Abstract

Researchers face challenges in balancing the demands of sound research methods and ethical standards in studies involving persons with mental retardation. While the autonomy of individuals must be respected, researchers also have an ethical responsibility to protect vulnerable individuals from social, psychological, or physical risks of research participation. Some individuals may face significant barriers in the research decision-making process due to cognitive limitations, lack of information and support, and limited experience and opportunities for exercising choice. The historical context of biomedical and behavioral research involving persons with mental retardation and current ethical issues related to informed consent and surrogate decision-making are examined. Implications for researchers and professionals are discussed, including ways to support individuals with mental retardation and their families in research decision-making.

Researchers conducting studies that involve persons with mental retardation face the challenge of balancing the demands of sound research methods and professional ethical standards. There is an underlying tension in applying the often competing ethical principles of autonomy of the individual versus protection of the individual from harm. It is increasingly recognized that individuals with mental retardation should have the right to self-determination in making important life decisions, including whether or not to participate in a research study. However, professionals have an obligation to protect vulnerable individuals from harm, including potential social, psychological, or health risks of research participation. Families and advocates may also experience conflicting desires to help persons with mental retardation exercise their autonomy, yet at the same time, protect their well-being.

Making informed decisions about research participation is often a difficult issue for the general public, given the complexities of research protocols. Persons with mental retardation may experience more significant difficulties and barriers in this decision-making process due to their cognitive limitations, lack of information and support appropriate to their needs, and limited experience with opportunities for choice and decision-making. Although a diagnosis of mental retardation should not preclude individuals from participation in research, they may require significant assistance in decision-making and, in some cases, family members or guardians may be legally designated to make these decisions on their behalf.

The ethical challenges of conducting research with persons who have mental retardation are particularly salient in the 1990s, given current trends in the disability field as well as in genetics, biomedical, and social/behavioral research. The emergence of the self-advocacy movement has resulted in increased emphasis on the importance of consumer participation and decision-making. There is growing recognition that persons with disabilities can speak for themselves and serve as effective agents of change (Dybwad & Bersani, 1996). Many individuals with mental retardation play a primary role in planning and choosing where they live and work and what types of supports and services they wish to use (Lindsey, 1996).

The emphasis upon empowerment and self-determination has also influenced the roles of consumers in research. The participatory action research paradigm is based upon a philosophy of empowerment of persons with disabilities as active collaborators in the research process, not only as objects of study. According to this approach, consumers and family members work with researchers to define research questions, shape the research agenda, assist with data collection and analysis, and disseminate results (Meyer, Park, Grenot-Scheyer, Schwartz, & Harry, 1998).

Another significant research trend affecting persons with mental retardation is the implementation in many states of outcome-based quality assurance systems that assess consumer “quality of life” outcomes, such as self-determination, satisfaction, personal growth, and social relationships. Researchers face particular challenges in assessing consumer outcomes for people with severe disabilities and limited or challenging communication (Bradley et al., 1996).

The emerging field of genetic research also poses serious ethical, legal, and social concerns. Persons with mental retardation and their families need to understand the implications of genetic research in order to make informed personal choices about participation in genetic testing, research, or therapy. Some individuals and families may face psychological risk of harm from learning genetic information about themselves that they may not wish to know. They may also face employment and insurance discrimination if information about their genetic status is not kept confidential (The Arc, 1996).

My purpose in this paper was to explore the challenges of conducting ethical and sound research involving persons with mental retardation. The focus of this analysis is biomedical and behavioral research; other areas of research may involve different ethical issues. After a brief review of the history of research with this population, critical ethical issues are examined related to informed consent and surrogate decision-making, and ways to facilitate the research decision-making process for individuals with mental retardation and their families are suggested.

Historical Context

There are numerous accounts of research in which persons with mental retardation have been used as subjects without their consent and without safeguards for protection from harm. The egregious Nazi medical experiments violated the most basic human rights of their “research” subjects—persons with disabilities, children, frail older people, and other prisoners. Individuals who conducted these experiments were put on trial in the Nuremberg Doctors' Trials following World War II. As part of the judgment issued in these proceedings, the Nuremberg Code was established, which set forth universal principles for the protection of the rights and well-being of human subjects. The Code established that the informed consent of the human subject is “absolutely essential” and must be obtained in a voluntary manner (Annas & Grodin, 1992). It also established the rights of individuals to refuse to participate in research and to terminate their research participation at any time. Moreover, the Code specified the responsibilities of researchers to avoid “all unnecessary physical and mental suffering and injury” of research subjects and to utilize “the highest degree of skill and care” in the conduct of research (Annas & Grodin, 1992).

In the United States, there have been numerous studies involving research practices now regarded as unethical, including the United States Public Health Service Tuskegee syphilis study, the Willowbrook State School hepatitis study, and human radiation experiments conducted by the United States Atomic Energy Commission in two Massachusetts institutions. These studies were not intended to benefit the subjects directly nor contribute to knowledge about their condition, nor did they utilize

satisfactory informed consent procedures, either because the subjects were not given adequate information, because they simply were not given a choice whether to participate or not, or because the subjects suffered from cognitive impairments that rendered them unable to consent. (Berg, 1996, p. 18)

In the Tuskegee study, which was conducted over a 40-year period beginning in the 1930s, African American men with syphilis remained untreated for their disease (even after penicillin therapy became available) so that the natural course of syphilis could be studied by the researchers. The research subjects were deceptively promised free treatment for their disease, when in fact the intent of the study was to study the progress of the disease without treatment. In the Willowbrook State School experiments conducted from the 1950s to 1970s, children with mental retardation who were residents of the state institution were injected with strains of the hepatitis virus in order to study the progression of the infectious disease. Although parents gave consent for their children to participate in this study, they were not informed of the actual risks of the research (Grodin & Glantz, 1994). Moreover, the institution failed to protect the children against the virus, in spite of new evidence at the time of the study of the effectiveness of injections of gamma globulin to treat the disease (Grodin & Glantz, 1994, p. 19).

