Chapter 3: Ethical Issues in Clinical Research

We will complete our conceptual framework of clinical research by discussing ethical issues related to the conduct of human studies. We ask people to participate as subjects in studies for the purpose of gaining new knowledge that may or may not have an impact on their lives. Clinical research is conducted because we have questions and do not have the answers; therefore, participation in research may come with some risk to the health and well-being of the subjects. In this society, we recognize a responsibility to research subjects for assuring their rights as individuals. Since the middle of the 20th century, the rules of conducting research have been and continue to be discussed, legislated and codified.

The purpose of this chapter is to present the principles and practices that have become standards in the planning and implementation of research involving human subjects. These principles elucidate the ethical obligation of researchers to engage in meaningful research and to acknowledge the participation of not only the subjects under study but also professional colleagues who contribute substantially to a research project. We will cite the major documents that define public policy.

Researchers have a responsibility for honesty and integrity in all phases of the research process, beginning with choice of a research question. Researchers who are health professionals must set priorities and pursue questions that are relevant to important health care issues. Today, for example, we see an emphasis on research to examine the outcomes of emerging health care strategies. Changes in the population profile, public health problems and scientific or technological advances have stimulated research in specific areas, such as geriatrics, acquired immune deficiency syndrome (AIDS) and stem cell research. External forces such as the priorities of private and governmental funding agencies affect the research direction of most clinical scientists. These complex factors influence the selection of ongoing research efforts.

Researchers also have an ethical responsibility to do clinical research that is meaningful. We should be able to justify a project based on the potential scientific value of its results. This implies an obligation to base our research on rational theoretical principles and to carry it out according to a sound research design with an appropriate sample. It also suggests that research should be conducted by competent investigators who have the expertise to do reliable and valid work. It is not ethical to involve patients in a study, with potential risks to them, when the study has little chance of making a scientific contribution.

During data collection researchers must be careful to minimize the effect of personal bias in measurement. Rosenthal¹ has described several types of experimenter bias that can have significant effects on experimental outcomes. These include unconscious or purposeful inaccuracies in measurement that will tend to support the research hypothesis and influential interactions between researcher and subject that alter the subject's behavior. Bias cannot be eliminated entirely simply because of human nature, but it should be recognized and controlled as much as possible.

Unfortunately, there have been several instances of misconduct in the medical research community such as falsification or misrepresentation of data that serve only to hinder the pursuit of truth and progress.² Statistical procedures should be appropriate, and should not be used to manipulate data for the sole purpose of obtaining a significant result. All data should be included in an analysis, and true differences or the lack of true differences should be reported. Researchers should be aware of potential conflicts of interest that may lead to misconduct. It is not unusual today to find research funded by government or private agencies or equipment manufacturers. The researcher should know who would control the dissemination of information. That agent may want to suppress results that do not conform to expectations. The researcher should be clear on who owns the data and who has access to it once the project is completed.

A researcher has an obligation to publish findings and to be thorough and honest in reporting results. In the 2004 update of the Uniform Requirements for Manuscripts Submitted to Biomedical Journals, the International Committee of Medical Journal Editors has included a section on the obligation to publish negative results.³ The Uniform Requirements are accepted by over 500 journal editorial boards.

Credit of authorship is an issue for research publications. Authors should have made major contributions to a project,³ although many research directors are routinely listed as a formality. Some journals now require a description of the role of each contributor. The order in which authors are listed in the byline is usually determined according to each individual's contribution. Agreements about authorship should be made early in the course of a project to avoid later conflicts. Most research involves the input of many contributors, and researchers should not hesitate to give authorship credit where it is due. Authors should also give credit to those who helped during the project, but who might not merit authorship, through an acknowledgment.

Our primary professional purpose in conducting clinical research is to document the effectiveness and efficiency of treatment intervention. Our patients or normal subjects are, therefore, the sine qua non of our research activities. Commitment to the protection of their rights and dignity must be inherent in the design and conduct of any clinical research project.

Guiding Principles

Research with human subjects requires adherence to three basic principles: autonomy of each individual, beneficence, and justice. Personal autonomy refers to self-determination and the capacity of individuals to make decisions affecting their lives and to act on those decisions. Beauchamp and Childress⁴ describe autonomous action as intentional, carried out with understanding and without controlling influence. It is essential that researchers demonstrate all due respect for the elements of autonomy. Some individuals who may be appropriate subjects for research, such as children or patients with cognitive problems, may be unable to understand adequately. In these cases, the researcher is obliged to assure that an authorized surrogate decision maker is available, has the ability to make a reasoned decision and is committed to the well-being of the compromised individual.

