Protecting the Boundaries
Health and human service professionals most frequently set boundaries through sampling plans that involve human subjects. Involving human subjects in research requires important ethical considerations and legally binding actions. The basic ethical principle that underlies boundary setting of human subjects in each of the research traditions is that the researcher is obligated to ensure the protection of human subjects.
What do we mean by “the protection of human subjects”? All research in which people are directly involved has potential risks to its participants, even if such risks are minimal or simply involve momentary discomfort with a personal question on a survey. According to federal law,1 investigators must submit a plan (proposal) for the ethical conduct of any inquiry involving human subjects to a board or group composed of both lay and scientific representatives. This board is mandated to examine proposals with regard to critical considerations that include: (1) the level of risk posed to study participants and relationship of risk to potential benefits to society; (2) the adequacy of the plan to provide participants with the necessary knowledge about study procedures, risks, and benefits, referred to as “full disclosure”; (3) the plan for ensuring that study participation and all procedures are voluntary; and (4) the plan for ensuring confidentiality.
Large institutions such as hospitals and universities have formal committees, usually called institutional review boards (IRBs). In smaller agencies, review boards may be ad hoc committees (with a particular purpose). Regardless of where you are conducting an inquiry, however, you must seek a human subjects review to protect those who are devoting time and effort as participants in your study. Also, if you obtain funding for your research, you will not be allowed to conduct the study until such a review has been conducted and formal approval obtained. Even if you are conducting a small or pilot-level research study, such as for a research class, you must seek IRB approval. IRBs are the main mechanisms through which protection strategies are reviewed and monitored in the conduct of any type of research study.
Principles for protecting human subjects
Human subject protection is based on three primary principles: full disclosure, confidentiality, and voluntary participation (Box 12-1). All investigators, regardless of the scope or type of research, must follow these principles.
Full Disclosure
Any person who participates in a study, whether participatory action research, single-subject design, or randomized trial, has the absolute right to full disclosure of the purpose and procedures of the study. Full disclosure means that the investigator must clearly share with the informant, subject, or research participant the types and content of interviews, length of time of participation, types and length of observations, and other data collection procedures that will occur, as well as the scope and nature of the person's involvement. Full disclosure also means that any risk to a subject, even if the potential is rare or minimal, must be clearly identified and a plan for remediation offered for each risk to every subject.
Identifying and sharing specific study procedures tend to be straightforward in experimental-type research. However, such disclosure can create difficulties for certain forms of naturalistic inquiry. In an experimental-type study, all the procedures are clearly articulated and determined before entering the field. Therefore, the researcher can identify in layperson's terms the purpose and scope of the study and the types of data collection efforts that will occur. In naturalistic designs, most bounding-type decisions are made in the field and change or evolve over time as knowledge about the context emerges. Researchers who work in this tradition must solve this dilemma in creative, thoughtful, and ongoing ways. In some studies, it may be necessary to introduce a consenting process for each data collection effort.
Discussion in the research literature is ongoing about effective approaches that investigators can use to remain ethical while preserving the integrity of their methodology. All researchers, regardless of tradition, must struggle over the best way to describe the study truthfully without revealing specific aims or hypotheses or introducing biased perspectives that may shape the informant's responses during the study. For example, although it is necessary to state the overall research objective, it is not appropriate to indicate a directional hypothesis that could influence how an informant responds during an interview.
Full disclosure of study intent and procedures is usually provided when initially enrolling participants into a study and when obtaining informed consent. In addition to consent forms, some investigators also provide an informational sheet to participants as a handy reference that outlines the study purpose and the procedures used. In studies that involve multiple testing occasions, the investigator may restate the study purpose and procedures at each follow-up to ensure that study participants understand what to expect next and at which stage they are in the research study.
Although full disclosure is mandatory in all studies and with all human subjects, it is important to recognize the importance of disclosure, particularly as it concerns vulnerable populations (e.g., children, individuals with cognitive impairments) and minority groups for which there has been a long history of unethical research practices. Take, for example, the Tuskegee syphilis experiment conducted between 1932 and 1972 in Tuskegee, Alabama, by the U.S. Public Health Service. The study recruited 399 poor African American sharecroppers with syphilis to evaluate the natural course of the disease. Even though by 1947, penicillin had become the standard of treatment for syphilis, this information and access to drug treatment was not offered to study participants. The failure to treat study participants resulted in unnecessary suffering and death of participants and their family members. The unethical practices in this and other studies led to major regulatory changes to protect participants of studies, including full disclosure through the informed consent process and accurate and timely reporting of study results including informing participants of changes in risk or treatment discoveries.
