in keeping with the relevant facts of the decision; the ability to reason with this relevant information so as to engage in a logical weighing and comparison of the positive and negative consequences; the ability to appreciate the significance of the information for one’s own situation; and the ability to make and communicate a decision relating to the decision at hand.

Historically, legislative approaches to incapacitated persons were paternalistic. These approaches have been criticised as not being in keeping with the spirit of the Universal Declaration of Human Rights [4] and more explicitly in the United Nations’ Convention on the Rights of Persons with Disabilities, Article 3 (a) [5] which states: “The principles of the present Convention shall be: Respect for inherent dignity, individual autonomy including the freedom to make one’s own choices, and independence of persons.”

In most jurisdictions, modern legislation now includes specific statutory means of protecting the rights of people with mental disabilities, as seen in New Zealand, Ontario, Scotland and England and Wales. All assume presence of mental capacity unless established otherwise. However, all have slightly different definitions for the test of mental incapacity – some require the presence of mental disorder as a first step to define incapacity – as seen in Scottish law [6], and others view capacity under separable domains [7‐9]. Newly published Irish legislation has also changed to a functional approach for capacity assessments [10]. In this article, we are primarily concerned with the clinical assessment of functional mental capacity. We regard this as a continuum from fully capable to seriously impaired, like any other neurocognitive mental ability or any physical ability. We are not primarily concerned with legal standards or thresholds for competence or incompetence.

The assessment of functional mental capacity is one of the most ethically complex tasks which clinicians are called upon to perform on a routine basis. In every case a patient’s autonomy and right to self-determination must be weighed against a clinician’s judgment that as a consequence of illness, mental disorder or intellectual disability, they may not be capable of acting in their own best interest. Clinicians have an ethical and legal responsibility to demonstrate how an illness, mental disorder or intellectual disability may compromise an individual’s right to autonomy and self-determination when faced with a specific situation. However the reasons for doing this go beyond the assessment of legal competence. Clinicians must assess responses to treatment for many purposes. It is increasingly recognised that functional abilities are the most important outcomes of treatment for mental disorders such as schizophrenia [2], and functional outcomes are for practical purposes more important than symptomatic improvement. Neurocognitive abilities, social cognition and metacognition all play a part in influencing symptoms and functional outcomes (Fett et al. [11], Lam et al. [12], Gallagher & Varga [13], Lysaker et al. [14]). Indeed deficits in metacognition may be one of the distinguishing features between schizophrenia and other debilitating physical illnesses (Lysaker et al. [15]).

In practice the assessment of functional mental capacity can be complex and challenging. The complexity arises not only because of the balancing of responsibilities, but because functional mental capacity is inextricably linked with contextual and situational factors, and also because it is underpinned by a number of sequential cognitive process. In contrast to illnesses, mental disorders or intellectual disabilities, all of which may be static in nature, functional capacity is dynamic and refers to the extent to which patients can apply their abilities to a particular situation and context independently as well as when offered assistance. The cognitive processes which underpin functional capacity include comprehension, appreciation, reasoning, and judgment and difficulties may occur at any step of this process (Grisso and Appelbaum) [3].

The potential gravity of recommending that a court find a person incompetent – i.e. denying that person the right to an autonomous choice, means that assessments of functional mental capacity can be lengthy and onerous. In such an assessment, the clinician must take account of the patient’s mental illness (or otherwise), cognitive function and social attitudes, as well as collateral information from a reliable informant where possible. Furthermore, the information shared by the clinician during the assessment of functional capacity should be such that it is easily understood by the person whose mental capacity is being assessed. This holds true even when the assessment is not for a medico-legal purpose.

