CURRENT STATUS OF THE FIELD

Research Directions

Functional Component

The following discussion relies on the instruments reviewed in the second section of this chapter to examine several issues in the develop­ment of methods for assessing the abilities of disabled persons in guardianship cases.

Table 4. Content Areas on IADL and FAI Instruments

SELECTING FUNCTIONAL ABILITY CONSTRUCTS.

The various instruments reviewed in this chapter exemplify three dif­ferent ways to identify the ability constructs relevant for describing every­day functioning of disabled adults. The MAI was based substantially on a conceptual model that defined domains of functioning—ranging from physical to social—and organized abilities within each domain in order of complexity. The AFABS, EPT, DAFS, and MFAQ derived their domains from a logical synthesis of gerontological literature on intellectual abilities (especially the EPT) and functional abilities, and the clinical experience of selected experts, as were the FCI and HCAI, with particular attention to issues of legal capacities. Selection of DIG and ILS dimensions was based on an empirical consensus of a wide range of groups, including the elderly and clinicians working in clinical and legal contexts.

How should future test development proceed? Developers of similar instruments in the future should not automatically select any one of these methods over the others, because none is inherently best. Each has its own values and limitations. Instruments based on a conceptual approach have a clear advantage over the other two approaches in that their results can then be interpreted in light of the conceptual model itself. On the other hand, concepts and theories sometimes confine thinking, so that the prod­uct might not be as responsive to real-world concerns and phenomena as an instrument based on a logical synthesis of practical, clinical experience. Clinical notions about what to measure, however, may be open to bias or narrowness of perspective. This can be mitigated by an empirical approach to gaining consensus among a broader range of individuals for whom the instrument will be relevant.

A fourth empirical approach, however, would offer advantages not afforded by any of the three approaches demonstrated in the current instruments.

Two methods would seem to be suitable for pursuing this objective. The more costly and difficult of the two would require researchers to observe, record, and categorize the specific functions performed by individuals in various settings (e.g., residential options for disabled eld­erly or nonelderly adults) requiring different degrees of independent functioning. These results would define the conceptual dimensions for subsequent instruments, each instrument having been designed to assess abilities relevant for functioning only in a particular type of setting.

The second method, exemplified by the initial investigations of Scheidt and Schaie (1978), would use survey procedures to determine the frequency with which residents in various living arrangements encounter various possible environmental demands (e.g., dealing with broken appli­ances, cleaning one's residence, responding to emergencies). The more frequent demands of various settings could then be translated into "criti­cal behaviors necessary to meet these demands" (Scheidt & Schaie, 1978, p. 85). Both the behavioral observation and survey-of-demand approach would provide a more direct, empirical base for the selection of functional-ability domains of measurement than the three approaches in the currentinstruments.

Whatever approach is used, the relevance of the instrument's con­cepts and scales for legal decisions about the need for a guardian will depend on the degree to which the use of the instrument in legal cases was considered at the outset.

Merely surveying judicial concerns may not be sufficient to determine which functional abilities are most relevant for legal guardianship cases. For example, although the ILS began with such a survey, this did not result in money-management content that might be relevant for assessing a person's capacity to manage larger estates. Test developers need to consider the range of case circumstances that actually confront legal deci­sion makers. Unfortunately, there has been no empirical research that describes specific types of guardianship cases typically seen in probate courts. Without this information, researchers are at a disadvantage in con­structing assessment instruments of relevance for that decision-making process.

In general, more study is needed to delineate the conceptual coher­ence of specific functional tasks and the relationship of performance on these with various criteria for "competent" functioning. In considering "capacity to care for self," it is unclear what items should be incorporated, and whether these should focus on direct behavior, reasoning about everyday activities, or higher level abilities (e.g., memory) important for safely carrying out everyday activities. It would be helpful to know how performance on different tasks relates to each other. Does an individual's abilities to prepare a meal relate to abilities to clean a house? Do different everyday activities rely on different cognitive abilities? Does the capacity to independently and prudently manage finances rely on different cogni­tive abilities then the capacity to independently and safely drive a car? There has been decades of research exploring the "structure of intellect," such as Guilford's work on model's of intelligence (1967), Cattell's work differentiating fluid from crystallized abilities (1963), and Schaie's (1994) and Baltes (1987) work on the course of intellectual development across the lifespan.

methods for assessing abilities. The various instruments manifest several different response modes for indexing functional abilities. Each of these offers different sources of potential error in relation to external criteria.