Confidentiality
The investigator must ensure that all information shared by a respondent in the course of a study is kept confidential. Confidentiality means that (1) no person other than specified members of the research team can have access to the respondent's information, unless those who have access to the data are identified to the participants before their participation (usually stated in informed consent), and (2) the information provided by a respondent cannot be linked to the person's identity. This second consideration, although relevant to all research involving human subjects, is especially important in studies that focus on sensitive topics, such as AIDS research, teenage pregnancy and birth control use, crime, and drug dealing or abuse.
An investigator can ensure confidentiality in several ways. The name of the respondent must be removed from the actual information that is obtained. This procedure ensures that the identity of respondents in your study is protected and that the information they provide will not be linked to their names in the future. One typical way to protect a study participant's identity is the assignment of identification numbers. However, this action presents some difficulty for studies that primarily use observation as the principal data collection effort. Also, ensuring confidentiality can be difficult when using audio and video recordings as data collection sources. In these instances, the investigator does not usually transcribe names that are recorded. The researcher would more likely establish procedures for coding and storing digital data in locked filing cabinets or offices with restricted access, as well as destroying electronic copies at the conclusion of the study.
Confidentiality of research participants must also be protected when results are reported. In reporting findings from a case study or naturalistic design, the names of individuals and key identifying information are modified so that there is no direct link between the person's identity and the information the person provides. Most experimental-type studies report findings that reflect summative scores or outcomes of an aggregate of individuals, which makes ensuring confidentiality less challenging in this research tradition than in naturalistic inquiry.
Researchers who investigate controversial topics must carefully plan how information will be stored and reported. Studies about HIV, sexual activity, drug trafficking, or sexual, physical, or substance abuse may obtain information of interest to the legal system. Investigators can refuse to turn over documentation but risk being called into court to testify.
Federal regulations such as the Health Insurance Portability and Accountability Act (HIPAA) also impose confidentiality rules and restrictions on research activity. HIPAA requires that all health-related information obtained in the course of a study be “de-identified” such that it is not possible to link a person's name to the health information provided. HIPAA also requires that any health information that is shared in the course of a study be documented in the informed consent form that is signed by a participant before entering a study. Furthermore, under HIPAA, researchers are not legally permitted to contact individuals about a study unless that person has given prior permission for such contact to be made. To abide by this regulation, many clinical sites ask patients or clients to sign a form that indicates their willingness to be contacted and informed about a study of potential interest. Researchers are developing similar procedures.
Ensuring confidentiality necessitates that certain office procedures be established. First, written personal identifying information concerning study participants is kept to a minimum. A master list of study participants that includes the assigned identification number (ID#), first and last names, and necessary group designations (e.g., control vs. experimental group) is maintained for tracking purposes on computers and is password protected; hard copies are kept in locked filing cabinets separate from the actual information obtained. In many studies, screening forms and interview cover sheets that include identifying information are used. All this information must be considered strictly confidential and is also kept in locked file cabinets separate from study data. Any information with subject identification that is not needed is shredded.
Second, confidentiality should also be maintained when making telephone contact with study participants or potentially eligible persons. Conversations with or about study participants require commonsense discretion. Attempts to schedule appointments with study participants or telephone conversations discussing issues related to a participant should be conducted in a manner that ensures confidentiality. When talking on the telephone, researchers should keep voices low and should not use last names except as necessary. When possible, they should conduct phone conversations behind closed office doors. If this is not possible, calls can be made when fewer people are in the office or only persons directly involved with the research project are present.
Third, computer files used for tracking study participants can be set up with password protection. Access to such files should be restricted to defined key personnel, and any backup disks must be stored in locked file cabinets.