Although psychological science has produced objective measures of cognitive capabilities, which can be usefully applied to capacity evaluations, it would be wrong to determine capacity solely on the basis of cognitive testing. This is because cut-off scores for a particular individual faced with a particular competency question can never be determined in advance. Moreover, the results of cognitive assessments are proxies and thus one step removed from the ultimate issue of whether a patient can understand, reason, appreciate and make and communicate a decision based on the unique demands of a particular situation. All of these factors can be more holistically and directly assessed using clinical interviews with the patient and informants. However, in contrast to cognitive assessments which have established psychometric properties, the reliability and validity of unstructured clinical interviews is unknown and unmeasurable.

Research by Volicer and Ganzini [16] showed that psychiatrists, psychologists and geriatricians do not adhere to a uniform approach for the purpose of assessments of functional capacity. Their work supports the idea that physicians’ competence assessments are often subjective and inconsistent (Marson, McInturff et al.) [17].

Case law [18, 19] internationally has increasingly placed emphasis on the patient’s right to all material information pertaining to a particular proposed treatment, in order to make an informed decision. However, if a person is suffering from a psychotic illness, it is foreseeable that sharing too much information about treatment options may overburden the person’s capacities to make an informed treatment decision, as demonstrated by Kennedy, Dornan et al. [20]. It follows that decision-making in other domains would be similarly affected. Furthermore, other illnesses such as dementia would have similar deleterious effects on decision-making ability.

There are a number of structured professional judgement tools which have been developed for assessment of functional mental capacity, the majority of which are based in Decision Theory [21]; that is identifying the values, uncertainties and other issues relevant in a given decision, its rationality, and the resultant optimal decision. Perhaps the best-known of these in a forensic arena are the MacArthur Competence Assessment Tools for Consent to Treatment [22] and Fitness to Plead [23]. In a psycho-geriatric setting, other tools have also been devised, e.g. The Hopkins Competence Assessment Tool [24]. Fazel, Hope and Jacoby [25] have also published a structured approach to the assessment of mental capacity to complete advance directives, emphasising a patient centred approach. A key feature of these assessment tools is that they provide clinicians with a structure for carrying out their own interviews thus improving transparency, reliability and objectivity. For example, in addition to using a validated functional capacity interview in a complementary and concurrent fashion, clinicians can use capacity instruments as a template to structure their own judgment by substituting the specific idiosyncratic question facing their patient into the structure of the instrument. Used in this manner a structured professional judgement approach to assessing functional mental capacity capacities could facilitate clinician’s ability to communicate their opinions in legal settings and to reduce the potential for idiosyncratic differences between experts. More importantly, the assessment of functional mental capacity as a treatment outcome can become a reliable and valid measure. One recent study employed the MacArthur instruments as outcome measures in a controlled trial of cognitive remediation therapy for psychosis (Naughton et al. 2012) [26]. In addition to improved transparency of clinical decision-making, functional capacity instruments are useful to identify targets for treatment and to measure change. The extent to which functional mental capacities change over time is of great importance when considering the legal and human rights protections necessary for mentally incapacitated patients, including those detained and treated under mental health legislation.

Although functional capacity assessments are often specific to the decision in hand, in practice clinicians are most often required to make decisions regarding a patient’s capacity to manage their finances, to consent to medical treatment and to make welfare decisions concerning living independently. Currently there is no single instrument to assist clinicians when assessing functional capacity in these three domains. No single instrument has a framework that is readily transferable to idiosyncratic situations. Because of the routine nature of these tasks a valid and reliable functional capacity assessment in these domains would be particularly useful to help clinicians structure their judgments.

In examining these tools, we noted the need for a different form of rating of functional mental capacity, in order to take account of its continuous rather than binary form.

We aimed to establish validity for a structured professional judgement tool to assess functional mental capacity in three domains – finances, welfare and healthcare. In this paper we present results of inter-rater reliability, construct validity (in the form of internal consistency) and criterion validity as compared to neurocognitive function. We wanted to rate each of these using a common decision-making structure for understanding, reasoning, appreciation and decision-making. We wanted these to be structured in such a way that even the most severely ill could be rated and that ratings were on a continuous scale rather than a dichotomous one. We have also examined the relationships between this measure of functional mental capacity, real world function and neurocognitive function.