Beneficence refers to the obligation to attend to the well-being of individuals. All who engage in clinical research are bound to "maximize possible benefits and minimize possible harm."⁵ The balance between risks and benefits must be weighed as a part of the decision to go forward with a specific project. Risks may be physical, economic, social or psychological. Potential benefits may include new knowledge that can be applied to future subjects or patients, or that may have a direct impact on study participants, such as improved health status. The point is that the "risks to the subjects [must be] reasonable in relation to anticipated benefits."⁶ The analysis of the risk-benefit relationship measures the probability and magnitude of benefit against the probability of anticipated harm or discomfort. For example, if a new powerful chemotherapeutic agent with known serious side effects requires testing, the selection of normal subjects would be unacceptable; whereas testing on subjects who are terminally ill, and for whom the drug has the potential to effect a positive change, might be acceptable.

Justice refers to fairness in the research process, or the equitable distribution of the benefits and burdens.⁴ This principle speaks to the fair selection of subjects who are appropriate for a given study, drawn from a defined population that is most likely to benefit from the research findings. The selection of subjects should not be discriminatory on some irrelevant criterion, but based on reasons directly related to the problem being studied.⁵ The principle of justice also has meaning in multigroup studies where subjects have an equal chance of being assigned to an experimental or control (or alternative) group. Current practices are followed to ensure consideration of and adherence to these three basic principles.

Use of Control Groups

A historical dilemma for medical researchers has been the need for a control group, or placebo group, as a basis for experimental comparison. Randomized controlled trials (RCTs) have become an accepted way to determine whether an intervention has a significant effect on subjects who receive the treatment compared to those who receive no treatment or a sham treatment. The RCT is preferred for this purpose over observation or retrospective analysis of data. Through the process of random assignment of subjects to the treatment or control group, researchers are best able to control confounding variables that could affect the outcome of the study.

A dilemma may arise when the researcher is also the clinician responsible for the care of the patient who might be a research subject. Rothman and Michels⁷ reported several examples of this problem in drug studies where a placebo condition was employed when, in fact, effective therapeutic methods existed. The current Declaration of Helsinki addresses this issue, stating that: "The benefits, risks, burdens and effectiveness of a new method should be tested against those of the best current prophylactic, diagnostic, and therapeutic methods" (principle 29).⁸ In this case, the research goal is to determine which of the alternative treatments is more effective. This allows us to say that one treatment is different from another, but it does not establish if treatment conditions are more effective than no intervention at all (see Chapter 10).

The researcher can support the concept of a placebo control group from several perspectives. First, there may be some clinical conditions for which no treatments have been effective.⁸ In that case it is necessary to compare a new treatment with a no-treatment condition. It is also necessary to make such a comparison when the purpose of the research is to determine whether a particular treatment approach is not effective. In situations where the efficacy of a treatment is being questioned because current knowledge is inadequate, it may actually be more ethical to take the time to make appropriate controlled comparisons than to continue clinical practice using potentially ineffective techniques. The researcher is obliged to inform the potential participants when a study is planned that includes a control group. Subjects should know that there is a chance that they will receive the treatment being studied and a chance that they will not receive the treatment. According to the 2002 edition of the Declaration of Helsinki, as a form of compensation, the experimental treatment may be offered to the control group patients after data collection is complete, if results indicate that treatment is beneficial (principle 30).⁸ As clinician-researchers, we are obliged to discuss alternative treatments that would be appropriate for each patient-subject if such alternatives are employed in clinical practice. This is an important element of the informed consent process that will be discussed later in this chapter.

Evolution of Regulations for the Conduct of Research Involving Human Subjects

Establishment of formal guidelines delineating rights of research subjects and obligations of professional investigators became a societal necessity as clear abuses of experimentation came to light. In the United States, many unconscionable studies have been identified over the years.⁹ In one case, elderly patients were injected with live malignant cancer cells without their knowledge.¹⁰ Another much publicized study, begun in the 1930s, withheld treatment from men with syphilis to observe the natural course of the disease.¹¹ This study continued until the early 1970s, long after penicillin had been identified as an effective cure.

The first formal guidelines document defining ethical human research practices was the Nuremberg Code of 1949. This document was developed in concert with the Nuremberg trials of Nazi physicians who conducted criminal experimentation on captive victims during World War II. The code clearly emphasized that every individual should voluntarily consent to participate as a research subject. Consent should be given only after the subject has sufficient knowledge of the purposes, procedures, inconveniences, and potential hazards of the experiment. This principle underlies the current practice of obtaining informed consent prior to initiation of clinical research or therapeutic intervention. The Nuremberg Code also addresses the competence of the investigator, stating that research "should be conducted only by scientifically qualified persons."

