For purposes of brevity, the model described in this paper is framed in terms of generic mental illness. Arguably, the discussion is most relevant to research involving people with more serious mental illnesses — such as schizophrenia, bipolar disorder, and dementia — and less relevant where people with milder disorders are involved.Google Scholar

National Bioethics Advisory Commission, Research Involving Persons with Mental Disorders That May Affect Decision-Making Capacity (December 1998), available at <http://bioethics.georgetown.edu/nbac/capacity/TOC.htm>..>Google Scholar

Ogloff, J.R.P. Otto, R.K., “Are Research Participants Truly Informed? Readability of Informed Consent Forms Used in Research,” Ethics and Behavior, 1 (1991): 239–52; Young, D.R. Hooker, D.T. Freeberg, F.E., “Informed Consent Documents: Increasing Comprehension by Reducing Reading Level,” IRB: A Review of Human Subjects Research, 12, no. 3 (1990): 1–5.Google Scholar

Waggoner, W.C. Mayo, D.M., “Who Understands? A Survey of 25 Words or Phrases Commonly Used in Proposed Clinical Research Consent Forms,” IRB: A Review of Human Subjects Research, 17, no. 1 (1995): 6–9.CrossRefGoogle Scholar

Peterson, B.T. et al. , “Improving Readability of Consent Forms: What the Computers May Not Tell You,” IRB: A Review of Human Subjects Research, 14, no. 6 (1992): 6–8.Google Scholar

DeRenzo, E.G., “The Ethics of Involving Psychiatrically Impaired Persons in Research,” IRB: A Review of Human Subjects Research, 16, no. 6 (1994): 7–9; Benson, P.R. et al. , “Information Disclosure, Subject Understanding, and Informed Consent in Psychiatric Research,” Law and Human Behavior, 12 (1988): 455–75;.Google Scholar

Reiser, S.J. Knudson, E, “Protecting Research Subjects After Consent: The Case for the ‘Research Intermediary,’” IRB: A Review of Human Subjects Research, 15, no. 2 (1993): 10–11.Google Scholar

Appelbaum, P.S. et al. , “Competence in Depressed Patients for Consent to Research,” American Journal of Psychiatry, 156 (1999): 1380–84; Carpenter, W.T. et al. , “Decisional Capacity for Informed Consent in Schizophrenia Research,” Archives of General Psychiatry, 57 (2000): 533–38.Google Scholar

Stiles, P.G. et al. , “Improving Understanding of Research Consent Disclosures Among Persons with Mental Illness,” Psychiatric Services, 52 (2001): 780–85.CrossRefGoogle Scholar

Miller, R. Willner, H.S., “The Two-Part Consent Form: Suggestions for Promoting Free and Informed Consent,” N. Engl. J. Med., 290 (1974): 964–66.Google Scholar

See, e.g., Appelbaum, P.S. Grisso, T., “The MacArthur Treatment Competence Study I: Mental Illness and Competence to Consent to Treatment,” Law and Human Behavior, 19 (1995): 105–26.Google Scholar

Bonnie, R., “The Competence of Criminal Defendants: Beyond Duskey and Drope,” University of Miami Law Review, 46 (1993): 539–601.Google Scholar

See also PharmSource Information Services, Inc., “Informed Consent: Model Capacity Assessment Program Provides Insights on Protecting Subjects Lacking Decisional Capacity,” Clinical Trials Advisor, 6, no. 18 (2001): 1–3.Google Scholar

Roth, L.H. Appelbaum, P.S., “Obtaining Informed Consent for Research with Psychiatric Patients,” Psychiatric Clinics of North America, 6 (1983): 551–65.CrossRefGoogle Scholar

Capacity may not necessarily be evaluated in normative terms. Particularly, where weak (i.e., undemanding) tests of comprehension are involved, arguments can be made that people must demonstrate an absolute level of performance in order to be considered competent.Google Scholar

My evidence for this assertion is necessarily anecdotal, as I have found no published studies where researchers require prospective research participants to pass a test of research disclosure comprehension in order to participate. Moreover, in over three years of service as a member of the University of South Florida institutional review board (IRB), I have not seen a single study that employed such a device to help inform the researcher's judgment about a prospective participant's capacity to consent.Google Scholar

Annual reviews by an IRB of ongoing research projects requires the reporting of participants who have been dropped from research. In our IRB, which conducts approximately thirty to forty such reviews monthly, prospective non-mentally ill patients are rarely excluded from studies for failing to meet various inclusion or exclusion criteria stated in the research protocols. In fact, to my knowledge, not a single patient has been excluded for incapacity to give informed consent. Furthermore, in published empirical studies with non-mentally ill participants, it is rare that investigators report the exclusion of potential participants for lack of capacity to consent.Google Scholar

Currently, neither our IRB staff nor anyone else monitors investigators' informed consent dialogues with prospective participants. There is probably a wide range of “effort” exerted by research team members across different studies. Anecdotally, I have observed investigators who took considerable time in the diligent presentation and discussion of consent forms. I have also interviewed a medical researcher who asserted that obtaining informed consent from a (presumably exhausted, potentially frightened) woman at the end of ten to twelve hours of unsuccessful labor using a consent form of some ten to twelve pages in length would take “about five minutes.”Google Scholar

