In 1974, the Center for Law and Health Sciences at the Boston University School of Law provided legal background papers on informed consent to research to the newly created National Commission for the Protection of Human Subjects of Biomedical and Behavioral Sciences. These papers were written by George J. Annas, the Center’s Director, as well as Barbara F. Katz and I, who were staff attorneys at the time. These papers can be found in the appendices to the Commission reportsFootnote 1 and in our book Informed Consent to Human Experimentation: The Subject’s Dilemma,Footnote 2 in which we present a refined version of those papers. This project introduced me to the world of human research ethics and the complexities of protecting the rights and welfare of research subjects.Footnote 3 Over the past fifty years, I have sat on Institutional Review Boards (IRBs), been a member of the FDA’s Pediatric Research Advisory Board, and engaged in varying activities related to human subject protection. During this time, I took for granted that consent forms were the best method for ensuring that subjects were thoroughly informed about all aspects of the proposed research. Like many IRB members, I spent considerable time reviewing, editing, and debating with other IRB members about the precise wording of these forms.
Having reviewed the expanding literature on the utility of consent forms, I have changed my mind. Even a cursory look at the literature on research consent forms demonstrates widespread recognition that they do not accomplish what they were supposed to.Footnote 4 They are too long, too complex, too full of irrelevant information, and often not understood, or even read, by potential subjects.Footnote 5
Of course, I still believe it is essential that human subjects be fully informed about what investigators are asking of them, but it has become increasingly clear that consent forms are a poor way achieving this goal, and a poor way to document that subjects have been properly informed by investigators. In this essay, I outline my concerns and offer an alternative that will seem radical to some and obvious to others.
Informed Consent to Therapeutic Interventions
Before “informed consent” to medical treatment, there was simply consent. Any invasion of a person’s body required consent from that person. Without consent, invasion of another person’s body constituted civil battery.Footnote 6 This legal requirement is exemplified by the much-quoted maxim from Schloendorff v. The Society of the New York Hospital that, “[e]very human being of adult years and sound mind has a right to determine what shall be done with his own body; and a surgeon who performs an operation without his patient’s consent, commits an assault, for which he is liable in damages.”Footnote 7 The lawsuit was an action for “trespass.”Footnote 8 Mary Schloendorff went to the hospital for a stomach ailment and was found to have a lump which appeared to be a fibroid tumor.Footnote 9 To determine the exact problem, the doctor suggested an “ether examination.”Footnote 10 Ms. Schloendorff consented to the “ether examination” but specifically told her doctor that “there must be no operation.”Footnote 11 Surgery was then performed despite these explicit instructions, after which serious complications ensued.Footnote 12 The issue presented was not the lack of informed consent but surgery with no consent at all.Footnote 13
In 1960, the Supreme Court of Kansas decided a case in which, unlike in Schloendorff, the patient consented to the bodily invasion but was provided no information about the risks of the procedure before doing so.Footnote 14 In Natanson v. Kline, Ms. Natanson was diagnosed with breast cancer and underwent a radical mastectomy.Footnote 15 After that procedure, it was recommended that she receive cobalt radiation treatment, a new and much more powerful intervention than previously used to treat cancer, and she agreed to the procedure.Footnote 16 Following this treatment, Ms. Natanson’s entire chest, skin, cartilage, and bone were completely destroyed.Footnote 17 While Ms. Kline’ negligence lawsuit raised the issue of whether the cobalt radiation treatment was properly administered, she also raised the issue of lack of informed consent.Footnote 18 Because she had consented to the intervention the doctors delivered, this was not brought as an assault case but as a negligence case alleging that Dr. Kline violated his duty to inform her of the risks associated with the treatment, even when the treatment is properly administered.Footnote 19 The Court ruled that “Dr. Kline was obligated to make a reasonable disclosure to the appellant of the nature and probable consequences of the suggested or recommended cobalt irradiation treatment, and he was also obligated to make a reasonable disclosure of the dangers within his knowledge which were incident to, or possible in, the treatment he proposed to administer.”Footnote 20 This case exemplifies how certain medical advances come with risks that are invisible and entirely unknown to patients. Without the disclosure by her physicians of the serious potential harms the intervention could cause, Ms. Natanson and similarly situated patients would be profoundly unaware of these risks and incapable of balancing these risks against the proffered benefit.