The MFAQ asks examinees to report whether or not they "can do" particular things. Some individuals, however, might not have had to per­form the function in question for some time (e.g., housecleaning, if they have hired a maid for several years) and may not be aware that their abil­ities have diminished over the years. In addition, some elderly individu­als might have a need to report being able to do things that in fact they cannot do. The MAI potentially mitigates these problems by asking not only what examinees "can do," but also whether or not they now perform the function. Nevertheless, this method is still open to error in self-report of one's performance.

Asking examinees to demonstrate the ability in question, as do many of the DAFS, EPT, and ILS items, avoids self-report error. Nevertheless, this method requires attention to another question of generalizability— whether individuals might sometimes perform the function better or more poorly in the artificial test situation than in the real world.

Another approach, used on the AFABS, is to ask individuals who are familiar with the examinee's everyday functioning to describe or rate it. This method, though, is highly dependent on the integrity and motivation of the informant. In guardianship cases, relatives who are close to the examinee may be the petitioners; they might be motivated to accentuate the individual's specific inabilities, consistent with their general opinion about the need for guardianship. The examiner's direct observation of examinees in their natural setting, also encouraged with the AFABS, may be possible in institutional or nursing home environments; but this may be more difficult when the examinee is living in a private setting.

criterion validity. The most common strategies for validation of the instruments reviewed here have been the comparison of their results to clinicians' ratings and comparison between groups of elderly individuals living in settings requiring various degrees of independence. These are important criteria, but they do not directly address the question of gener­alizability of test results to the real world. Mental health professionals and legal decision makers who will use the instruments and their results need to know whether performance on the test corresponds to functioning in everyday life. For example, one needs to know whether the behavior sampled in the ILS Money Management items provides results that correspond to the same levels of performance in the actual management of money in everyday life. Research efforts to validate these instruments eventually must turn to the use of behavioral criteria in examinees' natu­ral environments, rather than relying solely on clinical judgments and group classification as external criteria.

Another important validation strategy is to compare multiple instru­ments designed to measure the same abilities with one another in multi­ple homogenous subgroups. Given the expansion of the number of tests to measure everyday functioning, a critical next step of inquiry is to test the same sample with multiple measures intending to measure capacities in multiple areas (home, health, finances, etc). Multi-trait multi method analyses would serve to validate the instruments and the concepts.

psychometric properties. Many of the instruments presented have good reliability. Additional information on inter-rater reliability is needed for the AFABS and EPT, and on test-retest stability is needed for the AFABS, MFAQ, and DIG. Normative data are critically important to pro­vide information on the performance of healthy samples against which to compare the performance of impaired individuals to get a sense of the rel­ative deficit of any individual. Normative data are needed for the AFABS, DAFS, DIG, and HCAI. Furthermore, information on the performance of healthy individuals is important in establishing the meaningfulness (validity) of the test and concepts. If healthy people do poorly or change over time, concepts or their measurement need to be re-evaluated.

Many of the instruments demonstrate considerable ingenuity in the development of systems for combining item scores to obtain summary indexes of functioning. Many of the summary score systems will be quite useful in research for validating the instrument. On the other hand, researchers and test developers should be aware that summary scores will be of limited use in applications of instrument results to legal guardianship cases. An overall impairment score, for example, does not communicate a person's strengths and weaknesses on specific areas of functioning within the test. The latter differentiation will become increasingly important in crafting limited guardianships. Therefore, validation research should not focus entirely on summary impairment scores at the expense of validating individual items or subsets of items.

