IRB: Ethics & Human Research
Maintaining Informed Consent Validity during Lengthy Research Protocols
Valid informed consent is a cornerstone of ethical clinical research. A great deal of work has been undertaken to improve the informed consent process for clinical research, largely targeting the adequacy of consent at the time of a subject’s enrollment in a protocol.1Less attention has been paid to research subjects’ ability to maintain that knowledge over the course of participation in protocols that extend over substantial periods of time.2Many clinical trials are lengthy and easily confused with ordinary treatment.3Federal regulations governing research with humans4are silent on the extent to which information disclosed during the consent process must be retained for subjects’ continued participation to be warranted, and there is no consensus on this (and little explicit consideration) in the research ethics literature.5
There is ample evidence that participants in clinical studies frequently are unable to retain study information for the duration of their involvement in the research. Poor memory for information disclosed during the initial consent process has been observed in medically ill patients participating in research on cancer,6HIV,7hypertension,8and other types of research.9Participants in these studies exhibit poor knowledge of such protocol elements as risks, procedures, and the general purpose of the research. In some cases they exhibit confusion about the conceptual difference between the assignment of study medications or procedures and the individualized treatment decisions their own doctors would make. Some study participants report uncertainty about whether they were enrolled in research at all.
Why should we expect research subjects to retain information after protocol enrollment? After all, it is common for people in ordinary life to enter into contracts or sign leases that remain valid regardless of the lessee’s memory of the terms to which they consented. Unlike a lease agreement, though, where the lessee’s continued knowledge of the terms of the lease has no bearing on his or her obligations to the lessor, it can be argued that the justification for ongoing participation in research depends substantially on subjects’ continued knowledge of at least some protocol-related information. It is perhaps particularly important for subjects to remain aware that, in the research context, patient-centered decision-making is reduced as physicians and researchers are constrained by protocol requirements. Moreover, research subjects are expected to know that they can discontinue participation if they wish, without penalty, and return to ordinary clinical care.
If research subjects no longer grasp the nature of the compromises they have accepted, or if they fail to recognize their right to seek care outside of research settings, their ability to protect their interests will be materially impaired. Hence, it seems reasonable to demand that subjects have at least some ongoing knowledge about the nature of their involvement in a research study. This conclusion has been widely assumed, and is clearly recognized in dementia research where cognitive decline over time is predictable, and advance directives and surrogate decision-makers assume responsibility for continuing participation (or not) when subjects lose decisional capacity. But even here the quality and quantity of information to be retained is not resolved. In most clinical research studies, where loss of competence during protocol participation is not presumed, neither standards nor methods are in place to address sustained consent.
In the project described here we asked three questions: What study information should we expect research subjects to retain over time? How might study-related knowledge be assessed and monitored? What might be the appropriate course of action when a researcher suspects a subject has not successfully retained key knowledge? Our primary goal was to develop a methodology combining criteria to assess participants’ ongoing knowledge of critical study information with an opportunity to review the information. We sought expert advice from representatives of several relevant stakeholder groups. These advisors contributed expertise in clinical research; neurocognition; psychiatry and mental illness; bioethics; health care and human subjects protection law; legal and clinical competency assessment; decision science; clinical research oversight; and mental health consumers’ interests.
Useful criteria for the maintenance of critical study information is a relative construct that is meaningful only if we know what information was disclosed at the time of enrollment. In general we can assume that institutionally approved research (we will use a clinical trial as an example) conforms to existing guidelines in its consent procedures, disclosing at least: the identity of the researchers; the purpose of the study; subjects’ responsibilities (e.g., clinic visits, blood draws, clinical interviews); the possible treatments (e.g., placebo, active drug); method of treatment assignments (e.g., randomization); specific risks and benefits; confidentiality and safety measures; alternatives to participation (e.g., continuing current treatment); and subjects’ right to withdraw. Consent forms for clinical trials typically include other details such as the names of the drugs under study, specific medication doses, dosing schedules, and a list of all study sites. Also, researchers increasingly are encouraged to disclose to prospective subjects all financial interests in the recruitment and outcome of the study.