In the 1970s, courts developed more robust theories of informed consent and the doctor-patient relationship. In Cobbs v. Grant, the California Supreme Court held that, “as an integral part of the physician’s overall obligation to the patient there is a duty of reasonable disclosure of the available choices with respect to proposed therapy and of the dangers inherently and potentially involved in each.”Footnote 21 This obligation was part of the fiduciary relationship between doctors and patients:
A medical doctor, being the expert, appreciates the risks inherent in the procedure he is prescribing, the risks of a decision not to undergo the treatment, and the probability of a successful outcome of the treatment. But once this information has been disclosed, that aspect of the doctor’s expert function has been performed. The weighing of these risks against the individual subjective fears and hopes of the patient is not an expert skill. Such evaluation and decision is a nonmedical judgment reserved to the patient alone.Footnote 22
Patients, being unlearned in medical science, need to rely on their physicians to provide them with the information they need to weigh the risks and benefits of the proposed procedure so they can intelligently exercise their right to consent to or refuse treatment. Physicians, to act in their patients’ best interests, must affirmatively provide this information. Neither these cases, or similar cases requiring informed consent to medical treatment require, or even mention, consent forms.
Karp v. Cooley Footnote 23 falls between therapeutic and research interventions. Mr. Karp suffered from a severely diseased heart that required a surgical treatment using a “wedge” procedure.Footnote 24 If this procedure failed to treat his condition, Mr. Karp would require a human heart transplant to survive.Footnote 25 Unfortunately, no donor heart would be immediately available at the time the wedge procedure was to be performed.Footnote 26 Meanwhile, Dr. Denton Cooley had been developing an artificial heart.Footnote 27 At that time, his artificial heart had never been used in a human, though it had sustained the lives of seven calves for short periods of time.Footnote 28 Dr. Cooley suggested, in the event the wedge procedure failed, implanting the artificial heart to sustain Mr. Karp’s life until a donor heart could be implanted.Footnote 29 This scenario is precisely what transpired — the wedge procedure failed, and an artificial heart was implanted that kept Mr. Karp alive for sixty-four hours, at which time a human heart was implanted into him.Footnote 30 Mr. Karp died the next day.Footnote 31
Mrs. Karp brought a lawsuit which included a claim for failure to obtain informed consent from Mr. Karp. There was conflicting testimony regarding what Mr. Karp was told or knew about the use or efficacy of the artificial heart.Footnote 32 Referring to a form purporting to indicate Mr. Karp’s consent to the procedure, the court held, “[a]lthough not necessarily conclusive, what Haskell Karp consented to and was told is best evidenced by this document.”Footnote 33 Whether Mr. Karp read or understood this document could not be proved. His understanding ultimately was of no concern, since his signature on the form settled the matter of his “informed consent.”Footnote 34
Informed Consent to Research
Unlike informed consent to therapeutic interventions, which is generally unregulated,Footnote 35 informed consent to research subjects is strictly regulated by federal regulations.Footnote 36 This is the result of recognizing that research using human subjects, while a socially important activity, is also a troubling and suspect undertaking. Unlike medical treatment, in which benefit to patients is the goal of both the doctor and the patient, research subjects are a means researchers need to further their goal of creating new knowledge. The duty to obtain informed consent to therapy is derived from the fiduciary obligations physicians have with patients, but researchers are not in a fiduciary relationship with their subjects. If researchers had fiduciary obligations to subjects, it would be unlawful for researchers to put subjects at risk of discomfort or harm to benefit unknown others. In certain cases, there is some chance that research subjects may derive benefit by being in a research project, but that benefit is incidental and not the purpose of research interventions.
Given the history of well-documented abusive human subject research, federal regulation of research with human subjects was adopted to prevent the mistreatment of research subjects.Footnote 37
The Doctors’ Trials and the Nuremberg Code
The most atrocious abuses were committed by Nazi doctors, who claimed in their defense that their abhorrent acts constituted legitimate research conducted to benefit German war efforts during World War II and were lawful in the jurisdiction where the research was conducted.Footnote 38 The War Crimes Tribunal found these doctors guilty of crimes against humanity, which superseded the existing German laws.Footnote 39 Just following the law of the state was found to be no defense to the tortures and murders committed by doctors.Footnote 40 The law of humanity superseded the law of the state.