The Human Radiation Experiments involved a series of studies conducted by the Atomic Energy Commission from 1944 to 1974. One of these experiments, co-sponsored with Quaker Oats Company, involved residents of two Massachusetts institutions for persons with mental retardation as research participants. In the study, subjects were given minute tracer amounts of radiation in breakfast cereal in order to study mineral absorption in the human body. Although parents of the residents were asked to consent for their children to be enrolled in a special “Science Club,” they were not told the true purpose of the club or that research subjects would ingest tiny amounts of radioactive materials. These radiation experiments violated “basic moral principles that underlie informed consent—not treating others as mere means, promoting the welfare of others, and respecting self-determination” (Advisory Committee, 1996, p. 406).

Partly in response to public concerns about abuses of human experimentation, the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research was established in 1974 to examine ethical problems in research involving human subjects. The Commission's final Belmont Report set forth basic ethical principles for protection of human subjects, including (a) respect for persons (honoring the right of individuals to self-determination and autonomy); (b) beneficence/non-maleficence (maximizing possible benefits and minimizing possible harms); and (c) distributive justice (equitable distribution of burdens and benefits of research participation among populations) (National Commission, 1978). These principles have served as the ethical framework for professional and governmental guidelines regarding informed consent, assessment of risks and benefits of research, and selection of research subjects (Berg et al., 1991).

Since the mid-1970s, the federal government has issued various sets of regulations regarding the conduct of human subjects research that require the voluntary capable informed consent of the person, or if incapable, the consent of a legally authorized surrogate. Federal regulations also specify the roles and responsibilities of Institutional Review Boards (IRBs) in the review, approval, and monitoring of research protocols of studies receiving federal funds. Special research protections are required by federal regulation for certain populations (fetuses, pregnant women, prisoners, and children) because they are regarded as particularly vulnerable or unable to give meaningful consent in research. However, persons with cognitive or mental impairments have not been included in these protected populations. Although special guidelines for research with persons “institutionalized as mentally infirm” were proposed in the 1970s and 1980s by the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research and by the United States Department of Health, Education and Welfare, these regulations were never enacted, due in part to a lack of consensus among government officials and concerns that special regulations would unnecessarily restrict the conduct of much needed research or unfairly single out or stigmatize institutionalized persons with mental disorders (Moreno, 1998; National Bioethics, 1998).

In summary, although there is considerable public and professional concern about the conduct of research involving individuals with intellectual and mental disabilities, we currently lack clear federal regulations to protect the special interests of these individuals. Much discretion is currently left to states and to local Institutional Review Boards (IRBs) in providing guidance in these matters. To help provide guidance to local IRBs, The Office of Protection from Research Risks (1993) issued some guidelines for research involving persons with cognitive impairments and other special classes of subjects regarding selection of subjects, degree of risk, limiting risks, and problems of consent and competence. In addition, the National Bioethics Advisory Commission (1998) recently issued recommendations for changes in federal regulations and for other governmental, institutional, and organizational actions regarding research involving persons with mental disorders that affect decision-making capacity. These recommendations include special protections regarding research review bodies; research design; informed consent and capacity; categories of research; surrogate decision-making; and education, research, and support. The Commission's recommendations will, hopefully, serve as a momentum for the protection of persons with mental disorders and the conduct of sound ethical research.

In this paper I examine critical ethical issues related to informed consent and surrogate decision-making involving persons with mental retardation and their families. In the following discussion, case vignettes are used to illustrate some of these issues.

Consent Capacity and Information

Mary is a 28-year-old woman whose challenging aggressive behaviors interfere with her relationships at work and at home. She has been recruited to participate in a randomized clinical trial of a medication to help control these behaviors. Research staff have met with Mary and her parents to explain what the study would involve and to determine whether Mary has the capacity to consent to participate in this study. Although Mary says that she understands what the study is about, she does not seem to comprehend fully what participation will mean for her. For example, she does not understand the concept of a randomized trial—that she may be assigned to the experimental group or she may receive a placebo. Mary believes that if she participates in the study, she will definitely get the medication and that her behaviors will definitely improve. She is also confused about the potential side effects of the medication and questions why she would need to have weekly blood tests while on the medication.

One of the critical concepts underlying ethical research is consent, which involves three elements: capacity, information, and voluntariness (Turnbull, 1977). Capacity refers to the mental process or faculty by which an individual acquires knowledge, including the ability to “select and express one's choices” and to “engage in a rational process of making decisions” (Turnbull, 1977, p. 7). Information as a prerequisite of consent involves the substance of the information conveyed to the individual (“what” information is given) and the manner in which the information is communicated (“how” information is given) (Turnbull, 1977, p. 8). Voluntariness implies freedom to exercise choice without “overbearing coercion, duress, threats or inducements, and undue influence” (Turnbull, 1977, p. 11).