We hypothesised that patients currently cared for in more highly structured and supportive hospital environments of higher therapeutic security would score lower in terms of functional mental capacity than those who had progressed further along the therapeutic recovery pathway within the hospital. We also hypothesised that scores of functional mental capacity would correlate with scores of occupational, social and symptomatic function; and neurocognitive function.

The Central Mental Hospital provides high, medium and low therapeutic security and community follow-up mental health services for a population of 4.6 million [27]. At the time of the study there were eighty-two male in-patient beds at varying levels of therapeutic security, and sixteen patients under community supervision. Patients can be admitted to the hospital from prisons under the Criminal Law (Insanity) Act 2006 [28], if medically certified. They can also be admitted from the courts if found unfit to stand trial or not guilty by reason of insanity. Additionally, the service provides a tertiary referral and assessment service for local community mental health teams and admits patients under the civil Mental Health Act 2001 [29] on transfer from local psychiatric hospitals.

Male patients in the Central Mental Hospital are admitted to a high secure admission ward. From there they can move to a series of medium secure units and finally to low secure and pre-discharge units (Pillay et al. 2008) [30]. This is a patient-centred approach, as each service-user is placed at an appropriate level of therapeutic security according to their individual need. These placements correspond to levels of risk, symptom severity and the patient’s overall level of functioning. Location at the time of assessment was ranked according to the level of therapeutic security – high, medium or low. This is an ordinal ranking according to the staff-to-patient ratio, and the ranking system matches the position of each unit along the therapeutic pathway as service-users progress from admission to discharge. The units are arranged in clusters by level of therapeutic security – acute cluster (high secure), medium cluster (medium secure) and rehabilitation and recovery cluster (low secure) [30].

Three vignettes were drafted in simple language to assess functional mental capacity in three domains – finances, welfare and healthcare. The vignettes were designed to assess patients’ abilities to: understand and retain the information relevant to the dilemma in each vignette, reason through the dilemma, appreciate the need for a decision and make and communicate a decision in each vignette.

A stratified scoring system from zero to one hundred was designed for each of the four domains understanding, reasoning, appreciation and ability to make and communicate a decision with definitions tethering the ratings at every ten points so that scores could be assigned for each of the three vignettes.

The vignettes were then reviewed by three focus groups, including senior nursing staff, senior social work staff, occupational therapy, consultant forensic psychiatrists, clinical psychologists and patient advocates, and suggested alterations were made accordingly.

Ethical approval for the study was sought from and granted by the Research Ethics, Audit and Effectiveness Committee of the Central Mental Hospital.

Patients were given written information about the study, and all participants who were assessed by their treating psychiatrists as capable of giving consent gave written informed consent to participate.

All male patients with a diagnosis of schizophrenia or schizoaffective disorder using Structured Clinical Interview for Schizophrenia (Structured Clinical Interview for DSM-IV (SCID; First et al. 2002 [31]). resident in the Central Mental Hospital or under community supervision by the forensic recovery and rehabilitation service during the study period (January 2013 – January 2015) were considered eligible to participate. Female patients were excluded as this group makes up a very small proportion of forensic mental health service-users in this jurisdiction. Male patients who did not have English as a first language were excluded, as were those who were deemed too behaviourally disturbed to participate or unable to give consent as assessed by their treating consultant psychiatrists.

In order to rate the Dundrum Capacity Ladders (DCL), GM interviewed all patients who met inclusion criteria and consented to participate. Their responses were recorded in writing, and assigned a score on the ladder scoring system. For measures of inter-rater reliability, GM and JOC took a random sample of completed interviews and scored them using the DCL system. Data regarding contemporaneous level of therapeutic security, diagnosis as confirmed by treating consultant psychiatrist, and a validated measure of function, scored by the patients’ treating consultant psychiatrist, the MIRECC-GAF [32] was also recorded by the primary researcher. GM and JOC were blind to the ratings of the MIRECC-GAF and MCCB.