The “Doctors’ Trial”Footnote 41 led to the execution or imprisonment of the doctors who conducted the research.Footnote 42 In rendering their verdict, the Tribunal’s judges listed ten criteria that had to be met for research to be lawful.Footnote 43 This was not the act of a regulatory agency, but of a panel of judges guided by the American physicians Leo Alexander and Andrew Ivy.Footnote 44 The ten standards announced by the Tribunal are known as the Nuremberg code.Footnote 45 Only two of the criteria relate to subject consent and decision-making, with the rest creating obligations for physicians to protect research subjects from harm.Footnote 46 The first criterion is “[t]he voluntary consent of the human subject is absolutely essential.”Footnote 47 In terms of information this criterion requires
… before the acceptance of an affirmative decision by the experimental subject there should be made known to him the nature, duration, and purpose of the experiment; the method and means by which it is to be conducted; all inconveniences and hazards reasonably to be expected; and the effects upon his health or person which may possibly come from his participation in the experiment. The duty and responsibility for ascertaining the quality of the consent rests upon each individual who initiates, directs, or engages in the experiment. It is a personal duty and responsibility which may not be delegated to another with impunity.Footnote 48
This is far from an artfully written code of conduct due to its use of ambiguous terms and obligations. What is clear is the overall goal researchers must seek to attain. One reason for this drafting imprecision is that the judges answered a hypothetical question that was not an issue in the trial. Failing to obtain the minutiae of consent was not why these defendants were on trial. Informing someone that they would be frozen to death in icy water, or asphyxiated to death in a high-altitude chamber would need no detailed elaboration. Obviously, no one would voluntarily consent to being placed in such deadly scenarios. Consent would not make deadly, torturous research legitimate or legal. Failing to get consent is not why these doctors were executed or imprisoned. Assuming any part of the Nuremberg Code would apply retroactively to their atrocious research, the doctors would clearly be guilty of violating the provisions obligating researchers to not knowingly harm subjects regardless of consent.Footnote 49
The Nuremberg Code is not truly a “code,” in that it has not been formally adopted by a governmental body, but it remains an essential statement of ethics and human rights directed at protecting human subjects. Neither the term “informed consent” or a requirement for a written document is mentioned in the Code.Footnote 50
An important outcome of the Doctors’ Trial was the recognition that physicians who were trained and then trusted to diagnose and treat illnesses and to reduce suffering were capable of inflicting indescribable physical and mental torture on human beings. These German doctors were not outcasts from the German medical community. They were leaders of the medical profession and held highly esteemed positions in medical and academic institutions.Footnote 51 If they could inflict agony and death on human beings, any doctor could. The many attempts to explain how physicians could go from being healers to being torturers have been speculative and unsatisfying.Footnote 52
The Impact of the Tuskegee Study of Untreated Syphilis in the Negro Male
In 1972, Americans were shocked to discover that American doctors working for the U.S. Public Health Service (USPHS) were capable of conducting research they knew would lead to disability and death of a large portion of their subjects, as well as to serious illness in their subjects’ sexual partners and offspring. While the Tuskegee Study of Untreated Syphilis in the Negro Male (TSOUS), often referred to as the Tuskegee Syphilis Study, is well known for the overtly racist nature of the research and the effect of racism on the researchers and government officials who regularly reviewed and approved the continuation of this project, there is another important lesson to be learned.Footnote 53
The USPHS doctors who developed and conducted the TSOUS were originally sent to Macon County, Alabama, to treat impoverished Black sharecroppers who would not otherwise receive treatment in their community because they were Black and poor.Footnote 54 Even though no safe and effective treatment for syphilis existed, doctors would employ an intervention with arsenic therapy.Footnote 55 Many southern physicians, based on “research” published in southern medical journals, believed it was not necessary to treat Black men with syphilis because their alleged underdeveloped nervous system would not be affected by syphilis the same way White males’ were.Footnote 56 The USPHS doctors did not believe this racial difference was real, which was why they were sent to Alabama to provide treatment to Black people.Footnote 57 When the Depression led to falling federal government revenues, the USPHS doctors were no longer able to acquire drugs to treat syphilis.Footnote 58 Given this state of affairs, they decided they could now do a natural experiment which would prove that Black men suffer the same negative results of untreated syphilis as White men by following Black men with untreated syphilis.Footnote 59 Ultimately, they believed, this longitudinal study would disprove the belief that Black men could not benefit from treatment. In other words, the TSOUS started with the goal of proving the need to treat Black men infected with syphilis because if untreated they would suffer serious harm, but ended up withholding a safe and effective treatment, penicillin, when it became widely available in the early 1950s.
To recruit subjects, and to keep these subjects returning for the examinations this longitudinal research required, the researchers did not just fail to inform the subjects of the nature and requirements of the research, they intentionally lied to and defrauded the subjects. When subjects were subjected to spinal taps that were done solely for research purposes, they were lied to and told they were being treated for “bad blood” which supposedly was how syphilis was referred to in the Black community.Footnote 60
As immoral as this misconduct was, the most despicable part was the researchers’ failure to treat the subjects with syphilis upon the availability of the safe, simple, and effective treatment of penicillin.Footnote 61 The research was originally justified by its goal of proving Black men suffered severe consequences if treatment was not provided, but the researchers themselves withheld safe and effective treatment when it became available. Establishing the severe harms caused by nontreatment was the goal of the project.
