The right to individual self-governance (autonomy) is one of the core values of contemporary Western ethics. In medical ethics, this translates into the right to determine what happens to one’s own body in terms of health and mental health care, including the right to refuse treatment. But respect for individuals as autonomous agents also has broader and common application to self-management of civil, personal, and financial affairs, such as managing one’s assets and debts, making a will, entry into binding legal contracts, consenting to research participation, and consenting to sexual contact.
The most complex ethical dilemmas are not those with a clear choice between ethical and unethical actions, but rather those in which two important ethical mandates appear in conflict (DuBois, 2008). Although respect for autonomy is highly valued it is not absolute. In the context of those with limited capacity to make authentic autonomous decisions, there is a complementary ethical concern of “abandoning persons to their right to self-determination” (Freedom, Reference Biella, de Siqueira and Borges1984, p. 909) As described in The Belmont Report in reference to research ethics: “Respect for persons incorporates at least two ethical convictions: first, that individuals should be treated as autonomous agents, and second, that persons with diminished autonomy are entitled to protection” (unpaginated) (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, Reference DuBois1979). Achieving a balance between these two ethical imperatives is a profoundly important yet complex task. This issue comes up frequently for many mental health care providers, as a number of cognitive and other neuropsychiatric conditions may elevate the risk of impairment in decision-making capacity in one or more domains of independent functioning (Biella etal., Reference Kim and Appelbaum2020; Oshima etal., Reference Jeste, Eglit, Palmer, Martinis, Blanck and Saks2020; Parmigiani etal., Reference Dunn, Palmer, Karlawish, Miller and Cummings2021; Romdhani etal., Reference Kim, Appelbaum and Kim2020).
The importance of implementation efforts to ensure mental health professionals understand the ethical, legal, and practical underpinnings of capacity assessment was highlighted in seminal research from Marson and colleagues in the late 1990s. Specifically, they published a widely cited study regarding inter-rater reliability of physician’s capacity determinations in regard to capacity to consent to medical treatment (Marson etal., Reference Kohn1997). Twenty-nine patients with mild Alzheimer’s disease (AD) and 16 older comparison subjects were videotaped during a standardized consent capacity interview. Five experienced physicians board certified in their relevant subspecialties independently viewed each videotape and made a judgment as to whether the individual was competent or incompetent to consent to treatment.
Marson and colleagues found that all but one of the participants without AD were unanimously rated as competent, and the latter was rated as incompetent by only one of the five judges, yielding an agreement rate of 98%. However, there was substantially less consensus on ratings for people in the AD group. The percentage of “incompetent” ratings for the AD group varied from 90% by “physician 1” to 0% by “physician 5,” and the overall agreement rate for determinations of competence in the AD group was only 56%. As noted elsewhere by Marson etal. (Reference Marson, Earnst, Jamil, Bartolucci and Harrell1994) “It is precisely in these more equivocal cases that physicians are asked to render competency opinions and in which it is critical that experienced physicians demonstrate judgment consistency” (p. 12). The lack of agreement was further salient given that the physicians were each basing their determinations on the same videotaped interviews, whereas agreement among independent interviewers might be even lower. Similar results have subsequently been found by other researchers (Kim etal., Reference Marson, McInturff, Hawkins, Bartolucci and Harrell2011; Sessums etal., Reference Marson, Schmitt, Ingram and Harrell2011).
The lack of inter-physician consensus on capacity ratings represents a critical ethical dilemma. It suggest that when one is in the equivocal/border region of capable/incapable functioning, the ultimate determination of competency status for those on the border of capable/incapable status may depend more on the characteristics of the person conducting the evaluation rather than the characteristics of the person being evaluated. Fortunately, evidence from the same research group illustrated that education in competency standards may be key to improving the consistency of such determinations. Specifically, Marson and collegues gave five physicians individual standardized orientation to five legal standards of competency and their operationalization, before having the physicians view and rate competence of persons in the videotaped capacity interviews (Marson etal., Reference Moye, Sabatino and Weintraub Brendel2000). With such orientation, the physicians as a group had a 76% agreement rate on competency/incompetency judgments for 21 people diagnosed with probable AD, and even when restricted to the 10 patients with only “mild” severity of AD, agreement was 73%. Although a disagreement on one-fourth of the ratings remains concerning if one is on the receiving end of that one-fourth decision, education, even for experienced clinicians, is clearly important and helpful. The question then becomes one of how to efficiently implement education on a widespread basis.