Fourth, interviewers must be trained and certified in “protection of human subject” procedures before any telephone or face-to-face interviews with study participants. For face-to-face contact, interviewers introduce themselves, explain the study's purpose and procedures, and review the informed consent. Informed consent must be obtained before collecting any study-related information. For telephone interviews, interviewers must obtain verbal assent, which is recorded on an institutionally approved form.
Voluntary Participation
When humans are involved in studies, their participation must be strictly voluntary. Individuals have the right to choose to participate or not. Also, an individual who initially agrees to participate in a study has the right to withdraw from the study at any point and the right to refuse to answer a particular question or participate in a particular set of procedures. Thus, the voluntary quality of participation must be protected at three points in a study: initial enrollment, continuation in the study, and right to refuse answering specific questions or a particular study procedure. To ensure voluntary participation at each of these points, investigators must develop approaches to recruiting participants that are not coercive and that provide full disclosure of all study procedures.
Although a person may refuse to participate in a study or one of its procedures, it is important for the investigator to understand why refusals occur. Is withdrawal caused by the nature of the procedures or by excessive demands placed on participants? Is the research team offensive in any way? Is withdrawal based on a change in the health status of the participants or their relocation to another geographic region? Reasons for refusal to participate in a study, withdrawal from a study as a participant, or refusal to answer a particular question may have implications for the ethical conduct of the study, interpretation of results, and ability to generalize outcomes to other groups, as well as planning future studies. As such, it is important to keep track of and evaluate the reasons study participants withdraw from a study or refuse participation in a study component. This information enables the investigator to refine ethical plans, evaluate if differences exist between those who participate and those who do not, and prepare for future research.
Refusal to answer a particular question (e.g., “what is your yearly income?”) or to engage in a particular study component (e.g., allow observation of home interactions) presents a methodological challenge that the investigator must be prepared to meet. Missing information can be a greater problem in experimental-type research than in naturalistic inquiry because missing data limit the types of statistical analysis that can be used and the inferences that can be derived from the data. For example, a common question on survey studies is level of income as one indicator of socioeconomic status. However, participants may refuse to disclose this information. Missing information can be handled in numerous ways, including using the mean value of the group or using a statistical program to assign a value randomly.
Missing information is less problematic in naturalistic studies. The refusal to answer a question may be an indicator of the salience (or importance) of that particular topic or area, and its “missingness” becomes, in essence, a type of meaning that enters into an interpretive scheme. For the naturalistic inquirer, however, missing an observation, such as an important community event, may be more problematic. Adjustments in collecting information may need to occur to overcome this issue, such as prolonging engagement in a particular context or obtaining information about the event from newspaper reports, community meetings, and personal interviews.
Belmont report
The ethical issues of conducting research have only recently been a focus of national concern. In 1974, the National Research Act created a commission to delineate the ethical issues and guidelines for the involvement of humans in behavioral and biomedical research in the United States.1 This act and its subsequent activities arose from revelations of the Nuremberg war crime trials about the devastating human experiments conducted by medical scientists during the Holocaust. Other tragic abuses of human subjects involved in research had occurred in the United States as well. Most notably, the Tuskegee experiments, discussed earlier, that involved poor rural black men diagnosed with syphilis from whom investigators withheld known curative treatment to observe the natural course of the disease process.
The resulting Belmont Commission issued a report in 1979 that outlined three basic ethical principles to guide all research activity in order to protect human subjects.2 The Belmont Report is a brief document that is required reading by all those involved in human subject research. The first ethical principle is the importance of distinguishing the boundaries between research and practice. This differentiation may be more complex in health and human service research than in other types of science research. However, it is important to distinguish between daily or traditional practice versus systematic efforts to evaluate new approaches and service interventions.
The second principle in the Belmont Report describes three areas that must be addressed: respect for persons, beneficence, and justice. The first area, respect, states that individuals should be treated as autonomous individuals who are capable of personal choice and self-determination. A related mandate is that individuals who are not autonomous or who are vulnerable, such as the person with reduced cognitive capacity, must be protected. The second ethical area, beneficence, specifies that research will “do no harm” and will “maximize” benefits and “minimize possible harm” to individuals. The third ethical area, justice, specifies that people should be treated equitably; that is, research that poses a risk should not be conducted on vulnerable populations.