The TSOUS is a blatant example of the differences between the goals of treating physicians and those of researchers trying to increase knowledge. A physician would quickly provide penicillin to patients infected with syphilis, but if these researchers provided penicillin to subjects it would ruin their research plan. Indeed, the researchers were so concerned about penicillin contaminating their research that they provided lists of their subjects to other local physicians their subjects contacted for care and asked them not to provide their subjects with penicillin to treat any illness curable therewith.Footnote 62 There is likely no starker reminder that researchers — unlike physicians with fiduciary obligations — do not act in the best interests of subjects than how these researchers mistreated their subjects. In the doctor-patient relationship where the well-being of the patient is the end sought, penicillin would have been immediately employed. Whereas in the research-subject context the subject is a means to an end, and providing treatment to subjects could be a disaster for the researchers.
While the TSOUS began before the advent of the Nuremberg code, it continued well after its creation.Footnote 63 In addition to violating the Nuremberg Code’s consent requirement, the researchers violated a number of other criteria, including that “[n]o experiment should be conducted where there is an a priori reason to believe that death or disabling injury will occur; except, perhaps, in those experiments where the experimental physicians also serve as subjects.”Footnote 64 Indeed, death and disabling injury were the outcomes the researchers expected and needed to prove their hypothesis. Like the Nazi research that led to the development of the Nuremberg Code, the core problem with the TSOUS was not the lack of informed consent. Any person who understood the nature and consequences of agreeing to be a subject of the TSOUS would not consent to it. The niceties of consent were not an issue.
Unlike the German physicians, the TSOUS researchers and the CDC reviewers who regularly reviewed and approved the study’s continuation suffered no consequences for causing the death and disability of research subjects.Footnote 65 In 1974, the survivors and descendants of the study received a small monetary settlement.Footnote 66 Over two decades later, in 1997, President William Clinton publicly apologized to the survivors on behalf of the United States.Footnote 67
The public and members of Congress were shocked by the 1972 disclosure of this experiment by an Associated Press reporter. It was front-page news across the country.Footnote 68 The result was the creation of the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, which was charged with making recommendations to the Department of Health, Education and Welfare.Footnote 69 These recommendations ultimately resulted in the creation of the Office for Protection from Research Risks,Footnote 70 as well as federal regulations that required institutions that receive federal funds for human research to comply with these regulations.Footnote 71 These regulations were directed at institutions that were the recipients of the funds and did not directly apply to individual researchers.Footnote 72
Research Consent Forms
The requirement for written documentation of informed consent forms that must be signed by the subject or lawful surrogate for research, is based on the same mistrust of researchers that led to the overall regulation of human subject research. Virtually everything that occurs to a hospitalized patient is documented. The administration of drugs, changes in dosage, the existence and change in symptoms, changes in diet and so forth, are meticulously documented by doctors, nurses and other professionals. Their documentation consists of notes in the patient record that is signed by the provider, not the patient.Footnote 73 For invasive procedures like surgery, patients are often required to sign consent forms that look nothing like research consent forms regarding detail or length, clearly for the sole purpose of trying to minimize legal risk to the institution. They are written by the hospital lawyers and no “ethics board” review is required.
Research consent forms, unlike consent, have no ethical weight. They are legal documents that institutions use as evidence of compliance with federal regulations, and thus function primarily as risk management tools. The main problem with informed consent forms is that they have come to be seen as a substitute for the informed consent process by researchers and government and institutional officials.
As previously noted, the Nuremberg Code does not mention consent forms. The original Declaration of HelsinkiFootnote 74 divides research into two categories: Clinical Research Combined with Professional Care and Non-Therapeutic Clinical Research.Footnote 75 In the former group, researchers, “[i]f at all possible, consistent with patient psychology, … should obtain the patient’s freely given consent after the patient has been given a full explanation.”Footnote 76 No written document is required for this category of research. In fact, consent itself is not required. For the latter category, “[c]onsent should, as a rule, be obtained in writing.”Footnote 77 There is no indication what the phrase “as a rule” means, but it is clear that a written consent form is not strictly required. This is an early demonstration of the conflation of informed consent and informed consent forms. Consent can be documented in writing, but consent (not just a form) is something a researcher must obtain from a potential subject for the research to meet ethical and legal requirements.
The recently adopted 2024 revision of the Declaration of Helsinki provides,
After ensuring that the potential participant has understood the information, the physician or another qualified individual must then seek the potential participant’s freely given informed consent, formally documented on paper or electronically. If the consent cannot be expressed on paper or electronically, the non-written consent must be formally witnessed and documented.Footnote 78
This provision does not require the use of a specific form, or mandate that the “formal document” include any of the elements of informed consent. Rather it requires informed consent to be “formally documented” on paper or electronically. A signed document that states, “I have been informed of the risks and benefits of being a subject in XXX research, and voluntarily agree the be a subject in this research,” would provide “formal documentation” on paper. The “research ethics committee” review does not mention reviewing or approving the “formal documentation.”Footnote 79
The earliest FDA requirements regarding consent documents are somewhat ambiguous.Footnote 80 The FDA required that information about risks, benefits, and alternatives must be imparted to patients.Footnote 81 The last sentence says, “[s]aid patient’s consent must be obtained in writing,” but there is no indication that the “writing” must contain any of the elements of informed consent.Footnote 82 Note that these requirements refer to “patients,” who consent to the “use” of an investigational drug, not subjects or participants who agree to enroll in a research study.