Assessment of consent capacity should focus on

the ability of the potential subject to understand the nature of the research participation, appreciate the consequences of such participation, exhibit ability to deliberate on alternatives, including the alternative not to participate in the research, and evidence ability to make a reasoned choice. (High et al., 1994, p. 69)

Capacity involves not only an ability to grasp the facts of the particular research situation, but also an appreciation of how the proposed study actually applies to the individual in the context of his/her own life (Bioethics Interest Group, 1988).

There are no established standards, measures, or mental status tests that assess the specific capacity of individuals to provide consent for research participation. However, there are some general principles that may guide researchers and professionals in evaluating the capacities of prospective research subjects. First, persons with mental retardation should not be presumed to be incompetent to participate in research solely on the basis of their diagnosis. Second, assessment of consent capacity should be made on a case-by-case basis, specific to particular research projects. Third, capacity to consent involves the interaction of an individual's ability “in the context of the specific risks and benefits of research . . . thus, a patient may be effectively able to consent to participate in an observation study and not in a phase one drug trial” (Dubler, 1987, p. 547). Some researchers and ethicists maintain that a higher level of decisional capacity should be required when “research participation presents significant risks or no chance of benefit to subjects than when the risk-benefit ratio is more favorable for subjects” (Dresser, 1996, p. 68).

Consent not only involves cognitive functioning and reasoning but reflects emotional components. For example, an individual's appreciation of the nature and consequences of research participation may be heavily influenced by his or her overall psychological condition (e.g., anxiety, depression, paranoia). A person with paranoid delusions may refuse to participate in an experimental medication trial for fear of invasion by outside agents; a severely depressed person may be unable to appreciate the potential benefits of an experimental therapy due to feelings of hopelessness. Given the large numbers of persons who have dual diagnoses of mental retardation and mental illness (Reiss, 1994), assessment of an individual's affective state is a critical component of capacity determination.

Voluntariness

Linda lives in a residential facility. The staff in her home have recommended that she attend a sexual abuse prevention program designed to increase knowledge about sexual abuse and enhance self-protection skills. In order to evaluate the effectiveness of this program, Linda would be interviewed at pretest and posttest using a structured personal safety questionnaire and case scenarios. Linda is uneasy about being part of this program because she fears it will “get too personal” and that she might “get upset” when she is interviewed. However, she is afraid to say no—her residential staff have urged her to be part of the study, and all of her other housemates are participating. She agrees reluctantly to participate because she is afraid she will “get in trouble with staff” if she doesn't attend.

The premise of informed consent is that decisions are made freely and without coercion and that individuals may withdraw at any time from the study without adverse penalties. Persons with mental retardation may be vulnerable due to their general lack of experience with decision-making, reliance upon family and staff, desire to please others, and potential susceptibility to undue influence or coercion from others (e.g., caregivers, therapists, researchers). Certain populations (e.g., involuntary patients and prisoners) may be especially at risk due to the unequal power relationship between doctors and patients (Fulford & Howse, 1993). Dinerstein, Herr, and O'Sullivan (1999) noted that “residents of institutions are a convenient population on which to conduct studies because they are more easily observed and controlled than individuals living on their own” (p. 113). Institutional residents may be reluctant to refuse to participate or withdraw from a study because they fear angering or disappointing those who treat them (Dinerstein et al., 1999) or fear “subtle discriminatory practices or attitudes among the caregivers” (Cassel, 1988, p. 91). Dubler (1987) suggested that the more controlling and confining the setting, the more individual ability should be required to avoid duress and coercion in obtaining informed consent. As a safeguard against such coercion, the National Commission on the Protection of Human Subjects (1978) recommended that in research on mental disabilities, subjects should be recruited from among noninstitutionalized populations wherever possible. If investigators propose to involve institutionalized individuals, there should be sufficient justification for using that population.

Making informed decisions about research participation may be particularly problematic in clinical experimental studies that involve the use of randomization, control groups, and double-blind procedures. Potential research subjects may find it difficult to comprehend the complexities of experimental protocols and to recognize that participation in clinical trials may not necessarily be personally beneficial or therapeutic. Appelbaum, Roth, Lidz, Benson, and Winslade (1987) described the problem of “therapeutic misconception . . . a denial of the possibility that there may be major disadvantages to participation in clinical research that stem from the nature of the research process itself” (p. 20). In a study of psychiatric patients, Appelbaum et al. found that individuals failed to recognize that scientific methodology had other than a therapeutic purpose; patients often assumed that the research investigator was acting in their best interests and that participation would provide them individualized treatment and personal care. The potential for therapeutic misperception extends beyond psychiatric patients and beyond investigations in which the patient's physician is the investigator. This conflict may occur in any type of research in which a professional has a therapeutic role with potential research subjects.

Clinical researchers may find it difficult to acknowledge the uncertainty of therapeutic benefit to potential research subjects (who are often their patients). The subtleties and potential conflicts of interests in relationships between patients and clinicians or researchers may be particularly complex for persons with mental retardation to comprehend. Because research investigators are under great pressure to enlist study volunteers, some advocates have recommended that trained, neutral educators be utilized to present information and discuss the advantages and disadvantages of research with prospective subjects (Appelbaum et al., 1987; Dresser, 1996). Neutral educators may play an important role in explaining and clarifying the complexities of clinical research methods to persons with cognitive limitations and ensuring that information is provided in a clear, concrete, and nonthreatening way. This may be particularly advantageous when the investigator is also the patient's physician, therapist, or counselor, and the methodology used is likely to be interpreted as therapeutic intent (Appelbaum et al., 1987).

