Informed Consent for Clinical Research Involving Patients With Chest Disease in the United States FREE TO VIEW

Informed consent has been defined by Beauchamp and Childress¹ as “an autonomous authorization of individuals of medical interventions or involvement in research.” Terry² discussed informed consent in clinical medicine in an earlier article in this CHEST series on biomedical ethics. In this article, I discuss informed consent in clinical research, focusing on the evolution of the concept and its incorporation into federal guidelines in the United States. Although these guidelines apply to all adult and pediatric patients, I refer to research involving patients with chest disease whenever possible.

The concept of informed consent is supported by the right of consent and refusal contained within English and American common law. Despite its long history, this concept was not used for either clinical or research purposes until the 20th century. Thomas Percival's³ historical Medical Ethics, the first work to contain the term medical ethics within its title when it was published in 1803, makes no mention of obtaining consent from patients before procedures or other interventions. Similarly, the American Medical Association⁴ did not include considerations of informed consent when it published its first Code of Ethics in 1847.

The legal obligation of physicians to obtain consent before treating patients was established in the United States in the landmark decision in Schloendorff v Society of New York Hospitals⁵ in 1914. In this case, the Court of Appeal in New York stated that “Every being of adult years and sound mind has the 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, except in cases of emergency where the patient is unconscious, and where it is necessary to operate before consent can be obtained.”

The requirement of consent for clinical research was first adopted not in civilian courts in the United States but by American judges and physicians in Nuremberg, Germany, in 1947. Sitting in judgment of Nazi doctors accused of committing barbarous acts in the name of research on human subjects during World War II, these officials not only prosecuted the Nazis' actions but also developed the Nuremberg Code⁶ to help prevent their recurrence. The Code combined the Hippocratic obligation of physicians to “do no harm” with their responsibility to obtain informed consent that is not addressed in the Hippocratic corpus.

The first principle of the Nuremberg Code is “The voluntary consent of the human subject is absolutely essential. This means that the person involved should have legal capacity to give consent; should be so situated as to be able to exercise free power of choice, without the intervention of any element of force, fraud, deceit, duress, overreaching, or other ulterior form of constraint or coercion; and should have sufficient knowledge and comprehension of the elements of the subject matter involved as to enable him to make an understanding and enlightened decision.”

Following publication of the Nuremberg Code, the World Medical Association⁷ published its own Recommendations Guiding Physicians in Biomedical Research Involving Human Subjects, which are better known as the Declaration of Helsinki. Adopted at the 18th World Medical Assembly in Helsinki, Finland, in 1964 and amended at other assemblies thereafter, the Declaration has as its first and second principles the necessity of scientific validity of research and committee review of experimental protocols.

As Shuster⁸ notes, the World Medical Association was accused of trying to subvert the Nuremberg Code and distance physicians from Nazi medical crimes by attempting to have peer review supplant informed consent as a first principle in clinical research. Nevertheless, the Declaration of Helsinki acknowledges the Nuremberg Code's authority regarding consent and does not undermine it. Moreover, the documents together serve as models for US federal research regulations, which require both the consent of subjects and peer review of proposed research by institutional review boards (IRBs) before it can be performed.

The federal agencies most involved with human subjects research in the United States are the US Food and Drug Administration (FDA), which regulates experimental drugs, devices, and biologics; and the National Institutes of Health (NIH), which conducts and sponsors basic and clinical research. The Office for Human Research Protections (OHRP; formerly the Office of Protection from Research Risks [OPRR]) develops regulations for the protection of human subjects and oversees compliance with them by institutions and their IRBs. Much of the information covered in this article is available through the OHRP Web site (www.hhs.gov/ohrp.htm).

The FDA, NIH, and OHRP are part of the US Public Health Service, which itself is a division of the Department of Health and Human Services (HHS), formerly the Department of Health, Education, and Welfare (HEW). Clinical research conducted at all institutions that are funded by federal agencies is governed by the HHS Code of Federal Regulations (45 CFR 46, abbreviated herein as CFR) for the Protection of Human Subjects.⁹ The CFR is a living set of administrative laws that have evolved over time.¹⁰

Federal oversight of clinical research can be traced to 1962 when Congress passed legislation mandating that informed consent be obtained from patients involved in studies seeking FDA approval of new drugs.¹¹ In 1964, the NIH sponsored a survey of its grantee institutions and learned that only 9 of 52 departments of medicine had policies regarding the rights of research subjects.¹² In 1966, Beecher¹³ called for such policies in an article, “Ethics and Clinical Research,” in the New England Journal of Medicine. During the same year, the NIH created its first Policies for the Protection of Human Subjects.¹⁴

In 1972, the New York Times¹⁵ disclosed that uneducated black men from Tuskegee, Mississippi, had been enrolled in a longitudinal study of syphilis but not treated for the disease. The Tuskegee Study showed that unethical experimentation was not limited to Nazi Germany, and it highlighted the need for stronger oversight of clinical research in the United States. It prompted Congress to pass the National Research Act of 1974, which mandates IRB approval of studies conducted or funded by the federal government and established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.