The Belmont Report, published by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research in 1979, does not require signed consent forms as an ethical imperative.Footnote 83
The Federal Rules
In 1974, two years after the public exposure of the United States Public Health Service’s syphilis study, the Department of Health, Education, and Welfare issued its first comprehensive rules for the “Protection of Human Subjects.”Footnote 84 These regulations required institutions that receive federal funding to establish mechanisms to protect human subjects, including the creation of “a committee or committee structure” that would review and approve all research with human subjects prior to such research being conducted.Footnote 85
This new regulation provided a definition of informed consent which listed six straightforward “basic elements of informed consent.”Footnote 86 An “organization” — not an individual researcher — is obligated to “obtain informed consent”Footnote 87 and, separately, to document the subject’s informed consent.Footnote 88 The written consent document was required to embody all the elements of informed consent and was to be submitted to the research review “committee” for its consideration.Footnote 89
This structure of institutional committee review (now called Institutional Review Boards (IRBs)) and the separate requirements to obtain and to document informed consent have survived the revisions of the human subject protection regulations. The most recent version contains a section titled “General Requirements for Informed Consent,” which requires an “investigator” (not the institution) to obtain the informed consent of a subject before involving the human subject in research.Footnote 90 The basic element of informed consent consists of nine specific items that must always be included, as well as an additional nine items to be included when appropriate.Footnote 91 A separate section, titled “Documentation of Informed Consent,” requires a written consent form that incorporates all the previously specified elements of informed consent.Footnote 92 The current consent and consent form requirements provide,
Informed consent must begin with a concise and focused presentation of the key information that is most likely to assist a prospective subject or legally authorized representative in understanding the reasons why one might or might not want to participate in the research. This part of the informed consent must be organized and presented in a way that facilitates comprehension.Footnote 93
This new requirement for “a concise and focused presentation of the key information” is in response to the widely recognized problem that has plagued consent forms, namely that they have become so long as to be unreadable and include details and employ language that are incomprehensible to lay people.Footnote 94 It is not clear why being concise and focused is not the requirement for the entire document, rather than just at the beginning of the document. The excess length and incomprehensibility of these forms have not been a secret. It is impossible to find a study of consent forms (there are a considerable number) that finds they are an effective means of communicating the elements of informed consent to potential subjects. Rather, signed forms have become the primary goal of the informed consent process because they “prove” subjects provided informed consent.
There is an abundance of studies that conclude consent forms are of little, if any, value in providing information to potential subjects. A study published in 2010 demonstrated both the increasing length of research consent forms, while also demonstrating their failure to effectively communicate important information from the research protocol.Footnote 95 The authors found that consent forms became longer over the years – from less than one page in the 1970s to four-and-a-half pages (which was once considered long) in the mid-1990s.Footnote 96 In their discussion, the authors note that “studies have shown an inverse relationship between length and individuals’ comprehension of the information provided.”Footnote 97 Length does not simply negatively affect comprehension but also subjects’ willingness to read lengthy forms at all.Footnote 98 The authors go on to suggest that we should consider new approaches to obtaining informed consent.Footnote 99
A recent study of consent forms in industry-sponsored research found the average page length of informed consent forms (specifically, English versions) was 22.0 (range, 10–45 pages), with an average word count of 7,915.9 (range, 3,754–14,199 words).Footnote 100 To put this in context, this essay has about 7,500 words, excluding footnotes.
One study compared subjects’ perceived comprehension of information when receiving the standard versus a short informed consent form.Footnote 101 The investigators found there was no material differences in perceived comprehension of the elements of informed consent between the groups.Footnote 102 The problem, however, is that fifty percent or fewer subjects in either group rated their understanding of the elements of informed consent as “good” (the other categories being “moderate” and “poor”).Footnote 103 However, the results showed that 100% of the short form users and 98.7% of the standard form users knew their participation in the study was voluntary — a simple, non-technical point to make in writing.Footnote 104
An excellent summary of the problem is,
Over the past 50 years, the informed consent process has become increasingly regulated and standardized, while the challenges remain persistent and hard to overcome. Consent forms are increasingly long and complicated, obscuring important details, and are often designed to serve the interests of institutions and sponsors. Data show that participants often have a limited understanding of study information even when they have signed a consent form.Footnote 105
In addition to the special problems presented by the length and incomprehensibility of informed consent forms, there exists a more general problem when relying on forms: “more than half of Americans between the ages of 16 and 74 (54%) read below the equivalent of a sixth-grade level.”Footnote 106
Why are Informed Consent Forms So Bad?