Causal Component

Current research with the three instruments does not adequately prepare examiners to interpret the reasons or potential explanations for deficient performance on the instruments. For example, interpretations about causal relations between performance on these instruments and brain pathology or psychological symptoms would be aided by more research on the performance of individuals with known types of pathol­ogy. Research on the relation between results on these instruments and results on neuropsychological assessment instruments would also be helpful (e.g., see Dunn, 1984, in the ILS review). The inclusion of neu­ropsychological testing, as well as other data (such as medical record information), is necessary to establish causal links.

Some elderly individuals might manifest relatively poor perform­ance in certain areas of functioning not because of an inability to perform the function, but because of inexperience or a reversible atrophy of skills. Kane and Kane (1981), for example, point out that some residential set­tings for the elderly do not encourage the maintenance or development of "IADL skills," producing a risk that they will lose these skills perma­nently. On the other hand, some individuals may have become deficient in certain skills yet be capable of relearning them relatively quickly under proper conditions (Willis, 1990). Poor performances on the functional abil­ity instruments, therefore, should not automatically be interpreted as fixed incapacities. This observation has legal relevance, in that placing legal control of an individual's life in the hands of a guardian might seem unjust when the individual may be capable, with minimal assistance, of assuming self-control.

Examiners, however, have little research to guide them in determin­ing whether a current performance deficit suggests an irretrievable loss of function or a reversible deficit. Further, existing instruments do not provide methods for examining these alternative interpretations. One approach to developing such methods would be to devise supplementary test procedures designed to teach the skill in question to the examinee, then to reassess performance. For example, if an examinee performed poorly when locating a name in the telephone book (an item in the ILS), a supplementary procedure could instruct the examinee how to perform this task, then retest at a later time with a different name to locate. Thus, the development of a "learning capacity" supplement to an instrument like the ILS might be of assistance to examiners in interpreting the nature and implications of performance deficits on the instrument.

Overall, forensic examiners should bear in mind that while new statutes focus on "functional" definitions of incapacity, accurate diagnosis of the "disabling condition" is still an essential part of the forensic evalua­tion, especially in guardianship. Diagnosis will often be the best indicator of prognosis and will guide guardianship decisions, which focus on more than the competency of the individual for a single task at a single point in time (i.e., competency to stand trial), but on many tasks over an indefinite period of time. A diagnosis of dementia of the Alzheimer type would suggest the incapacities will progress; a diagnosis of alcohol dependence could mean the incapacities might remit if the drinking stopped.

Interactive Component

A disabled individual's functional deficit may be of greater or lesser consequence depending on the person's current living arrangements, family supports, financial estate, and other environmental or situational circumstances. This has several implications for the development of instruments like those reviewed earlier.

First, test developers should not encourage the use of cutoff scores for determining a person's impaired or unimpaired status. Absolute perform­ance criteria of this type may be useful for various research purposes. Yet impairment is a relative rather than fixed condition, depending in part on the demands of the examinee's own environmental circumstance. This is recognized not only in legal reasoning (see Interactive Component, in the first section of this chapter), but also in the ecological perspectives of gerontologists themselves (see the same discussion noted above). Kane and Kane (1981), for example, discuss the fact that "not all persons are required to manage skills at the same level of complexity... Some have more of the advantages of modern devices and human help than do oth­ers" (p. 57). They note also that certain geriatric rehabilitation settings in England evaluate competence "by performance relative to the actual home environment rather than by an absolute set of skills" (p. 57).

Second, this view suggests the need for ways to describe the demands of disabled individuals' specific environments relative to their own level of functional abilities. The MAI and MFAQ provide certain types of informa­tion that may be used to characterize the physical demands and social sup­ports in an elderly examinee's current environment. Part B of the MFAQ may be especially helpful in determining the supportive services currently at work in the examinee's life. Nevertheless, the ideal assessment system would provide a method for assessing and describing the examinee's environment on dimensions that directly parallel each of the functioning domains on which the examinee has been assessed. The ALSAR and CARE instruments attempt to measure resources for each capacity in question. They were not reviewed here since they have been the subject of limited study and valida­tion. Clearly more research on these instruments is warranted.