In recent years, researchers and ethicists generally have agreed that to justify initial enrollment in a research study, a prospective subject must be proficient in four basic domains:10understanding of the facts of the study (e.g., purpose, design, subjects’ responsibilities); appreciation of the study’s personal relevance and potential consequences of participation (e.g., risks and benefits, limited personalized treatment decisions); ability to reason logically about the options (e.g., assess the risk/benefit ratio); and ability to express a personal choice about whether to participate. It is argued that without these abilities, prospective subjects are likely not able to make a reasonable, genuinely informed decision about whether to enter a study. What abilities, then, must enrolled subjects have to make reasonable, genuinely informed decisions to continue participation in a study? The answer depends, in part, on whether the two decisions are the same. In most cases agreeing to enroll in a clinical trial is to agree to some change in the status quo regarding, for example, medication regimen or therapy activities. As time passes, the details of study participation become less abstract and subjects’ experience generates a different decision-making context. Now the decision is closer to a confirmation that the risk/benefit ratio is still in their favor. This ongoing calculation depends on participants’ continued knowledge of some critical protocol information.
In proposing that ongoing study knowledge should be assessed and monitored, our intention is not to reject the validity or meaningfulness of individuals’ initial consent to participate. We agree with Wendler and Rackoff’s argument that, in the absence of a material and/or significant change in the protocol or in a participant’s clinical state or personal situation, investigators should consider the individual’s initial consent to be valid.11We believe, however, that among those potentially important changes (e.g., newly discovered serious risks or side effects, worsening of the subject’s clinical state) may be a change in a participant’s ongoing knowledge of his own circumstances and options. Here we propose a method of detecting this type of change, just as many common symptom rating scales detect changes in a participant’s clinical state. Knowing the facts of a study is necessary but not sufficient to establish the ongoing validity of an individual’s consent to participate. Other factors, such as reasoning ability and external coercive influences, must also be taken into account when judging the validity of consent.
In the following sections we present the results of our work with stakeholder representatives to characterize the knowledge that research participants should retain over time, as well as methods to assess and document that knowledge and provide education where necessary.
Identifying the Knowledge Needed to Maintain Valid Consent
To determine the information that subjects should be expected to remember over time we sought advice from our stakeholder representatives, each of whom contributed a unique perspective. These stakeholders comprised two independent bodies: a nine-member Work Group (including the authors of this report) and a seven-member Advisory Group (see Acknowledgments). An initial outline of critical consent information, proposed by three of the authors (KJP, RRC, WTC), was reviewed and revised by the larger Work Group. Following extensive discussion within the Work Group, six elements of consent were deemed necessary to uphold the validity of consent over time:
Knowledge of Research Participation. Subjects’ continued knowledge that they are participating in a research project is a critical threshold issue. If after several weeks of participating in a study, the individual has completely lost sight of the fact of his participation, we would question the individual’s capacity to make a judgment about how to proceed. Subjects who are unaware that they are involved in research cannot monitor the risk/benefit ratio of participation and make meaningful decisions about whether to continue their involvement. This situation is fertile ground for the therapeutic misconception, since the individual may be taking a medication without the critical understanding that it has not been selected for him personally.
Understanding the Voluntary Nature of Research Participation and the Right to Withdraw. Subjects must at least maintain the knowledge that withdrawal from a study is permitted, without penalty, at any time. Awareness of research involvement by itself has little practical import unless subjects also appreciate that their participation is voluntary.
Understanding that Withdrawal from the Protocol Will Not Affect Usual Clinical Care Options. Ongoing participation is only truly voluntary if it is uncoerced. Subjects considering withdrawal from studies must be able to make their decisions free of concern that treatment options and clinical care otherwise available to them will be lost. It is possible that in some circumstances the treatment under study really is the only direct treatment option, and participants should be aware of that fact. However, they should also be aware that in no case would the standard clinical care otherwise available be withheld as a result of a decision not to participate. Subjects who are unaware of this may falsely believe that the alternatives to participation are constrained.