The third principle described in the Belmont report concerns the application of these general principles to research activities including the informed consent process, specifying a risk–benefit assessment and how human subjects are selected for participation in a study. Let's examine the elements of this third principle.
Institutional review board
On the basis of the Belmont Report, the involvement of humans in research must be overseen by a government-mandated board of experts that must be established at each institution engaged in the research process. These boards, referred to as institutional review boards (IRBs), are charged with monitoring the ethical conduct of research as outlined by the Belmont Report. Most academic settings have an IRB, but only a few health and human service settings have established research committees or IRBs. If you are located in a setting that does not have an established board, you may need to form a partnership with a university or hospital that can review your protocol. Some universities have arrangements with community-based agencies and organizations in which they agree to review protocols for a fee. This linkage is particularly important when seeking funding to support your research effort. All federally funded research studies must be approved by an official IRB. In most institutions, however, any research study—funded or not funded, small scale or large scale—must be reviewed by a committee to examine the nature of human involvement.
Before implementing a study, a researcher must write a proposal describing in detail the plans for involving humans, the procedures of the study, and analytic strategies. This proposal must be submitted for review to a designated office of research or to an IRB (see Chapter 21).
One of the initial determinations that must be made is whether a particular activity can or cannot be classified as research. This is not as straightforward as it may sound, particularly in clinical settings in which some clinical activities can become a form of research. This is also the case in educational settings, in which some evaluative activities may become a form of research.
Assume you are a health professional working in a rehabilitation setting, and (1) you read about a new therapeutic technique in the literature and want to evaluate its benefits for your clients, or (2) you want to examine case records to see whether you can identify a set of factors that predict rehabilitation improvement. Are these research studies? Would you need to submit these plans to the IRB for review and their approval?
Let's assume your clinical department decides to implement this new technique and track patient outcomes. However, the purpose of this activity is clinical and is designed to improve clinical services. Thus, it would not be considered research and hence require IRB approval. Suppose, however, that you want to compare outcomes from this new therapeutic approach systematically to traditional care and assign patients to receive the new or typical treatment, aggregate the data, and report the results formally. In this case, the activity would be considered research and fall under the purview of the IRB.
In the second example involving a chart extraction activity, given that you plan to review case records systematically and aggregate the data for reporting purposes, you need IRB approval. Because your methodology involves extracting information from the charts of patients but does not require patient contact or disclosure of patient names, you most likely will receive what is referred to as “exempt” status (see later discussion); this means your study has minimal risks and thus is exempt from continued reporting requirements. Although you are initially obligated to inform the IRB of any research study you plan to conduct, if it is designated as exempt by that board, there will be no further legal or ethical requirements for annual IRB reports and updates.
After the IRB determines that the proposal is research and requires a review, its main goal is to evaluate whether the research protocol will adversely affect study participants, whether study procedures are too burdensome, whether benefits outweigh risks, whether the design is appropriate to address the research questions, and whether the research itself justifies the involvement of humans. Also, the IRB evaluates the procedures that will be used to identify and enroll or engage humans in the study to make certain that participation is voluntary, that confidentiality will be ensured, and that there is full disclosure of study procedures.
Assume you are a rehabilitation clinician and plan to evaluate the relationship between client self-report of functional ability and observation of actual performance. You plan to evaluate the clients you see in the rehabilitation setting. Although this idea does not necessarily present ethical challenges, you must carefully consider how you will introduce the study to your clients. Also, you must establish clear boundaries between your clinical efforts and the information you will gather for research purposes. You must set up procedures to ensure that clients understand three points. First, their participation in the study is strictly voluntary; in other words, it is their decision to participate. You must establish procedures to enroll clients that ensure that they do not feel coerced into participating. Second, a decision not to participate will not affect the type and quality of service intervention they will receive. Third, if clients decide to participate, they can choose to discontinue participation at any point in the study with no consequences to them or their ability to receive other medical and social services for which they are eligible.
Moreover, the IRB will review all study procedures to make certain that they are ethical and not coercive. You will not be able to begin your study until you have received approval from the IRB.