Another key area for research is the individual-family dynamic in the interactive component. Although capacity assessments focus of the abilities of the identified individual, competency concerns often arise in a family system. Some family members support their parents' and grandparents' autonomous decision making and only step-in via guardianship when it is really necessary to protect the individual and estate from real danger to the individual (i.e., financial exploitation by a scam-artist that would result in the loss of a home for the elder). Unfortunately some guardianship petitions are from families with decades of conflict and the petition may really repre­sent a desire of an adult child or other relative to gain financial control for their own (the "fiduciary's" own) current and future financial needs. Drug addiction and excessive credit spending are among the habits that may lead adult children to look at compromised parents as a source of income; cohort differences whereby currently older cohorts see themselves in a role of sacri­ficing for their children may also shape the spending decisions and expecta­tions of elderly parents and their adult children. Determining when to respect the beliefs and norms of individual families versus when to intercede in situations of financial exploitation of the elders can be difficult. Clearly, the role of family dynamics needs to be evaluated in considering the dis­abled adult's capacities in the context of evaluations for guardianship, espe­cially where the guardian may benefit from their position.

Judgmental and Dispositional Components

Even with the eventual development of instruments for expressing person-situation incongruence, test developers should not be concerned with determining the degrees of incongruence that might define legal incompetence to care for one's property or oneself. Incompetence decisions require the application of judgments and moral values in order to conclude whether the weight of evidence tips the scales in favor of protection of the individual at the expense of loss of freedom and self-determination. This is not an empirical question and therefore is not properly addressed with predetermined, absolute criteria of a scientific nature.

It is important to recognize that decisions regarding self-determina­tion are moral and legal and must therefore be judicial. It is also important to acknowledge that such decisions are only as good as the data upon which they rely. Judges and juries generally do not have clinical training and rely upon the expertise of clinicians to complete evaluations of func­tioning that are comprehensive, valid, and age appropriate.

In addition, we should recognize that judicial determinations may be subject to the same ethical dilemmas clinicians face in balancing benefi­cence which promotes autonomy versus paternalism, especially for eld­erly adults about whom paternalistic instincts may run high. One area for future research consists of ethics studies of the judgmental and disposi­tional outcomes of guardianship proceedings. Potential racial (Vasvada, Masand, & Nasra, 1997) and age-based discrimination (Altman & Parmelee, 1989) in guardianship decisions could be further explored.

Clinical Application

Description

Not all guardianship cases involving disabled individuals will require a description on the full range of functional abilities represented in the instru­ments reviewed in this chapter. For example, some individuals may be so severely disabled physically (e.g., bed-ridden) that this condition alone will rule out the need to assess many behavioral tasks so that the evaluation may focus on cognitive capacities to direct others in self and estate care.

When a full assessment of capacities related to ability to care for self and/or property is needed, the instruments reviewed here offer several benefits, including standardization of procedure, generally acceptable reliability, and an assurance that the examinee will be described on a range of functions. Further, their use mitigates one's over-reliance on speculation about practical, functional abilities based on symptoms of physical or psychopathological disorder alone.

Examiners should not emphasize the use of summary scores at the expense of individual item responses when describing examinees' perform­ance on the instruments. The concept of limited guardianship is best served by providing information about discrete abilities; summary impairment scores may inappropriately encourage all-or-none notions of incompetency.

Explanation

Indications of poor ability on the instruments may be obtained for a variety of reasons. Among these are temporary conditions of fatigue, certain motivational states, misunderstanding of the examiner's inquiries (e.g., due to visual or hearing difficulties), atrophy of skills that conceiv­ably could be relearned, or actual deficit as a direct consequence of brain pathology or other disorder. Further, instruments that ask the examinee "Do you..." or "Can you..." (the MAI and MFAQ) may produce responses that are motivated by the examinee's self-perceptions and feel­ings of adequacy or inadequacy, such as exaggerated feelings of helpless­ness or a need to appear competent. Therefore, all indexes of functional strengths or deficits must be interpreted concerning their probable cause or their relation to other conditions of the examinee.