Surrogate Decision-Making

Michael, who has Down syndrome and Alzheimer's disease, lives in a nursing home. Although Michael lived fairly independently prior to the onset of Alzheimer's disease, he is now unable to make important life decisions. Barbara, his legal guardian and sister, is responsible for making health and financial decisions on his behalf. The nursing home physician has asked Barbara for permission to enroll Michael in a clinical study examining the course of Alzheimer's disease in persons with Down syndrome. This study will involve routine clinical tests and behavioral observations. The researchers assure Barbara that participation in the study will not cause any harm or risk to Michael. She decides to give consent for Michael's participation because she believes that he would have wanted to participate in the study if he were competent to make the decision. Prior to the onset of Alzheimer's disease, Michael worked in a health care facility and took pride in taking care of others. Barbara has made a subjective judgment that Michael therefore would want to help others by participating in this health research, based upon his past actions and values.

In cases where an individual is judged to be incapable of providing informed consent, a qualified surrogate or proxy is often asked to provide consent for research participation on behalf of the individual. The surrogate decision-maker is typically a family member or may be a legally authorized representative, such as a guardian or designated health care agent (National Bioethics, 1998). Families are generally regarded as best suited to make treatment decisions on behalf of incapacitated relatives and, likewise, to make research decisions (Berg, 1996). In fact, researchers have found that older individuals often prefer that family members serve as their surrogates in treatment and research decisions (High et al., 1994; Sachs et al., 1994).

Because most persons with mental retardation live with their families, many for their entire lives (Fujiura, 1998), it is common for family members to serve informally or formally as surrogate decision-makers in important life decisions, such as treatment or research participation. Most states, in fact, recognize the presumptive authority of families as qualified surrogates in treatment decisions, assuming that “families are generally most concerned about the welfare of the individual and are usually most knowledgeable of the subject's values and preferences” (High et al., 1994, p. 72).

Two standards have traditionally been utilized in making proxy decisions on behalf of decisionally incapacitated individuals—“substituted judgment” and “best interests.” These standards have guided surrogates' decisions regarding medical treatment and research participation as well as other critical life decisions. In the case of research consent, substituted judgment involves deciding whether or not an individual should participate in a particular research protocol, based upon the surrogate's knowledge of what the individual would have decided if she or he were competent. The surrogate makes a subjective judgment based upon what is understood of the incapacitated individual's personal values, preferences, and past experiences. The surrogate may infer from the individual's past behaviors or reactions (e.g., to treatments, tests, or examinations) how the individual might respond to participation in specific research protocols. For example, if an individual has exhibited significant anxiety and stress in the past when undergoing routine physical exams or medical procedures, the surrogate might infer that the individual would not want to participate in a research protocol involving medical tests. On the other hand, a surrogate might decide that an individual would want to participate in a consumer focus group about services if the individual enjoys socializing and likes to voice his or her opinions in group situations.

Unfortunately, it is often difficult for the surrogate to know or infer what the individual would have preferred or decided. Use of the substituted judgment standard presupposes that the incapacitated individual was competent in the past or expressed personal values or preferences about participation in research. However, many persons with mental retardation may never have been competent or had opportunities to express their thoughts or desires about research.

Research indicates that surrogates often do not know, misinterpret, or disregard the perceived preferences of the incapacitated individual. Warren et al. (1986) studied the decisions by proxies for 168 patients in nursing homes about whether to permit the patients' participation in a study involving minimal risk. They found that a third of the proxies who believed that the patient herself or himself would not consent to the study consented to the participation of the patient anyway, a finding that seems “…to run directly counter to the concept of substituted judgment” (p. 1127). Muncie, Magaziner, Hebel, and Warren (1997) examined proxies' decisions about their charges' participation in clinical research, based upon a series of hypothetical research scenarios encompassing a variety of risks and benefits. They found that proxies made research participation decisions for their charges based upon what they would have decided for themselves, not based upon the substituted judgment principle of “standing in the shoes of the charge” (p. 932).

The best interests standard is often used when a surrogate is unable to make a substituted judgment, either because the individual's preferences about research participation are unknown or because the individual never had the capacity to formulate and communicate preferences. The surrogate decides whether or not to permit the individual to participate in a research protocol, based upon an assessment of the individual's best interests. In making this determination, the surrogate should evaluate the risk–benefit ratio of the proposed research—the potential for direct therapeutic benefit to the subject in relation to the degree of risk or potential harm entailed. Ideally, the research should involve minimal risk to the individual and maximum potential for therapeutic benefit.

Weighing Risks and Benefits

Carol, who has mental retardation, is recruited to be part of a national research project to identify genetic determinants of her condition. Because Carol lacks the capacity to provide informed consent, her parents (who are her legal guardians) are asked to give consent on her behalf. Carol has not communicated any personal preferences or values that might shed light on her desire to be part of this study. The parents must, therefore, weigh the potential risks and benefits of this study and ultimately decide what is in Carol's “best interest.” Although this study may increase scientific knowledge about the genetic determinants of their daughter's condition, there is no potential for Carol to benefit directly from this research. On the other hand, the risks to Carol for participating in this study are minimal—the research involves routine blood tests. Making a decision on Carol's behalf is difficult because they do not know what Carol would have wanted if she were competent, and they themselves have a strong personal desire to help advance scientific knowledge about their daughter's condition.

