![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
Abstract
Over the last few decades, multiple studies have examined the understanding of participants in clinical research. They show variable and often poor understanding of key elements of disclosure, such as expected risks and the experimental nature of treatments. Did the participants in these studies give valid consent? According to the standard view of informed consent they did not. The standard view holds that the recipient of consent has a duty to disclose certain information to the profferer of consent because valid consent requires that information to be understood. The contents of the understanding and disclosure requirements are therefore conceptually linked. In this paper, we argue that the standard view is mistaken. The disclosure and understanding requirements have distinct grounds tied to two different ways in which a token of consent can be rendered invalid. Analysis of these grounds allows us to derive the contents of the two requirements. It also implies that it is sometimes permissible to enroll willing participants who have not understood everything that they ought to be told about their clinical trials.
ACKNOWLEDGMENTS
The authors thank Gopal Sreenivasan, David Wendler, three anonymous reviewers for the journal, and audiences at the American Philosophical Association Eastern Division Annual Meeting, the Canadian Philosophical Association Annual Congress, the NIH Clinical Center Department of Bioethics, the New Scholarship in Bioethics Annual Symposium, Northeastern University, and Oxford University.
DISCLOSURE STATEMENT
The views expressed are the authors’ own. They do not represent the position or policy of the National Institutes of Health, the U.S. Public Health Service, or the Department of Health and Human Services. No potential conflict of interest was reported by the author(s).
Notes
1 Note that even the most ardent proponent of the standard view can accept that there is some information that should be made available to participants but that need not actually be understood. For example, the consent form is a useful place to provide the contact information for the research coordinator and the location of the study. But while this information ought to be provided, its provision is not relevant to the validity of consent. It is therefore not a component of the “disclosure requirement” as we use the term here.
2 Some exceptions include: Manson and O’Neill (Citation2007); Miller and Wertheimer (Citation2011); O'Neill (Citation2002); Sreenivasan (Citation2003); and Wendler (Citation2004).
3 Cf. Sreenivasan (Citation2003). Sreenivasan also rejects the standard view, but suggests a different account of what the informational requirements for informed consent are. We reject Sreenivasan’s analysis of those requirements in Section 5 (see footnote 12).
4 Two notable exceptions are Gert et al. (Citation1997) who identify the purpose of disclosure as the avoidance of deception and Sreenivasan (Citation2003) who argues that the disclosure and understanding requirements should be separated.
5 Sreenivasan also suggests that what must be understood is related to what is in participant interests, though he does not explicitly endorse an interests view (Sreenivasan Citation2003, Citation2019).
6 Especially, that is, in clinical research where so much information that is likely to be relevant to a potential participant’s interests or would be material to his decision is unknown. To illustrate, consider Case 1 again. If the set of facts that the prospective participant needs to understand in order to protect his interests or would be material to his enrollment decision is identical to the set of facts that the researcher happens to know and believe is relevant at the time consent is requested—i.e. risks A, B, and that there may be further unknown risks—then the researcher would not need to disclose risk C were she to learn of it before the trial starts. After all, risk C is not in the set of facts that would be material to his decision or necessary to protect his interests on this modified version of the standard view because that set of facts is precisely the same as the set of facts the requestor of consent happened to know and believe was relevant at the time consent was requested. This is clearly implausible. But, more importantly, it is not implied by the interests or inducements views; it is only implied by a version of those views retrofitted to deal with the original objection.
7 Our thanks to an anonymous reviewer for this suggestion.
8 The standard view implies that she ought to disclose risk C because the content of the understanding requirement determines the content of the disclosure requirement. Since risk C is a fact that he could understand, she ought to disclose it. But this is not a duty she can discharge. Any attempt to avoid this consequence by limiting her duty by the “ought implies can” principle would only push the problem back to that of the original modification. If she were only required to disclose those facts that she knows, we could once again ask: why think that the participant’s interests or inducements are protected when he understands everything the researcher knows rather than everything he could understand? There is no satisfactory answer here. After all, in the case under discussion, she does not know about risk C, and yet that is a fact that he could understand and which is relevant to his interests and inducements.
9 We say “can lead to invalid consent” because control is a matter of degree and it is plausible that some degree of illegitimate control, while still disrespectful of a person’s autonomy, is nonetheless compatible with his giving valid informed consent. As Faden and Beauchamp put it, valid consent tokens must be “substantially noncontrolled,” but this is not the same as entirely uncontrolled (Citation1986, 256–262).
10 This has an interesting implication for cases in which the recipient of consent ought to know certain facts. For example, a professionally negligent researcher may be ignorant of well-known risks of an experimental procedure, and so fail to disclose them. Her disclosure is problematic, as we argue later. However, she’s not guilty of fraudulent disclosure because she is not in a position to invalidate her prospective participant’s enrollment decision by way of illegitimate control with respect to the risks of the procedure. She is in no better epistemic position with regards to the information than he is even if she ought to be.
11 Note that the question of whether consent is valid in these cases is distinct from the question of whether it is permissible for the recipient of the consent token to proceed. If the physician has no reason to think that she has been misunderstood, then she may have good reason to think that she has been given valid consent, in which case she may be ethically permitted to insert the needle. In this case she would be innocently mistaken about the validity of her patient’s consent.
12 Cf. Sreenivasan’s characterization of the “ultra-minimal” understanding requirement (Sreenivasan Citation2019). Note that there will be cases in which what must be understood on our view and what must be understood on the interests or inducements view will coincide. For example, when a trial is properly and independently assessed and when it has a favorable risk-direct benefit ratio, Sreenivasan argues that the trial is in the participant’s clinical interests anyway, and so the participant need only understand “what it means to consent and a basic description of what they will undergo—injections, for example” (Sreenivasan Citation2003, 2018). While our view implies that the same information ought to be understood in this unusual case in which participants’ interests are assumed to already be protected by other means, the similarity is a feature of the example, not the underlying theory. The implications of the views will diverge in any case where research poses net risks.