Knowledge of the General Purpose or Goal of the Research. Together with the participant’s memory of the study’s risks (below), ongoing knowledge of the study’s purpose is important to the risk/benefit analysis that lies at the heart of initial and ongoing consent. In many clinical trials, where there is no guarantee of a direct clinical benefit to the participant, the greater good that is served by the overall scientific purpose of the research is the most reliable benefit. Subjects who fail to grasp the purpose of a research study for which they are giving up a more individualized approach to treatment—and in which they may be subject to inconvenience, discomforts, and risks—are in no position to judge whether they wish to continue their involvement. Although participants make risk/benefit calculations at the time of study enrollment, as time goes on and life and attitudes change, participants need to recalculate that ratio to make sure it is still in their favor.
Knowledge of the Potential Risks of Participation. Research subjects in a longitudinal study are faced with an implicit question: should I continue to participate in this research project? That decision, as with the original decision to enter a study, should be based in part on an evaluation of the relative risks and benefits of participation. Without the ability to recall the risks of study participation, participants will not be able to evaluate the risk/benefit ratio.
Knowledge of the Differences Between Traditional, Individualized Clinical Care and Standardized Treatment during Research Participation. In ordinary clinical care, decisions about treatment are made on the basis of the best interests of the individual patient (though external constraints such as drug availability and funding are common). Research, in contrast, involves the application of protocols that usually restrict physicians’ discretion in altering dosages, changing medications, utilizing adjunctive treatments, and the like. When subjects are unable to recognize this difference, they are likely to overestimate the individualized therapeutic intent of the research (i.e., the therapeutic misconception12) and lose sight of the sacrifices they make in agreeing to participate. This issue gets at the heart of individuals’ motives for participation, and it is often very difficult to alleviate, especially in the context of lengthy clinical trials where traditional care and research protocols have overlapping methods and personnel.
These six aspects of informed consent were identified by the Work Group as critical to ongoing consent validity, thus representing consensus across a range of stakeholders. The list is not extensive, as we were looking ahead to an assessment approach that could easily and conveniently be implemented in most settings for clinical research. As attention to maintaining valid consent increases, modifications of these six elements and additions of others will no doubt emerge.
Assessment of Sustained Informed Consent
We propose that the continuing validity of informed consent, or “sustained consent,” depends in large part on participants’ ongoing knowledge of critical study information, and that it is therefore important to have a way of assessing that knowledge over time, and for educating participants when that knowledge has diminished. We worked with our Advisory Group to develop the Assessment of Sustained Informed Consent (ASIC) designed to address the six elements described above.
In a typical one-on-one assessment scenario, the interviewer elicits from the participant a response to each question in turn. If the participant is unable to provide an adequate response to any question, then the interviewer reviews the relevant information until the participant reaches a satisfactory level of understanding or, alternatively, shows a clear inability to do so. Subjects’ final responses are judged on a pass/fail basis, scoring 0 or 1 on each question. During administration, the interviewer records the appropriate score and indicates, for the record, whether the participant required remediation.
The most critical attribute of the ASIC is flexibility—it must be useful across a variety of protocols and participants. Accordingly, we designed the ASIC as a modifiable template with seven basic questions covering the six elements discussed previously (Table 1) and guidelines for what researchers should look for in participants’ answers when assessing sustained consent. The ASIC was conceived as a set of criteria for which individual investigators determine relative values and acceptable means of meeting the criteria, and not as a rating scale. The ASIC template is an electronic document with some fixed text and some blank form fields. The template provides the seven basic questions shown in Table 1, and blank input fields allow investigators to insert protocol-specific answer criteria, including knowledge required to earn a passing score and specific errors of omission or commission that would justify a failing score. These specific criteria would be determined during the protocol development stages for each protocol by the investigator in cooperation with the appropriate institutional research review board. This flexibility allows investigators and Institutional Review Boards (IRBs) to specify, for example, exactly which possible side effects a participant must recall. It might be important in some studies that participants know exactly which aspect of their disease is targeted by the study drug (e.g., schizophrenia-related cognitive deficits as opposed to hallucinations).