Three Levels of Review
Most IRBs have three levels of review: full, expedited, and exempt. A full board review involves a formal examination of an investigator's research protocol by members who have been officially appointed by an institution. The composition of board members is mandated to include a consumer representative and a member with scientific expertise in the protocols being reviewed. The board can be composed of as many as 20 individuals or as few as 5, depending on how the institution has set up its IRB. A study that involves a vulnerable population must receive a full review. Vulnerable populations refer to individuals who may not be able to represent themselves or participate in decision making or who may be at particular risk when participating in a research study (see later discussion). Examples of vulnerable populations are infants, children, pregnant women, prisoners, mentally incompetent individuals, or persons addicted to substances. Additionally, research studies involving human immunodeficiency virus (HIV) testing, acquired immunodeficiency syndrome (AIDS), investigational drugs, or medical devices also require a full board review. In a full board review, all members read the research protocol, discuss its merits and weaknesses, and vote to either approve (with changes or no changes to recruitment and consent procedures) or disapprove it.
An expedited review involves an evaluation of a research protocol by a subcommittee selected from the full IRB membership. Studies that receive an expedited review do not involve vulnerable populations and do not test invasive techniques, and they represent minimal risk to individual participants. Studies that can have an expedited review include those that collect data from individuals who are 18 years of age or older using noninvasive procedures routinely used in practice, studies that use existing data, and research on group behavior in which the investigator is not manipulating behavior. The previous example of the study in rehabilitation is appropriate for an expedited review.
Exempt status means that a study protocol is exempt from formal review from either the full board or its subcommittee. Although it is necessary to inform the IRB of the intent to conduct the research, the IRB will send a letter of approval of its exempt status and indicate that no future or annual review is necessary. Studies that are exempt from formal review procedures may include evaluation research such as that involving normal educational practices or the use of educational tests, as well as research involving the collection or study of existing data, documents, and records, provided these sources are publicly available or the information is recorded so that the individuals cannot be identified. An example of the latter situation is a retrospective study involving chart review or hospital census data from the previous 10 years of individuals who experienced strokes and their level of functional status at discharge.
In submitting a proposal to an IRB, you must address each of six considerations (Box 12-2). Although the actual format of a proposal submission to an IRB may differ across institutions, these six points are standard.
Informed consent process
The principles outlined in the Belmont Report are applied to the conduct of research through the informed consent process. Informed consent is the process through which potential study participants are informed of the study and its participation requirements. It usually takes the form of an official written document developed by the researcher that informs study participants of the purpose and scope of the study. Although the specific wording and format of informed consent forms vary widely across institutions, they must contain the basic elements listed in Box 12-3.
BOX 12-3 Typical Elements of Informed Consent
Statement of purpose of study in layperson's terms
Description of study procedures (e.g., number and length of interviews)
Disclosure of any risks or discomforts from study participation
Statement describing how confidentiality will be ensured
Health Insurance Portability and Accountability Act disclosures
Statement describing right of refusal and voluntary consent
Description of benefits of participation
Signatures of study participant, interviewer, and researcher
Important elements of informed consent include a description of the procedures in which you are asking the person to participate and your assurance that participation is voluntary. Also, you need to specify whether participation in the study carries any known risks, and if so, what these risks are and what measures should be taken if they occur. In proposing a study, you need to consider the elements to include in your informed consent and the procedures you will use to introduce it to study participants. A consent form must be read before collecting any information from a person. Usually the participant, the interviewer, and the principal investigator sign the form, and a copy is made for the participant to keep.
A written consent is not required for every type of study. For example, in conducting a mail survey, the act of returning the survey to the investigator is considered an indication of the respondent's consent to participate. Likewise, in a telephone survey, the act of agreeing to answer questions over the telephone is a sign of volunteering or assenting to the study.
Obtaining consent from participants, whether written or assent through verbal acknowledgment, is somewhat straightforward in experimental-type research that involves individuals who are not cognitively impaired. The researcher knows exactly who is eligible to participate in the study and can review informed consent before asking a set of standardized questions. It can be more difficult in naturalistic inquiry, especially when data collection involves observing different events in which you cannot predict who will be attending or involved in the setting. Researchers using a naturalistic approach must be creative and thoughtful as to the best and most ethical way of handling consent.