In 1979, the National Commission¹⁶ issued its seminal Belmont Report, which was developed at a meeting in Belmont House, a conference center of the Smithsonian Institution in Elkridge, MD. The Belmont Report is the American equivalent of the Nuremberg Code and the Declaration of Helsinki, although it introduces original concepts. As described by Jonsen,¹⁷ one of the National Commission's members, this document remains the clearest statement of how biomedical research involving human subjects is to be conducted in the United States.

The Belmont Report begins by describing the boundaries between research and clinical practice. Thus research is performed primarily to “test a hypothesis, permit conclusions to be drawn, and thereby to develop or contribute to generalizable knowledge” that may help future patients. “Practice,” on the other hand, “refers to interventions that are designed solely to enhance the well being of an individual patient or client and that have a reasonable expectation of success.”

The Report then explores the ethical principles relevant to clinical research, anticipating the publication of Beauchamp and Childress¹ (Beauchamp was a staff member of the National Commission). These principles include the following: (1) respect for persons (incorporating the conviction that individuals should be treated as autonomous agents and that persons with diminished autonomy are entitled to protections when participating in research), (2) beneficence (and its correlate, nonmaleficence), and (3) justice (particularly in the sense that persons should receive the benefits and bear the burdens of research equally).

After the National Commission disbanded in 1978, responsibility for implementing its recommendations in the Belmont Code and other writings was delegated by the secretary of HEW to the director of OPRR. The OPRR then organized a drafting committee that finalized the CFR. The CFR and the parallel FDA research regulations were signed by the secretary of HEW in 1981. In 1991, the revised Subpart A of the CFR was adopted by 16 federal agencies, including the NIH, and became known as the Common Rule. The FDA subsequently modified its own regulations to conform to the Common Rule.

Subpart A of the CFR contains the basic policy for the protection of human subjects. It begins with an iteration of the requirements for review and approval of research protocols by institutions engaged in research that is conducted or supported by any federal department or agency (most institutions apply the same requirements to nonfederally supported research). The institutions must provide written assurance that they will meet these requirements. These requirements include: (1) a statement of the principles governing the institutions in their responsibilities to protect research subjects, (2) designation of as many IRBs as necessary to oversee research, (3) lists of IRB members, (4) formal procedures for reviewing and revising research protocols, and (5) procedures for prompt reporting to IRBs, institutional officials, and sponsoring departments or agencies of any unanticipated problems involving risk to research subjects, investigator noncompliance with institutional procedures, or suspension or termination of IRB approval.

Subpart A places most of the responsibilities for oversight of research involving human subjects on individual institutions, whose IRBs carry out the review process. Nevertheless, once unanticipated problems, procedural noncompliance, or suspension or termination of IRB approval is reported from the institutions or from outside sources, further review can be conducted or requested by sponsoring departments or agencies such as the NIH or FDA, and by the OHRP.¹⁸ These organizations can suspend or terminate studies and can fine institutions for failing to honor the assurances they provided under Subpart A.^19,20

To approve research covered under Subpart A, IRBs must determine that the research meets several requirements. These requirements include: (1) “risks to subjects are minimized by using principles which are consistent with sound research design and which do not unnecessarily expose subjects to risk,” (2) “risks to subjects are reasonable in relation to anticipated benefits, if any,” (3) “selection of subjects is equitable,” (4) “informed consent will be sought,”(5) “informed consent will be appropriately documented,” (6) data monitoring will be used when appropriate, and (7) subject privacy will be protected.

Subpart A also contains general requirements for informed consent by subjects or their surrogates. In this regard, Subpart A states that “no investigation may involve a human being as a subject in research … unless the investigator has obtained the legally effective informed consent of the subject or the subject's legally authorized representative (LAR)” A LAR is defined as “an individual or judicial or other body authorized under applicable law to consent on behalf of a prospective subject to the subject's participation in the procedure(s) involved in the research.”