As the MIRECC-GAF and Dundrum Capacity Ladders have different maximum scores (300 and 1200 respectively), the total mean scores of each were corrected to a common total, 100.

Cognitive functioning was assed using the MATRICS Consensus Cognitive Battery [33] (MCCB) by Masters level assistant psychologists trained in its use. The MATRICS battery covers seven cognitive domains: processing speed; attention/ vigilance; working memory; verbal learning; visual learning; reasoning and problem solving and social cognition (assessed using social reasoning tasks for understanding and managing emotions taken from the ‘managing emotions’ subtest of the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT) which is a social reasoning test). The test comprises of vignettes of various situations, and options for coping with the emotions depicted in these vignettes. Participants are required to indicate the effectiveness of each solution ranging from one (very ineffective) to five (very effective). Overall performance on the MCCB is measured using a composite score which is the preferred outcome for treatment studies. Because there is growing awareness that non-social and social cognition are separable dimensions, the MCCB scoring system now provides an option for a neurocognitive composite that does not include the social cognition sub-scale. In validation studies, and in antipsychotic trials of stable patients, the MCCB demonstrated excellent reliability, minimal practice effects and significant correlations with measures of functional capacity. Those rating the MCCB were blind to the ratings of the Dundrum Capacity Ladders and also blind to the ratings of the MIRECC-GAF.

All data was entered using SPSS-22. Internal Consistency was assessed using Cronbach’s Alpha. Inter-rater Reliability was assessed using Spearman’s Rho and the Intra-class Correlation Coefficient. The relationships between Capacity Scores, function (MIRECC-GAF), Level of Therapeutic Security, age, length of stay and neurocognitive ability were assessed using Spearman’s Rho for non-parametric correlations. We took r values of 0.1 to 0.3 as small effect sizes, 0.3 to 0.5 as moderate effect sizes and r values greater than 0.5 as large effect sizes [34]. The relationship between Dundrum Capacity Ladder scores, MCCB scores and GAF scores was examined using regression analysis.

Forty-one patients’ neurocognitive and social cognitive abilities were assessed using the MATRICS Consensus Cognitive Battery. Spearman’s rho was assessed for non-parametric correlation between subscales of the neurocognitive battery, length of stay, age, MIRECC-GAF, and the three subscales of the DUNDRUM Capacity Ladders in a post-hoc analysis. Length of stay did not correlate with cognitive function. Age was found to correlate inversely with processing speed and reasoning. The GAF and all its sub-scales correlated with each of the domains of neurocognitive function. This relationship is shown in Table 3.

There were significant correlations between the Dundrum Capacity Ladders sub-scales and those in the MATRICS battery. This finding was replicated when examining the relationship between the MIRECC-GAF, its sub-scales and the Dundrum Capacity Ladders, as shown in Table 4. All correlations significant at the 0.01 level were in the range r = 0.358 to 0.693, effect sizes that were moderate to large.

The relationship between current place on the recovery pathway from acute / high security to medium security and minimum security was tested using ANOVA (Table 5). Post-hoc Bonferroni corrections showed that the mean GAF for the minimum security rehabilitation and recovery cluster patients was significantly better than the acute cluster patients, while for the DCL, each cluster differed significantly from each of the other clusters. Age did not differ significantly, but length of stay did differ as expected. The mean total GAF score and mean total DCL scores differed significantly, with mean DCL scores differing significantly at each step. It is notable that in the acute cluster, patients were severely impaired (GAF mean = 47.3 S.D. =16.2, DCL mean = 18.6 S.D. =26.1) while those in the minimum security / rehabilitation and recovery stage functioned significantly better (GAF mean = 72.7 S.D. =14.2, DCL mean = 78.8 S.D. =21.1).