There are no clear-cut definitions of degree of acceptable risk or potential for therapeutic benefit. In cases such as Carol's, where the individual lacks decisional capacity, the surrogate ultimately must determine the best interests of the individual in the absence of specific guidelines. According to federal “Common Rule” regulations, a study presents minimal risk if “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” (Federal Policy for the Protection of Human Subjects, 1991, p. 6). However, this normative concept is vague and, therefore, difficult to assess. What constitutes “minimal risk” for one person may pose much greater risk for another individual. Some persons with mental retardation may have special vulnerability to harm and discomfort and may experience greater risks from specific research protocols than persons who are competent to make their own decisions.

Decisions by persons with limited decision-making capacity or their surrogates about participation in research involving no therapeutic benefit to the subject or greater than minimal risk are particularly problematic and controversial. The Nuremberg Code prohibits nontherapeutic research involving mentally impaired subjects. Similarly, a recent New York court decision held that incapacitated adults and minors could not participate in any psychiatric research that contained a nontherapeutic element, irrespective of possible benefits to the subject or the importance of the knowledge to be gained (Haimowitz, Delano, & Oldham, 1997). Berg (1996) maintained that research “involving greater than minimal risk without the possibility of direct therapeutic benefit should be impermissible because it will never be in the best interests of the subject” (p. 24). However, other researchers and advisory groups maintain that nontherapeutic research should be permissible under certain limited conditions and with appropriate safeguards in place to protect subjects (American College of Physicians, 1989; National Bioethics, 1998). Weisstub and Arboleda-Florez (1997) suggested that members of vulnerable populations should be enrolled in nontherapeutic research only when (a) there is no alternative pool of subjects, (b) the least competent prospective subjects are considered last, and (c) the risks of participation in the study are minimal. High et al. (1994) argued that research on Alzheimer's disease that involves potential risks and no direct benefits “may be justified if the anticipated knowledge sought is deemed to be of vital importance for understanding and alleviating the disease in the future and the specific research protocol is reasonably likely to generate such knowledge” (p. 72).

The relative emphasis placed on an individual's autonomy to make research decisions is somewhat dependent on the specific risks and benefits of the research involved. For research involving high risks and minimal or no therapeutic benefit for an individual, a high standard of decisional capacity may be necessary to ensure that the individual can understand and assess the complexities of the risk–benefit ratio. On the other hand, a lower threshold of capacity may be sufficient when a person makes a decision about research that involves minimal or no risk and a strong likelihood of therapeutic benefit.

Medical and behavioral research may present different types of risks and benefits than do other forms of social research. Unfortunately, there is no clear consensus on appropriate thresholds of decisional capacity for various types of research. This leaves much subjective leeway to researchers, guardians, family, and staff in assessing whether an individual has sufficient capacity to weigh the risks and benefits of the specific research project and whether the preservation of the person's autonomy should outweigh any potential concerns about his/her best interests.

Some individuals (with or without disabilities) may make decisions about research participation that are contrary to what other persons perceive as their “best interests.” Assessment of research risks and benefits is often difficult and complex for the general adult population. Researchers and professionals should not impose stricter standards of decisional capacity for persons with mental retardation than are typically expected for persons without impairments.

Assent of Research Subjects

Frank, age 35, lives at home with his mother, who is his legal guardian. When Frank was recruited by a university clinic to be part of a research project on social skills development, his mother gave written consent for Frank to participate. An interviewer then met with Frank to explain what the study would involve and obtain his assent to be tested. Although Frank has very limited verbal skills, he made it quite clear nonverbally that he did not want to be part of this study—he shook his head adamantly, banged his fist on the desk, and started pacing. The interviewer calmed him and reassured him that he would not be part of the study if he did not want to. In spite of having obtained the legal consent of the guardian, Frank had made his own choice not to participate in this study. The interviewer respected this choice.

The assent of the prospective research subject is important to obtain, in addition to surrogate consent for research participation. Assent involves the “willingness of the subject to go along with, or not object to, the proposed study” (Sachs et al., 1994). Even if individuals have been judged incapable of providing informed consent, they may be capable of expressing their thoughts and feelings about participation in a study and of exercising a minimal level of choice (Dresser, 1996). If a prospective subject objects (verbally or nonverbally) to participation, she or he should not be enrolled as a research subject, regardless of the proxy's consent decision. This method of “double informed consent” enables individuals to participate to the best of their ability in the consent process and respects the ethical principles of autonomy and self-determination (Resau, 1995, p. 56).

In addition to the case scenario just described (the surrogate provides consent, but the individual dissents), there are also situations in which the surrogate does not consent to the individual's participation, but the individual desires to take part in the study.

Susan, who lives in a group residence, has been recruited to participate in a consumer satisfaction survey in which she would be interviewed by an independent evaluator about her satisfaction with the services she receives. Susan told the evaluator that she wants to be interviewed for the study. However, her legal guardian has refused to give permission for her to participate. The guardian is concerned that Susan's answers would be unreliable “since she tends to say yes to everything.”