Informed consent can be sought only under circumstances that provide potential subjects or their LARs “sufficient opportunity to consider whether or not to participate and that minimize the possibility of coercion or undue influence.” The information given to subjects or their LARs must be in language understandable to them. Consent usually must be in written form, and those consenting must receive a copy of the consent form. When oral consent is used, it must be witnessed by a third party who, along with the study investigator, must sign a written summary of what was said to the subjects or their representatives.

Subpart A allows that consent be waived in certain circumstances. Thus “An IRB may waive the requirements to obtain consent provided … that (1) the research involves no more than minimal risk to the subjects; (2) the waiver … will not adversely affect the rights and welfare of the subjects; (3) the research could not practicably be carried out without the waiver; … and (4) whenever appropriate, the subjects will be provided with additional pertinent information after participation.”

“Minimal risk” is defined in Subpart A as “the probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves … than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.” The “daily life” of most hospitalized patients with chest disease is far different from life outside the hospital, and most of the procedures they go through are more complicated than “routine physical or psychological examinations or tests.” As a result, research involving such patients usually presents more than minimal risk unless it consists of surveys or chart reviews.

In addition to providing the basic policy for research protections contained in Subpart A, the CFR includes additional subparts that provide protections for research subjects who are “likely to be vulnerable to coercion or undue influence.” These subjects include pregnant women, human fetuses, and neonates (Subpart B); prisoners (Subpart C); and children who have not obtained the legal age for consent (Subpart D). Children are considered vulnerable both because of their physical and psychological immaturity and their inability to consent, or refuse, for themselves.

Under Subpart D, research involving no greater than minimal risk may be conducted in children with their parents' consent and, if possible, their own assent (an affirmative agreement to participate, such as saying or nodding “yes”). Research involving greater than minimal risk but presenting the prospect of direct benefit to the individual subjects can be performed only if consent and assent are solicited, the risk is justified by the anticipated benefit, and the relationship of the risk to the benefit is as favorable as that presented by available therapies.

Research involving greater than minimal risk and no prospect of direct benefit can be conducted only if consent and assent are solicited, the risk represents a minor increase over minimal risk, the research intervention or procedure presents experiences commensurate with those inherent in the clinical situation, and the intervention or procedure is likely to yield generalizable knowledge of “vital importance for the understanding or amelioration of the subjects' disorder or condition.” Research that does not meet these criteria can only be conducted if approved by the secretary of HHS after consultation with a panel of experts.

A “minor increase over minimal risk” is not defined in Subpart D. A reasonable definition might be that any potential harms associated with a research procedure are transient and reversible, or the likelihood of incurring such harms is remote. Under this definition, a research procedure that yielded information potentially beneficial to a child might be permissible if parental consent was obtained. Thus obtaining a series of blood samples for novel microbiological analysis using DNA probes might be permitted in children with known infection because it could aid in their diagnosis and treatment. However, drawing repeated samples from healthy control children would not be allowed because these subjects could not benefit. The procedure could be performed only if it yielded important generalizable knowledge, for example, if the microbiological analysis helped determine the extent of a carrier state during a deadly pandemic, and was approved by the secretary of HHS.

In effect, Subpart A of the CFR allows adult subjects to participate in clinical research without regard to the potential benefits and risks inherent in it if IRBs approve the research and either the research presents only minimal risk or the subjects consent to participate. Most adult patients with chest disease who are not critically ill are capable of consenting for themselves and do not require LARs for this purpose. Furthermore, IRBs can approve research that provides no direct benefit and risk above a minimal level, although they are required to consider possible risks and benefits in granting approval.

Critically ill adults generally cannot make decisions and provide informed consent because their underlying diseases and the psychoactive drugs they received impair their decision-making capacity.¹⁰ Their decisional impairment is similar to that of some patients with psychiatric (eg, schizophrenia) and neurologic (eg, dementia) disorders. According to Subpart A, decisionally impaired patients can only be enrolled in research through the consent of LARs. As previously noted, a LAR is defined as “an individual or judicial or other body authorized under applicable law to consent on behalf of a prospective subject to the subject's participation in the procedure(s) involved in the research.”