The World Medical Association originally adopted the Declaration of Helsinki in 1964. This document addressed for the first time the concept of independent review of research protocols by a committee of individuals who are not associated with the proposed project (Principle 13 in the current edition of the Declaration).⁸ These and other essential principles set forth in this document have been incorporated into the U.S. Department of Health and Human Services (DHHS) Rules and Regulations.⁶ The Declaration of Helsinki also declares that reports of research that has not been conducted according to stated principles should not be accepted for publication (Principle 27).⁸ This principle has led to an editorial challenge to professional journals to obtain assurance that submitted reports of human studies do indeed reflect proper attention to ethical conduct. DeBakey¹² suggests that all reports should contain information on the procedures used for obtaining informed consent and ethical review. Most journals now require acknowledgment that informed consent was obtained from subjects used in any research project.

Based on the ethical principles guiding biomedical research and the rules and regulations promulgated pursuant to the 1974 National Research Act (PL-93-348), procedures to ensure protection of human subjects in research have been delineated and are now considered "standard" throughout the United States. There must be a fully developed research proposal that identifies the problem or question to be studied and provides the rationale of, need for, and importance of the study. The research design must be clearly stated and deemed appropriate to answer the question. Informed consent must be obtained from individuals or their legally authorized representatives. This is usually accomplished by signature affixed to a written informed consent document. Both the written proposal and the informed consent form are submitted to an Institutional Review Board for approval.

The National Research Act (1976) established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The deliberations and recommendations of this commission resulted in the Belmont Report (1979) that delineates the guiding principles discussed previously in this chapter and the manner in which these principles apply to human research studies.⁵ Based on that report, both the Department of Health and Human Services (DHHS) and the Food and Drug Administration (FDA) established the rules and regulations that govern the conduct of research in the United States today. The Office for Human Research Protection (OHPR) is the administrative arm of the DHHS that is responsible for implementing the regulations and providing guidance to those who conduct human studies.

The Health Insurance Portability and Accountability Act of 1996 (HIPAA) and the resultant regulations (the Privacy Rule) issued by the DHHS, Office of Civil Rights have added a new dimension to the process of protecting research subjects^13,14 The DHHS regulations (45 CFR 46),⁶ known as the Common Rule, and the FDA regulations (21 CFR, parts 50 and 56)¹⁵ have always included rules for protecting individual confidentiality.

According to federal regulations, an Institutional Review Board (IRB) must review research proposals prior to implementation to ensure that the rights of research subjects are protected.⁶ The IRB must be composed of at least five members. It may not consist of all males, or all females or all members of one professional group, although the IRB must be able to competently review the scientific details of proposed research. At least one member must be concerned primarily with nonscientific issues and may be a lawyer, clergyman or ethicist. One member, a "public" member, must not be otherwise affiliated with the institution where the research is to be conducted. One of the responsibilities of the OHRP is to approve IRB documents that assure compliance with ethical principles and regulations.

Review of Proposals

The responsibility of the IRB is to review research proposals at convened meetings. The decision to approve, require modifications in, defer or deny approval of a proposal must be that of a majority. In arriving at a decision, the IRB considers the scientific merit of the project, the competence of the investigators, the risk to subjects, and the feasibility based on identified resources. If the project is not scientifically sound or practical, there can be no benefit; therefore, no risk to subjects is justified. Reviewers will consider the evidence that the risks and discomforts to the subject have been minimized and are sufficiently outweighed by the potential benefits of the proposed study. This is the risk-benefit ratio. The Board also studies the procedures for selecting subjects, ensuring voluntary informed consent based on complete and understandable descriptions and conforming to the applicable elements of the Privacy Rule. The majority of proposals submitted to an IRB are reviewed in this detailed manner.

Expedited or Exempt Review

Some categories of research activity, however, may qualify for an expedited review or may be exempted from the review process. A project may qualify for expedited review in circumstances such as "recording data from subjects 18 years of age or older using noninvasive procedures routinely employed in clinical practice" and "moderate exercise by healthy volunteers."⁶ In the case of an expedited review, the chairman and at least one designated member of the IRB conduct the review. The advantage of expedited review is that it is usually completed in less time than that required for a full Board review.