The legal guardian in this case does have ultimate authority to decide whether or not Susan can participate in the study. When there are apparent disagreements between the surrogate's and individual's preferences about research participation, it may be helpful for the researcher or a trusted advisor to meet with the guardian and the individual in order to review information about the study, address questions and concerns, and explore options for reconciliation of their differences. Perhaps in Susan's case as just described, the guardian would feel more comfortable about the interview if she or he could review in advance the types of questions that Susan would be asked or could be present at the time of the interview.

Advance Directives

Dave, who lives independently in an apartment, makes most day-to-day decisions on his own. When he faces a difficult or complex decision about his finances or health, he often seeks out the informal advice of his trusted friend and advocate, Bob. When Dave recently decided to prepare a will, his lawyer suggested that he also prepare an advance directive that designates a health proxy. The health proxy would have the authority to make treatment or research decisions on Dave's behalf in the event that he becomes unable to make these decisions on his own in the future. Dave feels close to his friend Bob and trusts his judgment to make decisions on his behalf if this becomes necessary in the future. After several discussions with Bob about health care preferences, Dave prepares an advance directive that designates Bob as his health proxy.

There has been some discussion and debate about the use of advance research directives (including “research living wills” or research durable powers of attorney) to specify in advance one's personal preferences about research participation in the event of future incapacity (American College of Physicians, 1989; Dresser, 1996; Moreno, 1998; National Bioethics, 1998). Although advance directives have typically been used to make advance plans about medical treatment or end-of-life care, these planning tools may also have relevance for advance research consent decisions.

There are two types of advance directives: instruction directives and proxy directives. Instruction directives are written documents in which an individual issues advance consent (or refusal) for medical treatment or research participation in the event of future incapacity or provides specific instructions or preferences about future treatment or research involvement. A proxy directive (or research durable power of attorney) is a written document in which an individual names and authorizes a proxy or agent to make decisions about treatment or research participation on behalf of the individual.

Although both of these types of directives may offer opportunities to exercise autonomous choice, it is generally easier to prepare a proxy directive, in which individuals designate a surrogate in whom they have trust and confidence to make decisions on their behalf in the future. Many prospective subjects incapable of independent research decision-making may be capable of selecting a research proxy because the decision-making capacity that is required to designate a proxy is far less than the capacity required to understand a detailed protocol (Sachs et al., 1994). Proxy directives may, therefore, be easier for individuals with mental retardation to understand and execute than are instructional directives that require more hypothetical, abstract thinking.

Although all states have statutes regarding the use of advance directives for medical treatment or end-of-life decisions, it is unclear how many states specifically allow the use of advance health care directives for decisions regarding research participation. Because advance directives are not widely used even for medical treatment (Sachs et al., 1994), the use of advance directives for research participation is also likely very limited. However, these advance planning tools may become more widely used as public information about these directives increases and policies promoting their use are implemented in research settings.

Advance research directives may be most useful when they serve as a tool to stimulate discussion between researchers, prospective research subjects, and proxy-decision-makers. The process of communication about an individual's preferences and values may have broader value than the actual preparation of an advance directive. Even when individuals lack the capacity to give informed consent, they may be able to communicate their beliefs and feelings (verbally or nonverbally), which may help the proxy make effective decisions on their behalf. For example, one's personal values about privacy, commitment to help others, desire to please, comfort in sharing personal opinions, and reactions to unfamiliar people or settings may help shed light on an individual's research preferences.

Implications

There are many ways in which researchers and professionals can help facilitate the research decision-making process in a sound and ethical manner. Strategies are needed to inform and support individuals with mental retardation in making informed decisions about research participation, to educate surrogate decision-makers, and to involve individuals and their advocates in the planning and monitoring of research.

Enhancing the Communication Process

Researchers have an ethical responsibility to develop and exercise the best possible communication skills in helping potential research subjects and their families understand their roles in research participation (High et al., 1994). Obtaining informed consent is a critical step in the research process, but not an end in itself. Researchers often assume that their ethical obligations are fulfilled once research subjects have provided consent to participate. Because individuals' preferences and concerns about the research may change over the course of the study, informed consent should be part of an ongoing communicative process between researchers, research subjects, and their families.

Because many adults with mental retardation have traditionally been denied opportunities to exercise choice and decision-making, “basic” information and training may be needed to help educate them regarding their rights and responsibilities to make decisions about research, treatment, and other important issues affecting their daily lives. In addition to offering preparatory training about decision-making and risk-taking, specific information about the research must be communicated in ways that are appropriate to the needs and cognitive levels of the prospective research subjects. Researchers should provide repeated opportunities, if necessary, to communicate information to subjects about the study in order to enhance their understanding and establish trust. Written information about the research protocol in small increments and in basic nontechnical language should be provided to subjects and their families. Concrete visual aids, such as slides or a video, explaining the purpose of the study and the roles of research subjects would be helpful. Case vignettes or role-play scenarios may also help illustrate different research situations and the process of making decisions about research participation.