The ASIC’s inherent flexibility allows for its use in any number of clinical research settings and populations, although it is not reasonable or feasible to include such an assessment in all research protocols. Decisions about when and with whom this type of assessment is warranted probably depend at a minimum on whether the research poses substantial risk to participants or involves some alteration or interruption of usual clinical care, and whether the study population is likely to experience declines or fluctuations in cognitive capacity or insight. This type of assessment is not relevant to very short term, nontherapeutic studies or in research that poses minimal risk or only a modest increment above minimal risk. Practical considerations, such as how often the assessment is administered, are at the discretion of the researcher and the IRB.
When we did an informal test of the ASIC questions’ feasibility with a small sample of schizophrenia patients enrolled in clinical trials (N = 11), we found that most of the questions yielded consistently interpretable answers. Eight research staff members showed strong agreement, ranging from 0.67 (kappa coefficient) to 100% agreement, on how to score patients’ answers to questions about voluntariness and right to withdraw, alternatives to participation, and risks of participation. Agreement was only fair (kappa = 0.30 to 0.39) for the more complex questions pertaining to the purpose of the research and to the differences between individualized treatment and standardized study procedures. The lower kappa scores for some items likely reflect a need to refine their wording to better accommodate the specific study population and, indeed, individual participants. While the psychometric properties of individual items and formal inter-rater agreement statistics are of little generalizable value for instruments as fluid as the ASIC, these observations likely reflect equally the patients’ ability to express complex thoughts and the raters’ ability to interpret them. These are both important considerations for protocol-specific ASIC development and for staff training.
Next Steps toward Ensuring Continued Validity of Informed Consent
It is evident that many clinical trial participants lose track of the purpose of their participation, may develop a therapeutic misconception, and may not be aware of their rights as subjects. These problems, if widespread, must be addressed. We propose that periodic assessment and, where needed, review of protocols’ essential elements will be a useful reminder of key consent information and will identify subjects who may need close monitoring. This may already be accomplished informally in many studies, but a systematic assessment will simultaneously facilitate the process and document the results. Intermittent testing would also provide useful information for investigators about the characteristics of their research that are most difficult for participants to retain. This could inform investigators’ efforts to create maximally useful consent procedures.
The initial effort described here provides a starting point. Although we invite clinical investigators to evaluate the ASIC model and determine how best to implement it in their individual protocols, the chief value of this initial approach will be to stimulate discussion in the field and move toward a broad consensus on the essential elements of sustained informed consent. Once such a consensus is accomplished, assessment approaches can be refined and validated. The ASIC can serve as a model for an instrument that is practical to implement in a variety of clinical research settings.
An important question is whether attention to subjects’ retention of consent-related information will lead to increased study attrition. In undertaking this project, we neither sought nor anticipated an outcome of this sort. Subjects’ failure to grasp information critical to protection of their interests should trigger educational efforts, not an automatic exclusion from further research participation. Only for what we assume will be a small percentage of subjects who cannot grasp the simple information necessary for valid, ongoing consent may alternative authorization or removal from the study be necessary. However, determination of the extent of the problem awaits further research.
IRB approval was not obtained for this work. The University of Maryland Baltimore IRB granted a waiver of review for this project, having determined that it was a quality assurance activity. Accordingly, this report does not contain data collected from human research subjects.
This work was supported by a Distinguished Investigator Award from the National Alliance for Research on Schizophrenia and Depression, and by a National Institutes of Health grant (P30MH-068580).
We extend sincere gratitude to our fellow Work Group members and to the members of our Advisory Group, all of whom made important contributions to the development of the ideas presented in this paper.