In the beginning of the modern research consent form era (i.e., the promulgation of the 1974 human subject protection regulations), the form was meant to “document” that the investigator obtained the informed consent of the new subject. This is not the only method one could require to document the performance of a task. For example, everything that happens to a hospitalized patient is documented using a very simple method – a member of the health care team writes it in the patient’s notes. These notes are not read or signed by the patient. A note in the medical record that an investigator met with a patient to discuss the patient’s participation in a specified research study, that the investigator provided the patient with the elements of informed consent, and that the patient agreed to participate would constitute documentation.
The reason this method of documenting research subject recruitment and consent is not satisfactory is because researchers cannot be trusted to treat potential subjects with respect, as demonstrated by the history of documented research abuses. Since we cannot trust researchers to accurately impart information that might lead a potential subject to decline participation, which would run contrary to the researcher’s interests, regulators have required the use of consent forms that contain all the information a researcher is supposed to provide. This form is supposed to “document” what a potential subject was informed of by the investigator, and not to impart the information itself. In reality, no one has any idea what potential subjects are told by investigators. But it is highly unlikely that an informed consent document accurately documents what actually transpired between the investigator and potential subject, which is its supposed purpose.
As already mentioned, informed consent forms are risk management tools.Footnote 107 Institutions can “prove” to federal auditors that they have obtained informed consent from every subject by producing these signed forms, and auditors can assure their superiors that they saw signed, IRB-approved forms for research subjects. What the subject knew, or even was told, is impossible to ascertain from the forms. The signature on the bottom of the form is all that is required to meet the needs of federal auditors and research institutions.
They are also risk management tools to prevent subjects from successfully claiming in a lawsuit that they were not adequately informed about the research or their rights regarding the research. This is a lesson from Karp v. Cooley, discussed above – the signed consent form resolved any dispute about what Mr. Karp knew or consented to.Footnote 108 But signed consent forms do not always prevent successful lawsuits.Footnote 109 Even if they did, it is an extremely rare event for researchers and institutions to be sued as a result of conducting research. While informed consent forms are explicitly forbidden from using exculpatory language,Footnote 110 the form itself is designed to serve as an exculpatory document.
The perceived utility of consent forms as risk management tools led to them to be packed with other releases unnecessary to obtain consent to the research at hand, such HIPAA waivers or consent to the use of tissue donations in separate research by different researchers.
As one commentator put it,
[W]e have long known that an overwhelming majority of the public does not think of informed consent as something that aims to make them better decisionmakers but rather as a means used by physicians to protect themselves against liability. Instead of being a process of discussion that aims to reach a mutually agreed way forward, informed consent is equated with a document — the consent form. It is but one of a tidal wave of forms that patients sign (on paper or, increasingly, electronically) before being admitted to a medical facility … .
Patients are right to perceive that the consent form is all about foreclosing liability. After all, they can see as much from the forms themselves, which usually run on for mind-numbing page after mind-numbing page, filled with avowals of the benefits of the study alongside descriptions of every bad thing that could possibly happen. This means that the eventual act of signing such a form is often little more than a ritual, like clicking on “Accept” to download the newest App to one’s smartphone.Footnote 111
One important reason forms are an ineffective means of communication is that one form is supposed to serve a large array of potential subjects with widely different levels of education, reading aptitudes, awareness of scientific, biological concepts, and so forth. That is a lot to ask of one document.
Very smart and caring people have worked for decades trying to make consent forms better. The NIH, for example, has created an online resource page to help investigators produce better consent forms.Footnote 112 It has links to resources that include suggested language that it thinks will make the forms more readable.Footnote 113 Unfortunately, one section is “titled” “Getting your Consent Form (ICF) Approved Quickly,”Footnote 114 which clearly has appeal to researchers although this should not be the goal of those writing consent forms. The page also links to templates that it says complies with the revised research regulations.Footnote 115 The template for informed consent forms to use at the NIH Clinical Center is fifteen pages long before information about a specific study is entered into the template.Footnote 116 Recall the Office of Human Research Protection’s (OHRP) desire for conciseness.Footnote 117
These problems are not exclusive to consent forms. They share the same problems as virtually all forms people regularly sign. Forms are not written to protect the signers but to protect the users of forms. If you read your rental car agreement (and you know you do not) it is to serve the interests of the rental car company, not you. Internet companies want you to click that you read and agree to their terms (even though you haven’t) because it protects them, not the clicker.Footnote 118 If one’s interest is in educating people, forms are not how to achieve this. Schools and teachers do not use forms as educational tools. Books, pamphlets, videos, lectures, discussions and so forth are the means to provide information, not forms that must be signed on the dotted line. And if one is interested in accurately documenting what a specific person knows, or was told about a particular matter, forms are a poor way to accomplish this.