Eliciting Preferences and Values

In addition to communicating the “facts” of the protocol, researchers need to elicit prospective subjects' values and preferences about research involvement. Regardless of the individual's cognitive capacity, she or he may have strong feelings about participating (or not participating) in a study. Even if individuals have been judged unable to provide informed consent, obtaining their assent is critical in situations in which surrogate decision-makers have consented on their behalf. Researchers need to be sensitive to individuals' preferences, which may be expressed verbally or behaviorally. Involvement of families, close friends, or advocates may facilitate this communication process because they are generally most knowledgeable about the individuals. Close family and friends may provide important contextual information regarding the person's values about autonomy, privacy, relationships, and social responsibility, which may help shed light on the person's preferences regarding research.

Providing Opportunities for Shared Decision-Making

Families, guardians, or advocates may play an important role in assisting individuals with mental retardation with decision-making about research. Many persons may feel comfortable having close family members or friends helping them with complex decisions. “Shared” decision-making is preferred by many older individuals and also by families of non-Western cultures, who place great emphasis upon family interdependence and connectedness (Kapp, 1991). There is “increased disenchantment with the highly individualized model of decision-making . . . even prospective subjects capable of autonomous decision making may welcome the involvement of a relative or close friend in the information disclosure and decision-making process” (Dresser, 1996, p. 71).

Individuals with mental retardation often rely on the assistance and advice of trusted family members or friends in making important decisions. Members of the individual's informal support network often have direct knowledge of the individual's personal beliefs and values, which can help guide the decision-making process. Families and friends may explain proposed research protocol in terms that the individual can relate to and may help the individual develop an understanding of the research and its possible consequences for him or her. Individuals may feel more comfortable asking questions and expressing their concerns and preferences with family members or advisers rather than with research investigators, who may be total strangers or with clinicians who may be perceived as persons in power.

Educating Surrogate Decision-Makers

Surrogate decision-makers assume enormous responsibilities when they make research or treatment decisions on behalf of individuals with cognitive limitations. Families and other designated proxies need practical information and instruction about the use and appropriateness of substituted judgment and best interests standards in surrogate decision-making. Researchers have found significant discrepancies between proxies' decisions on behalf of individuals and the preferences of the individuals themselves (Muncie et al., 1997; Warren et al., 1986).

Proxies need to be educated about making decisions that reflect the individuals' preferences or best interests as opposed to their own desires or needs. They need to learn how to differentiate their own feelings from those of the individual. For example, they need to be aware of and guard against making research decisions for persons with decisional incapacity based on their own assumptions or personal biases about the value or meaning of the individual's life or contribution to society. Information is needed about ways to apply ethical principles in surrogate decision-making and ways to balance competing ethical principles, such as autonomy and protection from harm.

Involving Persons With Mental Retardation and Families or Advocates in Research Planning and Monitoring

One strategy for helping ensure the ethical conduct of research with persons who have mental retardation is to involve members of the relevant study population in the planning and monitoring of the research. The perspectives of consumers and their advocates help to educate researchers and IRB members about the needs and concerns of the study population and their research priorities. The National Bioethics Advisory Commission (1998) recommended that IRBs reviewing research proposals involving persons with mental disorders should include “at least two members who are familiar with the nature of these disorders and with the concerns of this population. At least one should be a member of this population, or a family member, or advocacy representative” (p. 161). Similarly, ethics committees and human rights committees in agencies and hospitals should include consumers and their advocates to represent their perspectives on the conduct of ethical research. Meaningful representation of consumers is consistent with the principles of participatory action research, in which persons with disabilities serve as active collaborators in the overall research process, not just as research subjects.

In summary, the advancement of knowledge in the field of disabilities is dependent upon methodologically rigorous and ethical research. New treatments, interventions, and preventive and habilitative techniques emerge only through the conduct of scientifically sound research. Although we must be vigilant in supporting the vital role of basic and applied research in the acquisition of new knowledge in our field, advocates must aggressively safeguard the rights of individuals with mental retardation in research settings. The “ethics of doing research . . . and of not doing it” are equally important considerations (Haywood, 1976, p. 311).