Although “applicable law” is not defined further in Subpart A, the OHRP has given such examples as a state statute, regulation, case law, or formal opinion of the state attorney general that addresses the issue of surrogate consent to “medical procedures.”²¹ Thus an “applicable law” that empowers surrogates to provide consent in the clinical context seemingly could allow them to consent for the “procedure(s) involved in the research.” In keeping with this interpretation, many IRBs in states with statutes sanctioning surrogate consent for medical treatment have relied on family members to serve as surrogate decision makers for research.²²

In 2000, the ARDS Network (ARDSnet)²³ reported the results of a clinical trial demonstrating that mechanical ventilation with low tidal volumes was associated with improved survival in ARDS patients. Shortly thereafter, the OHRP received a complaint that subjects had been enrolled in the ARDSnet trial at the 12 study sites in the absence of “applicable law.”²⁴ The OHRP then requested the IRBs at the study sites to justify the consent procedures they had allowed for the trial.²⁵

In most instances, the OHRP determined that the IRBs' procedures were appropriate under the IRBs' interpretation of state laws, including those allowing surrogate consent for clinical purposes. Nevertheless, several of the study sites took measures to develop laws specifically for research. At the urging of the University of California, for example, California enacted new legislation in 2003 allowing surrogate consent for decisionally impaired patients using a hierarchy of family members.²⁶

Review of state laws potentially applicable to consent clinical research at the time California's new legislation was passed revealed wide heterogeneity that persists today. Indeed, most states do not have laws similar to the California statute extending surrogate consent for patients with decisional impairment to the research context, and some do not have laws sanctioning surrogate consent even for clinical purposes. Critical care research generally is not conducted in states that do not sanction surrogate consent for clinical purposes unless subjects have other LARs (eg, court-appointed guardians).

Because research may benefit present individual patients and society as a whole, state laws allowing surrogate consent for decisionally impaired adults are desirable. At the same time, because decisionally impaired patients are vulnerable, it may also be desirable to provide protections for them beyond those that could be incorporated into state laws and even those called for in Subpart A. Such protections could be provided by development of a new subpart, comparable to that of Subpart D, focusing specifically on decisionally impaired patients, critically ill or otherwise.^22,27

However, because decisionally impaired adults are not necessarily as vulnerable on a physical and psychological basis as children are, the protections offered adults by a new subpart of the CFR might inhibit research unnecessarily. Other approaches would be to recommend a more rigorous risk-benefit analysis of research or of LAR qualifications by IRBs under the requirements of Subpart A. The Subcommittee on Inclusion of Individuals with Impaired Decision-Making in Research of the Secretary of HHS's Advisory Committee on Human Research Protections presently is evaluating these approaches and the possible usefulness of a new subpart to the CFR.²⁸

Glassberg and colleagues²⁹ recently demonstrated that enrollment of subjects in ARDSnet studies was lower at a public hospital than that at a university medical center in San Francisco because proportionately more patients with ARDS lacked surrogates at the former facility. Although trial enrollment would be enhanced if investigators consented themselves for decisionally impaired surrogates for whom surrogates were either not available or authorized under “applicable law,” the Common Rule does not sanction this practice, and the OHRP is not likely to favor it in the near future.

Even if critically ill patients have LARs under “applicable law,” the LARs may not be available in time to consider consenting for research in emergency situations, for example, immediately following cardiopulmonary arrest. Furthermore, although the Common Rule permits waiving the requirement for informed consent when the research involves no more than minimal risk to the subjects, its definition of “minimal risk” would preclude investigations of emergency interventions, such as cardiopulmonary resuscitation, that are potentially harmful and certainly not routine “in daily life.”

Given this situation, clinical investigators in the United States³⁰ once advocated a mechanism called “deferred consent” to conduct emergency research. Under this mechanism, patients were entered into studies without their consent, and they or their surrogates were informed as soon as possible after an experimental intervention. However, deferred consent was criticized on the grounds that true consent cannot be obtained for procedures that have already occurred.³¹ In 1993, both the OPRR³² and the FDA³³ cited this criticism in questioning the ethical and legal propriety of deferred consent.

In response to this situation, the Coalition Conference of Acute Resuscitation and Critical Care Researchers³⁴ argued in 1995 that patients who could not give personal or surrogate consent for emergency research were vulnerable “not only to research risks, but also to being denied potentially beneficial therapy when there is no known effective treatment for their life-threatening condition.” The Coalition members also proposed that the minimal risk standard be replaced by one of an “appropriate incremental risk” standard relative to that condition rather than to “daily life.”