The Brief Report provided by Peisah etal. (2019) in this volume illustrates the potential for global implementation of basic education in principles and methods of capacity assessment. Using a modification of the Delphi approach to achieve expert consensus, they constructed an e-learning tool for assessment and teaching some of the core aspects of competency assessments to health care professionals. Their tool includes a 25-item pretest of existing knowledge, a learning module in which the material presented to the individual is selected in reference to errors made on the pretest, followed by a posttest to evaluate improvement in knowledge. They tested the tool on 32 health care professionals from Israel, Canada, and Australia and found statistically significant improvements in core knowledge relevant to capacity assessment. Although acknowledging limitations of this one initial study, Pesaih etal. note the potential value of such an approach in improving understanding of the key aspects of capacity assessment.
One small study and educational tool tested in three Western nations cannot be expected to solve the complex ethical, legal, and practical issues of impaired decisional capacity and competency assessment. A strictly set tool or curriculum would be problematic. For example, outside of conservator or guardianship hearings, the capacity determination may not be one of global capacity for independent decision-making, but rather one limited to the specific decision at hand in that specific context. There are also individual legal and cultural considerations within each jurisdiction that prohibit a one-size-fits-all solution. But a brief tool such as that developed by Pesaih and colleagues can be helpful in increasing awareness of individuals regarding the complexities of capacity assessment. At minimum, raising the awareness of clinicians as to what they know and what they do not yet know can be of considerable service.
For individual health care providers who are called upon to conduct capacity evaluations, education is key. Education includes keeping abreast of ongoing discussions, debates, and developments within the professional literature on the very notion of competence, and the developing literature regarding the capacity/competency assessment process. In the context of consent for medical procedures or research participation, it is also important to consider deficits in standard methods in the consent process. Manifest capacity in these situations is affected not only by intraindividual characteristics (some transient, some stable but modifiable, and some that may be steadfast), but also by the inherent complexity of the information and decision at hand, and by the quality of the disclosure and consent discussion process (Dunn etal., Reference Oshima, Takenoshita and Iwai2007). Sometimes, it is not as much a matter that the individual is incapable of understanding, but more one of explaining and discussing the relevant information in ways appropriate to the individual’s abilities and preferences (Palmer and Harmell, Reference Palmer and Harmell2016; Palmer, Reference Palmer, Panicker and Stanley2021).
One helpful resource for clinicians who work with older adults is a freely available handbook on assessment of older adults with diminished capacity developed through a collaborative effort of the American Bar Association and American Psychological Association (American Bar Association Commission on Law and Aging & American Psychological Association, Reference Palmer, Harmell and Dunn2008). Seeking out mentorship, supervision, and consultation with other professionals is also important. Such mentors should obviously be persons experienced and remaining engaged in practice and in the evolving relevant ethical, legal, and clinical professional literature.
Alternatives to the dichotomous model of competence/incompetence should also be investigated and pursued where possible. A model that varies by degree rather than type could range from (1) fully capable of autonomous decision-making, (2) supported decision-making, (3) incapable of making the immediate decision but capable of appointing a proxy, to (4) incapable (Palmer etal., Reference Parmigiani, Del Casale and Mandarelli2018). There have been proponents of the various intermediate alternatives such as supported decision-making and capacity to appoint a proxy for several decades, but the legal changes in support of such options still lag in many instances (Jeste etal., Reference Thomasma2018; Kim and Appelbaum, Reference Peterson, Karlawish and Largent2006; Reference Peisah, Lerman, Herrmann, Rezmovitz and ShulmanKohn, in press; Moye etal., Reference Romdhani, Koskas, Kohler and Drunat2013; Peterson etal., Reference Sessums, Zembrzuska and Jackson2020).
Conclusions
Given the high stakes involved in balancing autonomy and protective considerations, the assessment of decision-making capacity is a task of paramount sensitivity and importance. The complexity of considerations in this intersection of ethics, law, and clinical care requires ongoing educational efforts and reappraisal of knowledge for those charged with conducting capacity assessments. Pesaih etal.’s e-learning tool represents an important initial step on one potential route to wider provision of such appraisal and assessment. There is clearly room for further improvement not only in raising inter-interviewer reliability, but also in ensuring the validity of competency determinations. There are no simple solutions to this problem, but the ongoing empirical research, including qualitative research with key stakeholders, is one essential part of the effort to find better, albeit never perfect, solutions to protect the rights and welfare of all persons, including those most vulnerable to impaired capacity for autonomy.