If Consent Forms Are Bad, What Is the Alternative?
In 2001, the National Bioethics Advisory Committee (NBAC) addressed this question. Recommendation 5.3 says, “…[f]ederal policy should require investigators to document that they have obtained voluntary informed consent, but should be flexible with respect to the form of such documentation[.]” Footnote 119 Its commentary to this recommendation says,
Although the federal regulations may have been attempting to reflect a legal standard for documentation of informed consent, NBAC is aware of no case law in which a signed, written consent form is required. To fulfill the substantive ethical standard of informed consent, depending upon the sort of research being proposed, it may be more appropriate to use other forms of documentation, such as audiotape, videotape, witnesses, or telephone calls to participants to verify informed consent and participation in the research study. Footnote 120
Why this sensible recommendation, made twenty-three years ago, has been entirely ignored is unknown to me. It is also unfortunate.
In 2013, the Supreme Court of India went further and ordered the government to videotape the clinical trials of five new drugs.Footnote 121 This was in response to allegations that multinational corporations were “exploit[ing] loopholes to make India the testing ground of their new drugs.”Footnote 122 High levels of suspicion require high levels of proof.
One reason the NBAC’s recommendation may have been ignored is because of the lack of inexpensive, easy to use, and unobtrusive technology to document informed consent in some way other than a written form. That is no longer the case.
I agree with the NBAC’s recommendation and propose that the informed consent interaction between investigators and potential subjects be audio-recorded.Footnote 123 Unlike in 1981, when the NBAC made its recommendation, this practice would not require audiotapes, microphones, or bulky and intrusive recording equipment. It requires only a mobile phone or small digital recorder. These recordings can be easily preserved on a computer or central drive and a copy could be made available to subjects who desire it. Unlike consent forms, this would precisely document an investigator’s informed consent discussion with each potential subject, as well as the questions asked by the potential subject and the investigator’s answers. Investigators could ask the potential subjects whether they understood what was being said, if they would like more or clearer information, or ask them to repeat their understanding of a particular item. It would document one-on-one instruction, which is the core of informed consent, and reading willingness or aptitude would be of no concern.
I also propose that standard written informed consent forms should not be allowed to be used as an alternative to such recordings. It has been repeatedly demonstrated that these forms are a poor way of providing or documenting the informed consent transaction.Footnote 124 Given this, it makes no sense to use such forms. Audio recordings would be a substitute for written forms, not an alternative.
Investigators obtaining audio consent would be guided by a checklist of the basic elements of informed consent that must be discussed. This would not be a script to be read verbatim, since an important goal is to give investigators flexibility to adapt to the learning and communication needs of the individual. At the end of conversation, the investigator should provide short clearly written materials such as a pamphlet containing brief reminders of what was covered during the conversation, and who to contact if any concerns or questions arise. The potential subject should also be informed that they can have a copy of the recorded discussion provided them.
For the first time, IRBs and institutional officials would be able to accurately ascertain the quality of informed consent in the institution. IRBs would not need to review every audio recording of the consent process, but could sample them from time-to-time to determine not just compliance but also the need for additional training for personnel who obtain consent. These recordings could be used to train other personnel once the subject’s identity has been removed from the recording.
Institutions might oppose this suggestion because of their fear of federal auditors whose job is to determine if institutions are complying with federal requirements. This is an entirely understandable concern. For audio-recorded consent to be acceptable to researchers and organizations, federal regulators, who would sample these informed consent discussions as part of an audit, must agree not to focus on every word or phrase of these discussions, as they have done when reviewing written forms.Footnote 125 Rather the auditors should determine if the investigator-potential subject discussion, in its entirety, fairly provided the potential subject with the information needed to make an informed decision.
Institutional lawyers will complain, without evidence, that this will open a new avenue for suing institutions and researchers. There is no evidence that lawsuits based on lack of informed consent to research is a significant problem, let alone lawsuits where the consent form saved the day for the defendants.Footnote 126 Indeed, such a concern would demonstrate that while signed consent forms may not provide the subject with useful information, all that is truly desired is a signed form to protect the institution. It also assumes that the institution’s researchers regularly do such a poor job of obtaining informed consent from potential subjects that the researchers are the problem. The institutional goal should be improving the consent process, not hiding its shortcomings.