The research of Bluebond-Langner,3 a medical anthropologist, illustrates the different approaches ethnographers must consider in addressing ethical dilemmas.
Several procedures were used to obtain consent in a study of children dying with leukemia and their parents. The study posed several challenges. First, it involved a vulnerable, protected population—children. Second, it involved data collection over time at different sites, in the physician's office, clinical settings, and hospitals. Third, it involved children who were dying and parents who were grieving. Bluebond-Langner first obtained consent from the parents by approaching them during their visits to the clinic. When she observed the same families admitting their children to the hospital, she reintroduced the study. Any parent who appeared uncomfortable or told someone else that he or she would rather not participate was dropped from the study. As the author describes, “I would never take any statement of consent as final, since I felt that the parents needed a way of getting out of the study any time they desired.”3 She also sought assent from the child by asking each for permission to speak with him or her about thoughts and feelings each time she approached him or her. These procedures not only ensured protection of participants but also led the children and their parents to trust the researcher. She believed that mutual trust and understanding were established very early in the study and were constantly reinforced.
As stated by the Office of Human Subjects Research (OHSR) of the National Institutes of Health, informed consent is best conceptualized as a process:
Informed consent is a process, not just a form. Information must be presented to enable persons to voluntarily decide whether or not to participate as a research subject. It is a fundamental mechanism to ensure respect for persons through provision of thoughtful consent for a voluntary act. The procedures used in obtaining informed consent should be designed to educate the subject population in terms that they can understand. Therefore, informed consent language and its documentation (especially explanation of the study's purpose, duration, experimental procedures, alternatives, risks, and benefits) must be written in “lay language” (i.e., understandable to the people being asked to participate). The written presentation of information is used to document the basis for consent and for the subjects' future reference. The consent document should be revised when deficiencies are noted or when additional information will improve the consent process.
Use of the first person (e.g., “I understand that …”) can be interpreted as suggestive, may be relied upon as a substitute for sufficient factual information, and can constitute coercive influence over a subject. Use of scientific jargon and legalese is not appropriate. Think of the document primarily as a teaching tool, not as a legal instrument.4
To understand fully the significance of informed consent and its pivotal role in the bounding processes of research, access the OHSR Web site. Also, samples of consent forms from our own research are provided in Appendix A. Although these consents contain the basic elements, each institution requires slightly different wording and approaches.
Study Approval and Monitoring
After you have submitted a proposal and a sample consent form and obtained IRB review, you will receive a letter indicating that your research protocol has been approved as submitted, conditionally approved, or not approved. Conditional approval indicates that the IRB will provide final approval after specific issues are clarified and elements of the consent are modified. However, an investigator cannot begin a study until final approval is obtained. If your study is not approved, you must carefully consider why and what types of changes are necessary.
In addition to receiving an approval letter, you will also be provided with your original informed consent document with a stamped date at the bottom of the page indicating when IRB approval was granted and the date that approval will expire. You must use only current, dated consent forms. Each year, a report must be submitted to the IRB indicating the number of study participants who have been enrolled, adverse events, and general study progress. Each year, the IRB will restamp a consent form with the current date. The dated consent document ensures that only the current, IRB-approved informed consent documents are presented to study participants. It also serves as a reminder to investigators of the need for continuing review.
Each subsequent year, you will be required to submit a brief annual report updating the IRB as to the number of subjects enrolled and any changes in protocol and level of risk. Additionally, adverse events related to study treatments must be reported to the IRB within 24 hours, then summarized at the annual reporting period.
Developing an informed consent document
Guidelines for developing the informed consent document reflect the ethical framework established by the Belmont Report. First, the consent should be written at a sixth-grade to eighth-grade level to ensure that persons with different levels of literacy can fully understand the research process in which they are being asked to participate. Second, familiar words should be used throughout the consent. Scientific jargon, including medical and legal terms, should be avoided; for example, using the term “cholesterol” is preferable to “blood lipids,” and when indicating “random assignment,” explain that assignment will be determined “by chance.” If a scientific or medical term must be used, be sure to define it. Avoid the use of abbreviations and acronyms.