As previously noted, interpretation of performance on functional ability instruments will require a variety of other types of data concerning the examinee. Generally this might require assessment with neuropsycho­logical instruments, psychological tests of affect and personality, medical and neurological examination, including labs and scans, or other meth­ods. In addition, several discussions in this chapter have pointed to the importance of information about examinees' current social and physical environments when explaining their functional abilities and deficits. Thus a direct examination (observation) of their home environment and associ­ates may be necessary. The range of information that will be needed in many of these assessments again supports an interdisciplinary approach.

Special consideration should be given to the possibility that some disabled individuals may be capable of learning skills that they do not possess currently, rather than merely assuming that current inabilities are fixed and unchangeable (Willis, 1990). Thus assessments should include an evaluation of learning potential.

Prediction

Generalization of the instruments' results to functioning in everyday life presents several difficulties. Currently there is evidence that perform­ance on such instruments reflects systematic differences between elderly or nonelderly disabled groups who function in living arrangements requiring more or less self-sufficiency. Yet the instruments vary regarding empirical evidence that their indexes of specific abilities correspond to actual per­formance of those abilities outside the testing situation. Construct validity studies suggest that they do, but direct evidence is minimal.

Any use of the instruments' results to infer functioning or to predict future functioning in everyday settings must consider the fact that envi­ronmental settings themselves contribute greatly to an individual's mani­festation of the ability in question outside the test situation. Thus generalization of the results to external situations will require data on the current or future physical and social environment in which the examinee will be functioning. Any speculations on the basis of examination of the individual alone, without a corresponding environmental assessment, are improper and highly suspect. Certain features of the instruments them­selves provide some information about the examinee's environment. Nevertheless, the self report does not offer a substitute for obtaining and considering environmental data from other sources.

Conclusions

The instruments obviously cannot define legal competence or incom­petence, because these determinations require moral and social judgments about justice that are not empirical in nature (see Judgmental and Dispositional Components, in the first section of this chapter). However, these instruments can be extremely helpful in providing a way to assess behavioral and cognitive skills important for everyday functioning and for organizing the courts' thinking about these matters for the purpose of deciding questions of legal competence. Like all assessment instruments, those reviewed here must be used by qualified examiners familiar with the population and disorders presented, as well as the psychometric proper­ties that support or limit interpretation of tests (Moye, 2oO0).

Examiners may be especially helpful for the legal decision maker if they can characterize the congruency or discrepancy between the disabled individual's functional abilities and the demands or supports that are characteristic of specific environments. Expert testimony about the significance or severity of functional disability should be based on the previously mentioned person-situation comparisons, not on an absolute level of functioning alone. This is because very poor functioning in a certain area of ability may be of little consequence or of great con­sequence, depending on the nature of the examinee's everyday environ­ment. Thus describing an individual as impaired should be qualified, that is, impaired in relation to some particular demand of the relevant environment.

Degrees of discrepancy between an individual's manifest abilities and environmental demands might allow the examiner to testify about the risks associated with a decision not to provide the individual with guardianship assistance in various spheres of self-care or management of property. This testimony would not violate the general principle to avoid stating a conclusion on the ultimate legal question. That is, the assessment results might allow the examiner to conclude that the individual would be exposed to a particular degree of peril without a guardian's assistance; this could be stated without concluding that a legal threshold for a deter­mination of incompetency has or has not been reached.

However, as previously noted, forensic evaluations for the purposes of guardianship of person and estate rarely involve in-person testimony and cross-examination of the evaluator. This means there is not an oppor­tunity for the proposed ward and their counsel (if they are even present) to challenge the adequacy and appropriateness of the evaluation. There should be continued efforts to advocate for legislation that can increase the quality of competency evaluations in this area (e.g., bills which require the evaluator to provide detailed reports or to appear in person).

In conclusion, numerous instruments are available for use in guardianship evaluations. The examiner can choose the most appropriate instrument based on the best method and task appropriateness for the specific capacity area in question. Future research should extend the validity and normative data for these instruments. In the interim, clini­cians involved in these evaluations can carefully evaluate the disabling condition, cognitive abilities, and behavioral outcomes specific to any individual and guardianship issue presented and support the courts in arriving at fair and, if appropriate, limited guardianships.

This page intentionally left blank