This theoretical liability concern assumes that if the theoretical subject testifies at the theoretical trial that he did not understand the long and complicated form (and there is a lot of literature that proves this subject is not alone),Footnote 127 that a signed consent form — which the imaginary jury of laypeople would also not understand — is somehow superior evidence compared to an audio recording of the discussion between the investigator and the subject. If done properly, the jury would hear an investigator present the basic elements of informed consent in terms the potential subject (and imaginary jurors) can understand, the investigator offer to further explain anything the potential subject says is unclear, and the investigator answering any specific question the subject asks. I posit this would impress jurors more than a lengthy, incomprehensible form.
Investigators and institutions might complain this process would take too much time. If investigators were already properly informing potential subjects of the elements of informed consent, this process should take no extra time at all. The audio-recording would not add any additional length to the informed consent discussion. On the other hand, because they would now know that their conversations with potential subjects may be heard by authorized persons in the institution and federal auditors, investigators may be a bit more punctilious in their conversations. This may take a bit more time, but trying to ensure potential subject comprehension is the goal of the informed consent process, and doing so properly takes time.
I imagine someone will accurately point out that I have no evidence that this proposal would work any better than the current system. That is largely correct, but there is some evidence that talking works better than forms.Footnote 128 I would look forward to social scientists conducting additional research on this question.
A very serious barrier is that current federal regulations require the use of some sort of written form.Footnote 129 The possible good news is that the regulations permit the use of a “short form” that comes very close to what I am proposing.Footnote 130 The short form allows the “basic elements of informed consent” to be presented orally.Footnote 131 The regulations require IRBs to approve “the written summary [not a script] of what is to be said to the subject.”Footnote 132 This is similar to the checklist I propose to help those obtaining informed consent to cover the essential topics. The current regulations require the presence of a witness to the presentation.Footnote 133 After the presentation, the subject must sign the “short form,” the witness must sign the short form and the summary, and the “person actually obtaining consent” must sign a copy of the summary.Footnote 134 Additionally, a copy of the summary must be left with the subject.Footnote 135
This seems to be an improvement over the use of long, complicated forms. However, the role of the witness is unclear. The regulations do not specify who can act as a witness or the witness’s purpose. Can it be a family member or friend who is also not scientifically trained? All the witness can do is verify that this process occurred. A lay witness likely cannot verify the accuracy or completeness of the information provided. While better than the long form, the use of the “short form” shares the problem that it provides no meaningful evidence of what was actually discussed during the informed consent discussion.
If regulators were willing to be flexible and open to the audio-recording proposal, they could adopt a position (perhaps in a guidance) that a recording of the informed consent process could serve as a substitute for a witness. Indeed, a recording is more reliable than a witness in assuring this process was properly conducted. What I have referred to as a “pamphlet” could be deemed to meet the regulatory requirement of a “written summary,” so long as it is short, simple, and clear.
Again, this requires that the OHRP and IRBs be willing to allow flexibility in the informed consent process. Specifically, they would need to be willing to take a more holistic approach to the content of the investigator-subject discussion, including with respect to the terms the investigator uses, so that the person obtaining consent can use different words, illustrations, and so forth to try to aid the subject’s understanding.
It is not clear why using the long form has continued to be the default method for documenting informed consent. If OHRP believes the short form process is effective in obtaining informed consent, as it must, what is the point of using incomprehensible long forms?
The greatest barrier to implementing this proposal is institutional inertia. This is the likely reason the NBAC recommendation discussed above was not adopted or even given serious consideration. Consent forms currently meet the needs of institutions, researchers, and federal regulators, so why rock the boat? Even if recording the informed consent discussion proved to be a better method of obtaining and documenting informed consent, changing the status quo that has existed for over fifty years requires a huge reconceptualization of the informed consent process. Consent forms have become deified, part of an important ritual, and doing away with them may be viewed as an act of heresy.
Conclusion
It is time to change what we (including me) thought was a good idea for over fifty years when the intervening evidence proves otherwise. During the COVID pandemic, and in the face of influence from anti-vaccine propagandists, it was repeatedly proclaimed that we must “follow the science.”Footnote 136 And we should. There now exists a body of science which provides powerful evidence that consent forms do not serve the admirable goals they were created to serve — protecting the rights and welfare of research subjects by assuring they are provided with the information they need to intelligently exercise their autonomy rights when deciding whether to become a research subject. Since the science shows consent forms do not serve the goals they are meant to accomplish, we should cease using such forms and try another means to accomplish the worthy goal of ensuring and documenting informed consent.
We should stop trying to fiddle with these forms to make them better. Fiddling has not worked. The NIH and others should stop providing advice to investigators about how to write better consent forms. Better consent forms are not supposed to be the goal of the informed consent process. A better consent process and corresponding documentation should be the goals.
My proposal to audio record the informed consent conversation will be met with opposition and controversy. I look forward to this, to the extent it provides an opportunity to discuss why this proposal is flawed, how we can better assure that subjects receive the information they need and deserve, and how to better document that this information was truly provided. This should be our shared goal.
Open access