Paul S. Appelbaum, MD, Director, Division of Psychiatry, Law, and Ethics, and Elizabeth K. Dollard Professor of Psychiatry, Medicine, and Law, Columbia University College of Physicians and Surgeons; Alan Bellack, PhD, ABPP, Director, VA Capital Health Care Network MIRECC, Baltimore Veterans Administration Medical Center, and Professor of Psychiatry and Director, Division of Psychology, University of Maryland School of Medicine; William T. Carpenter, Jr., MD, Director, Maryland Psychiatric Research Center (MPRC), and Professor of Psychiatry and Pharmacology, University of Maryland School of Medicine; Robert R. Conley, MD, Professor of Psychiatry, University of Maryland School of Medicine, Chief, Treatment Research Program, MPRC, and IRB Chair, National Institute on Drug Abuse; Jonathan D. Moreno, PhD, David and Lyn Silfen University Professor, Professor of Medical Ethics and of History and Sociology of Science, University of Pennsylvania; Kristen J. Prentice, PhD, Assistant Professor of Psychiatry, MPRC, University of Maryland School of Medicine, and Research Health Scientist, VA Capital Health Care Network MIRECC, Baltimore Veterans Administration Medical Center; Peter Rabins, MD, MPH, Professor of Psychiatry and Behavioral Sciences, and Codirector, Division of Geriatric Psychiatry and Neuropsychiatry, Johns Hopkins School of Medicine; Jack Schwartz, JD, Assistant Attorney General, Director of Health Policy Development, Maryland Attorney General’s Office; and Paul Root Wolpe, PhD, Senior Faculty Associate, Department of Psychiatry, Senior Fellow, Center for Bioethics, and Director, Program in Psychiatry and Ethics, University of Pennsylvania.
Richard J. Bonnie, LLB (Chair), John S. Battle Professor of Law, Professor of Psychiatric Medicine, and Director, Institute of Law, Psychiatry and Public Policy, University of Virginia School of Law; David Shore, MD, Associate Director for Clinical Research, National Institute of Mental Health; Ron Honberg, JD, National Director, Policy and Legal Affairs, National Alliance on Mental Illness; Charles McCarthy, PhD, Adjunct Professor of Research Ethics, Virginia Commonwealth University, and former Director, Office for Protection from Research Risks, National Institutes of Health; Ruth R. Faden, PhD, MPH, Philip Franklin Wagley Professor of Biomedical Ethics, Executive Director, Johns Hopkins Berman Institute of Bioethics, The Johns Hopkins University, and Senior Research Scholar, Kennedy Institute of Ethics, Georgetown University; James F. Childress, PhD, John Allen Hollingsworth Professor of Ethics, Professor of Medical Education, and Director, Institute for Practical Ethics and Public Life, University of Virginia; Suzanne Vogel-Scibilia, MD, Assistant Clinical Professor, Western Psychiatric Institute and Clinic, and Past President, National Alliance on Mental Illness.
Kristen J. Prentice, PhD, is Assistant Professor of Psychiatry, University of Maryland School of Medicine, and Research Health Scientist, VA Capitol Health Care Network (VISN 5) Mental Illness Research, Education, and Clinical Center (MIRECC), Baltimore, MD; Paul S. Appelbaum, MD, is Professor and Director, Division of Psychiatry, Law, and Ethics, Columbia University College of Physicians and Surgeons, New York, NY; Robert R. Conley, MD, is Professor of Psychiatry, University of Maryland School of Medicine and Maryland Psychiatric Research Center, Baltimore, MD; and William T. Carpenter, Jr., MD, is Professor of Psychiatry and Pharmacology, University of Maryland School of Medicine, and Director, Maryland Psychiatric Research Center, Baltimore, MD.
1. Dunn LB, Candilis PJ, Roberts LW. Emerging empirical evidence on the ethics of schizophrenia research.Schizophrenia Bulletin2006;32(1):47-68 (Epub October 19, 2005); Moser DJ, Reese RL, Hey CT, et al. Using a brief intervention to improve decisional capacity in schizophrenia research.Schizophrenia Bulletin2006;32(1):116-120 (Epub September 21, 2005).