The right to individual self-governance (autonomy) is one of the core values of contemporary Western ethics. In medical ethics, this translates into the right to determine what happens to one’s own body in terms of health and mental health care, including the right to refuse treatment. But respect for individuals as autonomous agents also has broader and common application to self-management of civil, personal, and financial affairs, such as managing one’s assets and debts, making a will, entry into binding legal contracts, consenting to research participation, and consenting to sexual contact.
The most complex ethical dilemmas are not those with a clear choice between ethical and unethical actions, but rather those in which two important ethical mandates appear in conflict (DuBois, 2008). Although respect for autonomy is highly valued it is not absolute. In the context of those with limited capacity to make authentic autonomous decisions, there is a complementary ethical concern of “abandoning persons to their right to self-determination” (Freedom, Reference Biella, de Siqueira and Borges1984, p. 909) As described in The Belmont Report in reference to research ethics: “Respect for persons incorporates at least two ethical convictions: first, that individuals should be treated as autonomous agents, and second, that persons with diminished autonomy are entitled to protection” (unpaginated) (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, Reference DuBois1979). Achieving a balance between these two ethical imperatives is a profoundly important yet complex task. This issue comes up frequently for many mental health care providers, as a number of cognitive and other neuropsychiatric conditions may elevate the risk of impairment in decision-making capacity in one or more domains of independent functioning (Biella etal., Reference Kim and Appelbaum2020; Oshima etal., Reference Jeste, Eglit, Palmer, Martinis, Blanck and Saks2020; Parmigiani etal., Reference Dunn, Palmer, Karlawish, Miller and Cummings2021; Romdhani etal., Reference Kim, Appelbaum and Kim2020).
The importance of implementation efforts to ensure mental health professionals understand the ethical, legal, and practical underpinnings of capacity assessment was highlighted in seminal research from Marson and colleagues in the late 1990s. Specifically, they published a widely cited study regarding inter-rater reliability of physician’s capacity determinations in regard to capacity to consent to medical treatment (Marson etal., Reference Kohn1997). Twenty-nine patients with mild Alzheimer’s disease (AD) and 16 older comparison subjects were videotaped during a standardized consent capacity interview. Five experienced physicians board certified in their relevant subspecialties independently viewed each videotape and made a judgment as to whether the individual was competent or incompetent to consent to treatment.
Marson and colleagues found that all but one of the participants without AD were unanimously rated as competent, and the latter was rated as incompetent by only one of the five judges, yielding an agreement rate of 98%. However, there was substantially less consensus on ratings for people in the AD group. The percentage of “incompetent” ratings for the AD group varied from 90% by “physician 1” to 0% by “physician 5,” and the overall agreement rate for determinations of competence in the AD group was only 56%. As noted elsewhere by Marson etal. (Reference Marson, Earnst, Jamil, Bartolucci and Harrell1994) “It is precisely in these more equivocal cases that physicians are asked to render competency opinions and in which it is critical that experienced physicians demonstrate judgment consistency” (p. 12). The lack of agreement was further salient given that the physicians were each basing their determinations on the same videotaped interviews, whereas agreement among independent interviewers might be even lower. Similar results have subsequently been found by other researchers (Kim etal., Reference Marson, McInturff, Hawkins, Bartolucci and Harrell2011; Sessums etal., Reference Marson, Schmitt, Ingram and Harrell2011).
The lack of inter-physician consensus on capacity ratings represents a critical ethical dilemma. It suggest that when one is in the equivocal/border region of capable/incapable functioning, the ultimate determination of competency status for those on the border of capable/incapable status may depend more on the characteristics of the person conducting the evaluation rather than the characteristics of the person being evaluated. Fortunately, evidence from the same research group illustrated that education in competency standards may be key to improving the consistency of such determinations. Specifically, Marson and collegues gave five physicians individual standardized orientation to five legal standards of competency and their operationalization, before having the physicians view and rate competence of persons in the videotaped capacity interviews (Marson etal., Reference Moye, Sabatino and Weintraub Brendel2000). With such orientation, the physicians as a group had a 76% agreement rate on competency/incompetency judgments for 21 people diagnosed with probable AD, and even when restricted to the 10 patients with only “mild” severity of AD, agreement was 73%. Although a disagreement on one-fourth of the ratings remains concerning if one is on the receiving end of that one-fourth decision, education, even for experienced clinicians, is clearly important and helpful. The question then becomes one of how to efficiently implement education on a widespread basis.
