by Jerry Menikoff, MD, JD
In a world with far too much dissonance, sometimes things nonetheless manage to come together. Such is the case regarding the article by Dickert and colleagues in this issue, “Partnering with Patients to Bridge Gaps in Consent for Acute Care Research”, and the recent changes in the Common Rule relating to improving research consent.
Dickert and colleagues’ work is a refreshing and very timely attempt at determining how we can improve informed consent for acute care clinical research trials. Their patient-centered approach was especially attuned to the challenges in these types of trials that are created by there being “time limitations, stress, and severe illness.” What I particularly want to highlight is how much overlap exists between the improvements suggested by the group put together by the authors, and the overall direction provided by the new Common Rule provisions regarding informed consent in general. Put more simply, the suggestions from Dickert’s group are even more useful than their article suggests. Many of those suggestions will be useful in a wide range of clinical trials—not just those involving acute care. And the reason is that they address broader, longstanding problems with consent forms.
Consider the scenario of a clinical trial involving randomizing a patient with a serious medical problem to one among two or more possible treatments. Unlike the scenarios considered by Dickert’s group, decisions about the interventions being considered would not necessarily have to be made within a very short period of time. So, for example, we could image someone with cancer, or with an immune disorder, or—not to ignore the stark reality society is currently confronting—an infectious disease that is difficult to treat.
Even though we are discussing a decision about participation in a research study, we can ask ourselves how different this decision is from a decision about clinical care. Perhaps not all that different: in both contexts, the patient will often be trying to determine what choice will best meet their goals in terms of treating their medical problem. The decision with regard to participating in the research study may be more difficult, due to greater possible uncertainty about the risks and benefits of the treatments, and the uncertainty and complexity created by the fact that the choice of a treatment is to be made by randomization. But the underlying structure of the problem confronting the patient is very similar in both scenarios.
Which suggests that we might look to how consent is obtained in the clinical setting as an analogy for determining how to obtain good consent for participating in a clinical trial. In this regard, it is noteworthy what the authors’ group concluded about research consent: they “felt that the consent form should mimic an effective conversation.” Which sounds very much like how we might conceptualize ideal consent in the clinical setting: a doctor sits down with the patient, explains to them about their medical problem and what their options are, and helps them work their way through a process of reasoning leading to a choice among those options. The doctor is not merely providing isolated pieces of information, but rather helping the patient digest and process what often is very complicated information about their condition, treatments options, and the often uncertain risks and benefits of one or another choice.
Compare the Dickert group’s desire for the consent form to function as an effective conversation about the clinical trial to what the revisions to the Common Rule say about informed consent. The information provided “must be organized and presented in a way that does not merely provide lists of isolated facts”. Rather, it should facilitate the person’s “understanding of the reasons why one might or might not want to participate.” There is quite a bit of similarity between how the Dickert group thought this should be approached, and the direction taken by the Common Rule’s revisions. Merely providing someone with “isolated facts”—particularly lists of them—isn’t all that helpful in improving understanding. So, just telling a person that here are the risks and benefits of various possible treatments—you figure it all out—is far from ideal. This is not the sort of conversation we would hope to have with a thoughtful and caring clinician. Rather, that clinician should ideally be putting the “isolated facts” together, and explaining how to use them to come to the best decision consistent with the person’s goals.
The Dickert group also came up with a variety of more specific recommendations with regard to the content of consent forms. Many of these—consistent with the fact that they were dealing with scenarios in which there would perhaps be only minutes before a decision had to be made—were of the “cut to the chase” variety. So, among other things, they were critical of too much “generic” or “irrelevant” information, particularly at the beginning of consent forms, which they viewed as critical “real estate” that should be used to the best advantage. They “felt that important information about the study itself and what it means to participate should be presented early.”
Given that the revisions to the Common Rule were not specifically targeting acute care research, it is striking how much they overlap with this aspect of the Dickert group’s recommendations (as that group itself noted). Under the revised regulations, consent forms “must begin with a concise and focused presentation of the key information that is most likely to assist a prospective subject … in understanding why one might or might not want to participate.” The initial pages of the consent form for any clinical trial—even if there is no time pressure to sign it—should ideally end up providing the most important information. Ideally, after reading those crucial pages, most prospective subjects would already be able to determine whether they likely do or do not want to participate.
As an example, consider how that initial portion of the consent form might deal with information about risks. In consent forms for cancer clinical trials, for example, it is not uncommon to see pages and pages of risk information about the several chemotherapy drugs that a patient might receive. For each of the drugs, there can be complicated charts with separate columns for risks based on whether they are very common, less common, or rare. In effect, there may end up being hundreds of “isolated” pieces of information, often encompassing almost all of the risk information that exists in the FDA-approved labeling information for the drugs.
Contrast this with the conversation that takes place between an oncologist and her patient, in the context of consent to clinical care: the physician will likely provide a short and relatively comprehensible synopsis about how “sick” the combination of drugs will likely make the patient, There will rarely be a hyper-detailed description of every risk spelled out for every drug, because that amount of information is not particularly digestible by most patients, let alone helpful to them in making a decision. As Dickert’s group observed, research consent forms had “too much information” about risks, which constituted an “overdisclosure [that] reduced clarity and potentially inflated perceived risks.” The new key and concise section at the beginning of consent forms, channeling a similar concern, will hopefully provide information about risks that is relatively short and meaningful, more like what the thoughtful oncologist would provide in the clinical setting (while allowing the remainder of the form to include more comprehensive risk information for those who might want it).
On the other hand, sometimes consent forms provide too little information about topics that are crucial to good decision-making. Dickert’s group identified this as a concern particularly with regard to information about benefits, including the use of “generic” statements such as telling subjects that they “may or may not benefit from participating in the study.” Such a statement conveys no information about the particular study at hand, even though it is often quite easy to provide very useful, specific information, in only a few sentences. In an oncology study, for example, it is hugely important for a prospective subject to know whether the current thinking is that the treatment being tested might at best minimize their suffering, or perhaps slightly extend their life, as contrasted with possibly curing their cancer. In my view, it is also important, as an ethical matter, to give a sense (where meaningful information is indeed available) of the likelihood of that possible benefit taking place. While I defer from commenting on exactly what the Common Rule does require in this regard, it is nonetheless noteworthy that the recent changes include a new requirement that information must be presented in “sufficient detail” to allow a person to understand the decision they are confronting.
Bottom line, by collecting the real-world wisdom of patients and others, Dickert and colleagues have put together some uncommonly good advice about how to make informed consent more ethical. Their article deserves a wide audience.