It is critical that all persons be given the opportunity to participate in the research process. You need to consider different strategies for involving persons with low literacy so that they can understand your study procedures and participate in an informed way. For persons with low cognitive capacity as well, approaches can be used to ensure their understanding of the research process. Different approaches may be necessary to be inclusive for different study populations. For example, you might develop an explanation of your study, with its risks, benefits, and confidentiality procedures, that can be translated into pictures, languages other than English, or Braille, if necessary, depending on the target group. The point is that you must consider the characteristics of your target study population and how best to ensure informed participation.
Involving vulnerable populations
The involvement of vulnerable persons remains a widely discussed and debated issue in research. As noted earlier, vulnerable populations refer to a wide range of persons who may be at risk in a research setting because of not only their intrinsic characteristics but also their life situation or circumstance. For example, persons may be vulnerable because of a medical condition (e.g., terminal illness), a particular setting (e.g., emergency room of hospital, homeless shelter, prison), a baseline limitation of intellectual function (e.g., developmental disability), a psychosocial stressor (e.g., posttraumatic stress disorder), or an illness that compromises comprehension and decision-making abilities (e.g., dementia).5
The concern with involving vulnerable persons is twofold. First, no ethical justification exists for excluding vulnerable populations. In fact, it is unethical to exclude such populations from research in that their exclusion restricts knowledge development in areas that would ultimately benefit that group. Vulnerable populations are typically underrepresented in studies, and it is usually not possible to generalize research findings to the group. A second concern is determining the best way to involve vulnerable populations. Here the investigator must strike a balance between not excluding a particular study group that is vulnerable and avoiding inducement.
Consider how pharmacological studies that use monetary inducements for participation may be particularly attractive to individuals rendered vulnerable because of their life circumstances, such as persons who are homeless or live in prison.6 Participating in a study based on a hefty monetary inducement, however, may not be in the best interest of these groups.
A related issue is how best to apprise persons with compromised decision-making abilities about the procedures of a study so that they can make an informed decision to participate or not. Remember that an essential principle established in the Belmont Report and overseen by IRBs is that persons must be fully informed before agreeing to participate in a study. For persons with dementia and other populations with impaired comprehension and judgment, how can we be assured that these groups willingly and knowingly agree to participate in a study?
The research community is just beginning to address this important issue and develop tools that can reliably and validly distinguish between persons who can make a judgment regarding their participation and those for whom proxy consent is necessary.7 For example, persons at the early stages of dementia may be capable of comprehending a typical informed consent written document, whereas persons at the moderate stage may not fully understand what they are committing to in the study. In persons unable to provide their own consent, consent from a legal guardian or family caregiver can be obtained. Even when proxy consent is obtained, however, before administering a test, interview, or any type of procedure, the person with dementia should be provided a brief explanation, and assent should be obtained (e.g., “Hello Mr. Smith, I would like to ask you a few questions about how you are doing; is that okay?”).
Although simplifying the language of an informed consent may be helpful for some groups with compromised intellect or decision making, use of pictorial or other visual representations of study procedures may be useful for other populations. Knowing exactly the most effective way to present informed consent can be difficult and is only now becoming the focus of research.
Suppose you want to conduct a study of the quality of life of persons with developmental disabilities living in a group home. You want to include persons living in the home who have a wide range of capabilities. In this case, your presentation of informed consent may need to be tailored to match persons at different levels of intellectual and cognitive abilities. However, there is no standardized approach to do this, or even to determine how to identify what type of approach a person may need to become fully informed.
Despite the difficulties and challenges of involving vulnerable populations, the scientific community must be committed to develop meaningful approaches that can appropriately involve them. Remember that these populations are vulnerable only because of our lack of knowledge and expertise about their involvement in research and their prior exploitation in the research enterprise. Given the challenges in developing informed consent documents for different populations and, in particular, for individuals with low literacy or cognitive impairments, the Agency for Healthcare Research and Quality has developed a helpful toolkit. Available on the Web: “Informed Consent and Authorization Toolkit for Minimal Risk Research” (http://www.ahrq.gov/fund/informedconsent/). It provides templates for informed consents and strategies to evaluate the literacy level and appropriateness of the documents you develop.