2. Wendler D, Rackoff J. Consent for continuing research participation: What is it and when should it be obtained?IRB: Ethics & Human Research2002;24(3):1-6.
3. Appelbaum PS, Roth LH, Lidz CW, et al. False hopes and best data: Consent to research and the therapeutic misconception.Hastings Center Report1987;17(2):20-24; Horng S, Grady C. Misunderstanding in clinical research: Distinguishing therapeutic misconception, therapeutic misestimation, and therapeutic optimism.IRB: Ethics & Human Research2003;25(1):11-16; Lidz CW, Appelbaum PS. The therapeutic misconception: Problems and solutions.Medical Care2002;40(9 Suppl):V55-63; Lidz CW, Appelbaum PS, Grisso T, et al. Therapeutic misconception and the appreciation of risks in clinical trials.Social Science & Medicine2004;58(9):1689-1697.
4. Department of Health and Human Services. Protection of Human Subjects. 45 CFR 46.
5. See ref. 2, Wendler and Rackoff 2002.
6. Olver IN, Buchanan L, Laidlaw C, et al. The adequacy of consent forms for informing patients entering oncological clinical trials.Annals of Oncology1995;6(9):867-870; Schaeffer MH, Krantz DS, Wichman A, et al. The impact of disease severity on the informed consent process in clinical research.American Journal of Medicine1996;100(3):261-268.
7. See ref. 6, Schaeffer et al. 1996.
8. Bergler JH, Pennington AC, Metcalfe M, et al. Informed consent: How much does the patient understand?Clinical Pharmacology & Therapeutics1980;27(4):435-440.
9. See ref. 3, Lidz 2004; Criscione LG, Sugarman J, Sanders L, et al. Informed consent in a clinical trial of a novel treatment for rheumatoid arthritis.Arthritis & Rheumatism2003;49(3):361-367; Estey A, Wilkin G, Dossetor J. Are research subjects able to retain the information they are given during the consent process?Health Law Review1994;3(2):37-41; Hassar M, Weintraub M. “Uninformed” consent and the wealthy volunteer: An analysis of patient volunteers in a clinical trial of a new anti-inflammatory drug.Clinical Pharmacology & Therapeutics1976;20(4):379-386; Lynoe N, Sandlund M, Dahlqvist G, et al. Informed consent: Study of quality of information given to participants in a clinical trial.BMJ1991;303(6803):610-613; Searight HR, Miller CK. Remembering and interpreting informed consent: A qualitative study of drug trial participants.Journal of the American Board of Family Practice1996;9(1):14-22; Sugarman J, Kass NE, Goodman SN, et al. What patients say about medical research.IRB: A Review of Human Subjects Research1998;20(4):1-7.
10. Appelbaum PS, Roth LH. Competency to consent to research: A psychiatric overview.Archives of General Psychiatry1982;39(8): 951-958; Dunn LB, Nowrangi MA, Palmer BW, et al. Assessing decisional capacity for clinical research or treatment: A review of instruments.American Journal of Psychiatry2006;163(8):1323-1334; Sturman ED. The capacity to consent to treatment and research: A review of standardized assessment tools.Clinical Psychology Review2005;25(7):954-974.
11. See ref. 2, Wendler and Rackoff 2002.
12. See ref. 3, Appelbaum 1987, Lidz 2002, Lidz 2004; Appelbaum PS, Lidz CW, Grisso T. Therapeutic misconception in clinical research: Frequency and risk factors.IRB: Ethics & Human Research2004;26(2):1-8; Fisher JA. Procedural misconceptions and informed consent: Insights from empirical research on the clinical trials industry.Kennedy Institute of Ethics Journal2006; 16(3):251-268.
Kristen J. Prentice, Paul S. Appelbaum, Robert R. Conley, and William T. Carpenter, “Maintaining Informed Consent Validity during Lengthy Research Protocols,”IRB: Ethics & Human Research29, no. 6 (2007): 1-6.