The Brief Report provided by Peisah etal. (2019) in this volume illustrates the potential for global implementation of basic education in principles and methods of capacity assessment. Using a modification of the Delphi approach to achieve expert consensus, they constructed an e-learning tool for assessment and teaching some of the core aspects of competency assessments to health care professionals. Their tool includes a 25-item pretest of existing knowledge, a learning module in which the material presented to the individual is selected in reference to errors made on the pretest, followed by a posttest to evaluate improvement in knowledge. They tested the tool on 32 health care professionals from Israel, Canada, and Australia and found statistically significant improvements in core knowledge relevant to capacity assessment. Although acknowledging limitations of this one initial study, Pesaih etal. note the potential value of such an approach in improving understanding of the key aspects of capacity assessment.
One small study and educational tool tested in three Western nations cannot be expected to solve the complex ethical, legal, and practical issues of impaired decisional capacity and competency assessment. A strictly set tool or curriculum would be problematic. For example, outside of conservator or guardianship hearings, the capacity determination may not be one of global capacity for independent decision-making, but rather one limited to the specific decision at hand in that specific context. There are also individual legal and cultural considerations within each jurisdiction that prohibit a one-size-fits-all solution. But a brief tool such as that developed by Pesaih and colleagues can be helpful in increasing awareness of individuals regarding the complexities of capacity assessment. At minimum, raising the awareness of clinicians as to what they know and what they do not yet know can be of considerable service.
For individual health care providers who are called upon to conduct capacity evaluations, education is key. Education includes keeping abreast of ongoing discussions, debates, and developments within the professional literature on the very notion of competence, and the developing literature regarding the capacity/competency assessment process. In the context of consent for medical procedures or research participation, it is also important to consider deficits in standard methods in the consent process. Manifest capacity in these situations is affected not only by intraindividual characteristics (some transient, some stable but modifiable, and some that may be steadfast), but also by the inherent complexity of the information and decision at hand, and by the quality of the disclosure and consent discussion process (Dunn etal., Reference Oshima, Takenoshita and Iwai2007). Sometimes, it is not as much a matter that the individual is incapable of understanding, but more one of explaining and discussing the relevant information in ways appropriate to the individual’s abilities and preferences (Palmer and Harmell, Reference Palmer and Harmell2016; Palmer, Reference Palmer, Panicker and Stanley2021).
One helpful resource for clinicians who work with older adults is a freely available handbook on assessment of older adults with diminished capacity developed through a collaborative effort of the American Bar Association and American Psychological Association (American Bar Association Commission on Law and Aging & American Psychological Association, Reference Palmer, Harmell and Dunn2008). Seeking out mentorship, supervision, and consultation with other professionals is also important. Such mentors should obviously be persons experienced and remaining engaged in practice and in the evolving relevant ethical, legal, and clinical professional literature.
Alternatives to the dichotomous model of competence/incompetence should also be investigated and pursued where possible. A model that varies by degree rather than type could range from (1) fully capable of autonomous decision-making, (2) supported decision-making, (3) incapable of making the immediate decision but capable of appointing a proxy, to (4) incapable (Palmer etal., Reference Parmigiani, Del Casale and Mandarelli2018). There have been proponents of the various intermediate alternatives such as supported decision-making and capacity to appoint a proxy for several decades, but the legal changes in support of such options still lag in many instances (Jeste etal., Reference Thomasma2018; Kim and Appelbaum, Reference Peterson, Karlawish and Largent2006; Reference Peisah, Lerman, Herrmann, Rezmovitz and ShulmanKohn, in press; Moye etal., Reference Romdhani, Koskas, Kohler and Drunat2013; Peterson etal., Reference Sessums, Zembrzuska and Jackson2020).
Conclusions
Given the high stakes involved in balancing autonomy and protective considerations, the assessment of decision-making capacity is a task of paramount sensitivity and importance. The complexity of considerations in this intersection of ethics, law, and clinical care requires ongoing educational efforts and reappraisal of knowledge for those charged with conducting capacity assessments. Pesaih etal.’s e-learning tool represents an important initial step on one potential route to wider provision of such appraisal and assessment. There is clearly room for further improvement not only in raising inter-interviewer reliability, but also in ensuring the validity of competency determinations. There are no simple solutions to this problem, but the ongoing empirical research, including qualitative research with key stakeholders, is one essential part of the effort to find better, albeit never perfect, solutions to protect the rights and welfare of all persons, including those most vulnerable to impaired capacity for autonomy.