Projects may be exempted from committee review if they are surveys, interviews, or studies of existing records, provided that the data are collected in such a way that subjects cannot be identified. Surveys or interviews will only be given exempt status if they do not deal with sensitive issues, such as drug abuse, sexual behavior or criminal activity. Secondary analysis of existing data from databases ("limited data sets") where subject information is appropriately de-identified may be exempt. On the other hand, when actual patient clinic records are used, IRB approval is required.¹⁶ Researchers must describe methods of coding data to assure patient confidentiality. Given the advent of the HIPAA Privacy Rule, some institutions have eliminated the exempt review category.

All proposals for studies involving human subjects must be submitted to the IRB, which then determines whether a project qualifies for full review, expedited review or is exempt.

Institutional Guidelines

Each institution establishes its own guidelines for review in accordance with federal and state regulations. Clinicians should, therefore, become familiar with the requirements in their own institutions. Clinicians should also be aware that the process could take several weeks, depending on the IRB's schedule and whether or not the proposal needs revision. This review process should be included in the timetable for any research project. No research on human subjects should be done without prior review and approval of a designated review committee.

Perhaps the most important ethical tenet in human studies is the individual's ability to agree to participate with full understanding of what will happen to him. The informed consent process and all of its elements address the basic principles of autonomy, beneficence and justice. The components of the process consist of information elements, including disclosure of information and the subject's comprehension of that information, consent elements, ensuring the voluntary nature of participation and the subject's competence to consent and authorization to use data in a manner specified in the protocol.^15,17

Information Elements

Subjects Must Be Fully Informed

The informed consent process begins with an invitation to participate. A statement of the purpose of the study permits potential subjects to decide whether they believe in or agree with the worth and importance of the research. The process then requires that the researcher provide, in writing, a fair explanation of the procedures to be used and how they will be applied. This explanation must be complete, with no deception by virtue of commission or omission. Subjects should know what will be done to them, how long it will take, what they will feel, what side effects can be expected, and what types of questions they may be asked. If subjects cannot read the informed consent document, it should be read to them. Children should be informed to whatever extent is reasonable for their age. Subjects should also know why they have been selected to participate in terms of inclusion criteria for the study, such as clinical condition or age. If the subjects are patients, they should understand the distinction between procedures that are experimental and procedures, if any, that are proposed to serve their personal needs.

An ethical dilemma occurs when complete disclosure of procedures might hinder the outcomes of a study by biasing the subjects so that they do not respond in a typical way. When the risks are not great, review boards may allow researchers to pursue a deceptive course, but subjects must be told that information is being withheld and that they will be informed of all procedures after completion of data collection. For example, when the research design includes a control or placebo group, subjects will not know what treatment they are receiving. They should know that they will be told their group assignment at the completion of the study.

An important aspect of informed consent is the description of all reasonable foreseeable risks or discomforts to which the patient will be subjected, directly or indirectly, as part of the study. Risk refers to physical, psychological or social harm that goes beyond expected experiences in daily life. The researcher should detail the steps that will be taken to protect against these risks and the treatments that are available for potential side effects. For example, if a patient is likely to become fatigued as a result of performing maximal physical exercise, the researcher may include rest periods during the experimental trial. If a patient were receiving electrical stimulation, the researcher would explain the risk of shock and how that risk is minimized by proper grounding and by regular inspection of the equipment. The subject should be advised against such behaviors as taking certain medications or driving a car, which could be hazardous during or after the experimental period. A statement should be included whereby subjects agree to exercise appropriate caution. They are not bound by this, but they should understand the potential harm of not honoring the agreement. If the research involves more than a minimal risk, a statement should be included concerning the availability of medical care and whether compensation will be provided. Subjects should also be informed of new information, such as the identification of previously unknown risks that becomes available during the course of the study. This may affect their willingness to continue participation.

The researcher also delineates the potential benefits of participation. Some studies may result in a beneficial reduction of symptoms. For example, subjects who participate in a study to test the effectiveness of treatment for migraine headache may find their pain relieved by the experimental treatment. The subject should be advised that such a benefit is possible but is not guaranteed. Studies that are geared more toward theory testing may provide no direct benefits. The researcher should explain the potential application of theoretical findings and how the findings will contribute to future research or future patient care.

When a study involves a form of therapeutic intervention, subjects must be informed that alternative treatments are available and that they have the right to choose among them instead of accepting the experimental intervention. Patients must also be told if "standard" treatments to which they are entitled are being withheld as part of the study. This information is essential if a patient is to make an informed decision about accepting the experimental conditions.