Specialized oversight of experimental-type designs
The true-experimental design, often referred to as a “randomized controlled trial,” requires a specialized level of human subject oversight in addition to IRB review. Given that such trials test the efficacy of a new technology, behavioral intervention, device, or therapeutic program, there is heightened concern for the safety of study participants in this type of design strategy. Thus, in addition to the IRB, another level of oversight is required to monitor the safety of study participants in randomized trials.
Assume you plan to test an intervention that involves introducing balance and strength exercises to older adults to reduce fear of falling. A potential harmful outcome of the intervention might be muscle strain, back injury, or possibly a fall as a result of increased movement and activity. Thus, the study must be monitored for the occurrence of these possible events. The investigator must track the occurrence of such events and determine whether a given event is a direct consequence of the intervention.
To monitor the safety of participants in randomized trials, the investigator is responsible for establishing an independent group of experts as a data and safety monitoring board (DSMB). The extent of involvement of a DSMB in oversight of a study depends on the level of risk associated with the treatment and study procedures. Studies that place participants at high risk require greater monitoring, whereas studies with minimal risk require less oversight.
Regardless of risk level, the primary responsibilities of a DSMB are (1) to review and approve all study procedures; (2) to provide oversight for procedures regarding the safety of human subjects and ethical research practices, including reviewing the investigator's approach to recruitment and the informed consent process; and (3) to determine whether and when a study should be terminated because of adverse events that can be attributed to study procedures.
The requirement for DSMB monitoring recently has expanded beyond pharmaceutical research to include behavioral treatments. Some institutions have created a DSMB to serve as a monitoring group for all clinical trials conducted at that setting. In most institutions, however, the investigator is responsible for establishing a committee composed of persons who have expertise in randomized trials and in the specific content of the treatment and who are not involved with members of the research team such that the potential exists for a conflict of interest.
It is important to recognize that even a controlled trial with minimal risk, such as testing a telephone information service to homebound patients, necessitates careful oversight to ensure human subject protection and ethical research practices. The DSMB can be a helpful mechanism for providing feedback and recommendations to the investigative team to enhance study rigor. Search the Internet for DSMBs to see how they function at different institutions. You will find a wide range of approaches, although all cover the essentials of safety monitoring and human subject protection oversight.8
Summary
It is your obligation as a researcher to design protection protocols so that all populations, regardless of literacy level or physical or cognitive capacity, can engage in the research process in a fully informed and ethical way. This is both an ethical and a legal obligation of researchers working in any research tradition involving human subjects. Before engaging in any type of research, you must submit a proposal to the IRB for review and approval. If a written consent is necessary, it must contain critical elements describing study procedures, voluntary capacity, risk-to-benefit ratio, and confidentiality procedures. A current, dated, and stamped consent must be used to enroll study participants. The process of obtaining consent is an essential aspect of bounding one's study.
As in all other thinking and action processes of research, the construction of protection protocols requires careful and thoughtful consideration. Unethical or inappropriate research harms not only study participants but also the target population, society, and the overall research enterprise.
References
1. National Research Act (Public Law 93–348)
2. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report: ethical principles and guidelines for the protection of human subjects of research. Washington, DC: The Commission, 1979.
3. Bluebond-Langner, M. The private worlds of dying children. Princeton, N.J.: Princeton University Press, 1978.
4. Office for Protection from Research Risks Tips on informed consent. US Department of Health and Human Services, 1993. http://www.hhs.gov/ohrp/humansubjects/guidance/ictips.htm.
5. Quest, T., Marco, C.A. Ethics seminars: vulnerable populations in emergency medicine research. Acad Emerg Med. 2003;10:1294.
6. Beauchamp, T.L., Jennings, B., Kinney, E.D., Levine, R.J. Pharmaceutical research involving the homeless. J Med Philos. 2002;27:547–564.
7. Kim, S.Y.H., Caine, E.D., Currier, G.W., et al. Assessing the competence of persons with Alzheimer's disease in providing informed consent for participation in research. Am J Psychiatry. 2001;158:712–717.
8. Ellenberg, S.S., Fleming, T.R., DeMets, D.L. Data monitoring committees in clinical trials: a practical perspective. Chichester, England: Wiley & Sons, 2003.