In keeping with the proposed appropriate incremental risk standard, the CFR was amended in 1996³⁵ to allow research to be conducted in emergency settings without patient or surrogate consent if “(1) the human subjects are in a life-threatening situation …, (2) obtaining consent is not feasible …, (3) participation in the research holds the prospect of direct benefit to the subjects …, (4) the clinical investigation could not practicably be carried out without the waiver, (5) … the investigator has committed to attempting to contact a legally authorized representative for each subject within (the therapeutic) window …, (6) the IRB has reviewed and approved informed consent procedures …, and (7) … consultation with representatives of the communities in which the clinical investigation will be conducted and from which the subjects will be drawn.”

The provision for consultation with community representatives contained within the CFR waiver for emergency research might be problematic, particularly because “consultation” and “representatives of the communities” are not defined. However, studies in 2006 and 2008^36,37 reveal that consultation based on telephone surveys of and public meetings with potential clinical trial enrollees is logistically feasible. The studies also show that many respondents would not choose to participate in emergency research or provide consent for family members despite knowledge about the specific studies in which they might be enrolled.

Despite these findings, the waiver has facilitated research in outpatient and emergency department settings involving patients who have suffered conditions such as trauma and cardiopulmonary arrest. Yet because few hospitalized patients with chest disease face true emergencies, they would not be considered eligible for clinical research under the terms of the waiver. For example, studies of new modes of mechanical ventilation have a relatively long therapeutic window during which obtaining consent from surrogates may be possible.

Although deferred consent and a waiver of consent for emergency research are not widely used in the United States, these approaches have been endorsed by members of the Canadian Critical Care Trials Group and Australian and New Zealand Intensive Care Society Clinical Trials Group as methods of improving subject enrollment in a recent survey conducted by Cook and colleagues.³⁸ The same survey respondents also endorsed allowing investigators to consent for patients without surrogates. How often the respondents actually employed these methods is not clear from this survey. Nevertheless, deferred or waived consent were used to improve enrollment in several recent clinical trials^39,40 outside the United States.

Clinical research involving patients with chest disease is essential in understanding illness and improving its treatment. At the same time, such patients require protection from the risks of research because of their physical and psychological vulnerability and, in the case of critically ill patients, their decisional impairment. Ethics and law in the United States, which evolved in response to research abuses and are more restrictive than those of other countries, seek to advance autonomy and protect research subjects, in large part through the requirement of informed consent.

A clinical investigator in the United States asks the IRB at her medical center how to ethically and legally enroll adult patients in a proposed trial of mechanical ventilation 2 days after the onset of ARDS. She wonders whether (1) consent is necessary in research involving patients with life-threatening illnesses; (2) consent could be deferred until the patients recover from ARDS; (3) patients can give informed consent if they are intubated but only minimally sedated and are not known to have psychiatric or neurologic diagnoses; and (4) surrogates could consent for heavily sedated patients, even if the surrogates are distant relatives.

The IRB explains that (1) consent is necessary for all clinical research in the United States, as mandated by the CFR under which the IRB operates; (2) a waiver of consent, not deferred consent, is allowable only in emergency research and would not be applicable in a study of mechanical ventilation; (3) intubated patients cannot give informed consent because their mental status cannot be adequately assessed and because intubation precludes their asking questions; and (4) even distant relatives can provide consent if they qualify as LARs under applicable state law.

A clinical investigator in the United States asks the IRB at his medical center how to ethically and legally enroll pediatric patients with cystic fibrosis in a proposed study involving bronchoscopy and BAL to determine cytokine levels and thereby extend knowledge about the extent of inflammation in the patients. He wonders whether (1) teenage children can consent for themselves in such a study; (2) a single parent can give consent if the parents are divorced; (3) children must assent for bronchoscopy and BAL even if parents consent; and (4) healthy children can serve as control subjects.

The IRB explains that (1) children can give consent only if they have attained the legal age to do so under state law (in most but not all states, 18 years of age is the legal age of adulthood, although exceptions can be made for emancipated minors); (2) a single parent can give consent (consent generally should be obtained from both parents unless one is deceased, decisionally impaired, or not readily available); (3) the children's assent is necessary, assuming that their age, maturity, and psychological state permit it; and (4) healthy children cannot serve as control subjects without the approval of the Secretary of HHS because the proposed research involves greater than minimal risk and does not present the prospect of direct benefit to the individual subjects.

Further information about the cases is available at www.hhs.gov/ohrp/researchfaq.html.