References

References
Advisory Committee on Human Radiation Experiments.
1996
.
Research ethics and the medical profession: Report of the Advisory Committee on Human Radiation Experiments.
Journal of the American Medical Association
276(5)
:
403
409
.
American College of Physicians.
1989
.
Cognitively impaired subjects.
Annals of Internal Medicine
111
:
843
848
.
Annas
,
G. J.
and
M. A.
Grodin
.
1992
.
The Nazi doctors and the Nuremberg Code.
New York: Oxford University Press
.
Appelbaum
,
P. S.
,
L. H.
Roth
,
C. W.
Lidz
,
P.
Benson
, and
W.
Winslade
.
1987, April
.
False hopes and best data: Consent to research and the therapeutic misconception.
Hastings Center Report, 20–24
.
The Arc.
1996, October
.
Genetic discrimination.
In Genetic issues in mental retardation (Rep. 401–2). Arlington, TX: Author
.
Berg
,
J.
1996
.
Legal and ethical complexities of consent with cognitively impaired research subjects: Proposed guidelines.
Journal of Law, Medicine, and Ethics
24
:
18
35
.
Bioethics Interest Group.
1998
.
Research involving individuals with questionable capacity to consent: Ethical issues and practical considerations for institutional review boards.
Expert Panel Report to the National Institutes of Health. Bethesda, MD: Author
.
Bradley
,
V.
,
T.
Jaskulski
,
C.
Feinstein
,
L.
Brown
,
J.
Silver
, and
B.
Rabb
.
1996
.
Intermediate care facilities for persons with mental retardation: Compendium of state quality assurance systems.
Cambridge, MA: Human Services Research Institute
.
Cassel
,
C. K.
1988
.
Ethical issues in the conduct of research in long term care.
Gerontologist
28(Suppl.)
:
90
96
.
Dinerstein
,
R. D.
,
S. S.
Herr
, and
J. L.
O'Sullivan
.
(Eds.).
1999
.
A guide to consent.
Washington DC: American Association on Mental Retardation
.
Dresser
,
R.
1996
.
Mentally disabled research subjects: The enduring policy issues.
Journal of the American Medical Association
276
:
67
72
.
Dubler
,
N. N.
1987
.
Legal judgments and informed consent in geriatric research.
Journal of American Geriatrics Society
35
:
545
549
.
Dybwad
,
G.
and
H.
Bersani
.
1996
.
New voices: Self-advocacy by people with disabilities.
Cambridge MA: Brookline Books
.
Federal Policy for the Protection of Human Subjects,
.
45 CFR 46.102(i), 1991
.
Fujiura
,
G. T.
1998
.
Demography of family households.
American Journal on Mental Retardation
103
:
225
235
.
Fulford
,
K.
and
K.
Howse
.
1993
.
Ethics of research with psychiatric patients: Principles, problems and the primary responsibilities of researchers.
Journal of Medical Ethics
19
:
85
91
.
Grodin
,
M. A.
and
L. H.
Glantz
.
(Eds.).
1994
.
Children as research subjects: Science, ethics, and law.
New York: Oxford University Press
.
Haimowitz
,
S.
,
S. J.
Delano
, and
J. M.
Oldham
.
1997, November–December
.
Uninformed decisionmaking: The case of surrogate research consent.
Hasting's Center Report, 9–16
.
Haywood
,
H. C.
1976
.
The ethics of doing research . . . and of not doing it.
American Journal of Mental Deficiency
81
:
311
317
.
High
,
D. M.
,
P. J.
Whitehouse
,
S. G.
Post
, and
L.
Berg
.
1994
.
Guidelines for addressing ethical and legal issues in Alzheimer disease research: A position paper.
Alzheimer Disease and Associated Disorders
8(4)
:
66
74
.
Kapp
,
M. B.
1991
.
Health care decision making by the elderly: I get by with a little help from my family.
Gerontologist
31
:
619
623
.
Lindsey
,
P.
1996
.
The right to choose: Informed consent in the lives of adults with mental retardation and developmental disabilities.
Education and Training in Mental Retardation and Developmental Disabilities
30
:
171
176
.
Meyer
,
L. H.
,
H.-S.
Park
,
M.
Grenot-Scheyer
,
I.
Schwartz
, and
B.
Harry
.
1998
.
Participatory research: New approaches to the research to practice dilemma.
Journal of the Association for Persons with Severe Handicaps
23
:
165
177
.
Moreno
,
J. D.
1998
.
Regulation of research on the decisionally impaired: History and gaps in the current regulatory system.
Journal of Health Care Law & Policy
1(1)
:
1
21
.
Muncie
,
H. L.
,
J.
Magaziner
,
R.
Hebel
, and
J. W.
Warren
.
1997
.
Proxies' decisions about clinical research participation for their charges.
Journal of the American Geriatrics Society
45
:
929
933
.
National Bioethics Advisory Commission.
1998
.
Research involving subjects with mental disorders that may affect decisionmaking capacity (Vol. 1).
Rockville, MD: Author
.
National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research.
1978
.
The Belmont Report: Ethical principles and guidelines for the protection of human subjects of research (Vol. 1, Appendix).
Washington, DC: U.S. Government Printing Office
.
Office for Protection From Research Risks.
1993
.
Protecting human research subjects: Institutional review board guidebook.
Washington, DC: Author
.
Reiss
,
S.
1994
.
Psychopathology in mental retardation.
In N. Bouras (Ed.), Mental health in mental retardation: Recent advances and practices (pp. 67–78). Cambridge, England: Cambridge University Press
.
Resau
,
L. S.
1995
.
Obtaining informed consent in Alzheimer's research.
Journal of Neuroscience Nursing
27(1)
:
57
60
.
Sachs
,
G. A.
,
C. B.
Stocking
,
R.
Stern
,
D. M.
Cox
,
G.
Hougham
, and
R. S.
Sachs
.
1994
.
Ethical aspects of dementia research: Informed consent and proxy consent.
Clinical Research
42
:
403
412
.
Turnbull
III,
H. R.
(Ed.).
1977
.
Consent handbook.
Washington DC: American Association on Mental Deficiency
.
Warren
,
J. W.
,
J.
Sobal
,
J. H.
Tenney
,
J. M.
Hoopes
,
D.
Damron
,
S.
Levenson
,
B. R.
DeForge
, and
H. L.
Muncie
.
1986
.
Informed consent by proxy: An issue in research with elderly patients.
New England Journal of Medicine
315
:
1124
1128
.
Weisstub
,
D. N.
and
J.
Arboleda-Florez
.
1997
.
Ethical research with the developmentally disabled.
Canadian Journal of Psychiatry
45
:
492
496
.

Author notes

Author:Ruth I. Freedman, PhD, School of Social Work, 264 Bay State Rd., Boston University, Boston, MA 02215. (freedman@bu.edu)