Subject Information Should Be Confidential and Anonymous

Research subjects should be told what steps are being taken to ensure confidentiality of all information, including descriptive and experimental data. Whenever possible, a subject's anonymity should be protected. This becomes an issue with surveys, for example, when respondents wish to remain unidentified. In experimental situations anonymity is often not feasible, but the researcher can code the data without using names. Identifying codes can be kept separate from the rest of the data. Researchers should also be aware of this responsibility when disseminating results. Researchers also have a responsibility to know the requirements of the Privacy Rule^15,17 and the procedures established by their Institutional Review Boards.

If the subject is to be videotaped or photographed during the study, this should be disclosed in the consent form. The subject should know who will have access to tapes or photographs, who will keep them, and how they will be used. Subjects retain the right to review such material and to withdraw permission for its use at any time. Subjects should also be informed if one-way windows will be used and who the observers will be.

The Informed Consent Form Must Be Written in Lay Language

Informed consent is more than telling subjects about the research; the process implies that they understand what they are being told and what they are reading. The language must be clear and basic so that the average reasonable individual can follow it. Professional jargon is unacceptable. Instead of "perform a maximal isometric contraction," the subject should be told to "pull up as hard as you can without moving." This is the language that clinicians use routinely in patient education. As a rule of thumb, language should be written for the lowest educational level that would be expected for subjects.

The Researcher Must Offer to Answer Questions at Any Time

The researcher is responsible for ensuring that the subject understands all relevant information. A verbal description is almost always a part of the process, so that the researcher can "personalize" the information for each subject. The subjects should have sufficient time to assimilate the details of the proposed project, prior to making their decision to participate. They should feel free to question the procedures at any time during the course of the study, and should be provided with the name and telephone number of an appropriate contact person.

Consent Elements

Consent Must Be Voluntary

Subjects should participate in a research project of their own free will. Patients are usually quite motivated to help, but they must be informed that there is no penalty to them if they refuse. Some studies may involve monetary compensation for participation. It should be clear if such compensation will be received whether or not the subject completes the study.

Special Consideration Must Be Given to Subjects Who Are Particularly "Vulnerable"

Some individuals cannot give informed consent because they may not be able to understand the information. In cases of mental illness, developmental disability, or diminished mental capacity, the ability of the subject to consent must be evaluated by the researcher and others who know the subject well. If the subject is not competent, consent must be provided by a legal guardian or advocate.⁶

The regulations regarding children as research subjects require that parents or guardians give permission for participation. Furthermore, if a child is considered competent to understand, regardless of age, his or her assent, that is, his or her affirmative agreement to participate must be obtained and documented. Researchers should be particularly cautious about influencing subjects who are considered "captive."⁷ For example, there are specific regulations regarding research involving prisoners.⁹ More subtle circumstances exist with the use of students or nursing home residents. In both cases, the sense of pleasing those in authority may affect the subjects' decisions.

Subjects Must Be Free to Withdraw Consent at Any Time

The informed consent document must indicate that the subject is free to discontinue participation for any reason at any time without prejudice; that is, the subject should be assured that no steps will be taken against him, and, if he is a patient, that the quality of his care will not diminish. This can occur before or during an experiment, or even after data collection when a subject might request that his data be discarded. It should also be clear that the researcher would discontinue the experiment at any time if necessary for the subject's safety or comfort.

The Informed Consent Form

All subjects must give informed consent prior to participating in a project. This is done by providing a written informed consent form that is signed and dated by the subject, researcher and a witness. Subjects should receive a copy of this form and the researcher must retain a signed copy. Although it is a general contractual agreement, the informed consent form is not binding on subjects. Subjects never waive their rights to redress if their participation should cause them harm. The form should not contain language that appears to release the researcher from liability. The required elements of an informed consent form are listed in Table 3.1. A sample informed consent form can be found in Appendix E. The format of this sample informed consent form is used in many institutions. It specifically identifies and acknowledges all of the elements of informed consent. Of special note, the signature page should not stand alone, but must contain some of the text of the document to show that the signatures are applied in the context of the larger document.

Although written consent is preferable, some agencies allow oral consent in selected circumstances. In this case, a written "short form" can be used that describes the information presented orally to the subject or his or her legally authorized representative.⁶ This short form is submitted to the review committee for approval.

Informed Consent and Usual Care

Clinical research projects are often designed to test specific treatment protocols that are accepted as standard care, and subjects are recruited from those who would receive such treatments. Therapists often ask if informed consent is necessary for a research project when the procedures would have been used anyway. The answer is yes! Even where treatment is viewed as usual care, patients are entitled to understand alternatives that are available to them. Patients must always be informed of the use of the data that are collected during their treatments, and they should have sufficient information to decide to participate or not, regardless of whether treatment is viewed as experimental or accepted clinical practice.