The choice of two standard deviations below the mean performance cut-off score of a non-mentally ill group, though arbitrary, is rational. As one reviewer noted, it is a commonly used choice in distinguishing “normal” from “abnormal” physiological parameters. It is also frequently used to distinguish “normal” from “pathological” scores on psychological measures. As the text notes, it is a cut-off score that would exclude few non-mentally ill individuals from research participation, and thus is one that comports closely with current practice. Although this is a defensible criterion in principle, it may be a flawed criterion in practice if investigators fail to adequately test the competence-related capacities of the “normal” group. Furthermore, as a second reviewer noted, it is not necessary that the research community accept any single standard — two standard deviations below the “normal” group mean or otherwise, for all studies. The research community could well consider more stringent standards for studies that involve greater risk and lesser ones for studies involving no more than minimal risk. These considerations go beyond the scope of this paper, though they are highly relevant to the broader development of research guidelines. As noted supra note 14, it is possible to fashion competence measures that rely on absolute rather than normative performance levels. Relevant to the present context, the previous work of Taub and colleagues employed an 80 percent performance cut-off to evaluate the capacities of elderly individuals to consent to research participation.Google Scholar

See Taub, H.A. Kline, G.E. Baker, M.T., “The Elderly and Informed Consent: Effects of Vocabulary Level and Corrective Feedback,” Experimental Aging Research, 7 (1981): 137–46;.Google Scholar

Taub, H.A. Baker, M.T., “A Reevaluation of Informed Consent in the Elderly: A Method for Improving Comprehension Through Direct Testing,” Clinical Research, 32 (1984): 17–21.Google Scholar

Other arguments for the use of recognition measures include the claims that they are easier to construct and that they provide more precise documentation of the testing results. Optimally, recall measures require verbatim recording of a person's response to the query and the development of a scoring manual for use in determining the kinds of narrative responses (e.g., key words and phrases) for which the person will be given credit.Google Scholar

For example, the Fair Housing Amendments Act of 1988 and the Americans with Disabilities Act, effective in 1992, both prohibit discrimination on the basis of disabilityGoogle Scholar

Each also incorporates the principle of “reasonable accommodation.” To illustrate, if a person seeking employment has a disability, the employer is obligated to provide “reasonable accommodation” (e.g., flexible scheduling of work hours or providing access ramps for a person in a wheelchair) before deciding whether the disability is so severe that the person is unable to perform the job in question.Google Scholar

See Fair Housing Amendments Act, 42 U.S.C. §;§; 3601–31 (2002); Americans with Disabilities Act, 42 U.S.C. §;§; 12101–213 (2002).Google Scholar

The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research, DHEW Pub. No. OS 78-0012 (1978), available at <http://ohsr.od.nih.gov/mpa/belmont.php3>..>Google Scholar

Id. at 13.Google Scholar

Stiles, et al. , supra note 8.Google Scholar

Because this was an analogue design, participants were informed that the “clinical trial” was only hypothetical and that the investigators were merely evaluating their comprehension of the consent disclosures. Group members who were not mentally ill were asked to assume participation in a study of medication for heart disease. Despite the variations in the groups' illnesses, the consent forms were identical for all groups.Google Scholar

Both mentally ill and non-mentally ill people perform better on recognition than recall tasks. A shift from a minimum performance standard on recall tasks to a minimum performance standard based on recognition tasks of a non-mentally ill group would exclude some of the people (both mentally ill and non-mentally ill) that have been accepted as research participants (i.e., some of those who passed the recall standard would not pass the recognition standard).Google Scholar

Grisso, T. Appelbaum, P.S., “The MacArthur Treatment Competence Study III: Abilities of Patients to Consent to Psychiatric and Medical Treatments,” Law and Human Behavior, 19 (1995): 149–74.Google Scholar

Regarding this approach — and perhaps the previous one as well (utilizing normative data from archival studies) —investigators and IRBs would also have to consider the appropriateness of the proposed non-mentally ill group. Although a requirement that the normative group be matched case-by-case with study participants would almost certainly be unworkable, some consideration would need to be given to sample variables (e.g., age, education, intellectual capacity) that might affect the performance of the non-mentally ill group (and therefore the criterion cut-off score).Google Scholar

Carpenter, et al. , supra note 7.Google Scholar

Stiles, et al. , supra note 8.Google Scholar

Requiring that such justifications be reported to, and approved by, the IRB (perhaps by a designated IRB member rather than full board review) prior to engaging the participant in the research would provide additional protection for those mentally ill participants of marginal capacity.Google Scholar

Current federal guidelines identify the required elements of informed consent, but they do not explicitly make distinctions among them in terms of their relative importance as asserted here. My thanks to Thomas Grisso for alerting me to this feature of, and concern with, this model.Google Scholar

National Bioethics Advisory Commission, supra note 2.Google Scholar

See Recommendation 2 in National Bioethics Advisory Commission, “Chapter Five: Moving Ahead in Research Involving Persons with Mental Disorders: Summary and Recommendations,” Research Involving Persons with Mental Disorders That May Affect Decision-Making Capacity (December 1998), available at <http://bioethics.georgetown.edu/nbac/capacity/Moving.htm>..>Google Scholar

At present the creation of a special standing panel is merely a recommendation. To my knowledge, no formal governmental action (or congressional funding) has been taken to establish such a panel.Google Scholar

Recommendation 8 in National Bioethics Advisory Commission, supra note 33.Google Scholar

Arguably, the discretionary judgment of these “independent” evaluators might be less subject to bias in some circumstances (e.g., where there are professional or financial incentives to recruit participants to fill the cells of a research design) than that of the study investigators who would otherwise obtain consent.Google Scholar

Michels, R., “Are Research Ethics Bad for Our Mental Health?,” N. Engl. J. Med., 340 (1999): 1427–30.CrossRefGoogle Scholar

Capron, A.M., “Ethical and Human-Rights Issues in Research on Mental Disorders That May Affect Decision-Making Capacity,” N. Engl. J. Med., 340 (1999): 1430–34.Google Scholar

Recommendation 8 in National Bioethics Advisory Commission, supra note 33.Google Scholar