Comminiity Dent. Oral Epidemiol. 1976: 4: 43-50
Ethics
(Key words: dcntat caries; ethics, dental)
Ethical considerations of study participants in dental caries clinical trials HERSCHEL S. HOROV/ITZ
Caries Prevention and Research Branch, National Caries Prograin, National Institute of Dental Researeh, NIH, Bethesda, Maryland, U. S. A. AiisTRACT - During the past 30 years there has been increasing concern for ethical considerations that pertain to the conduct of human biomedical research. Consequently, many national and international medical and dental organizations and agencies have developed reguhitions, policies or ethical guidelines for the protection of human subjects who take part in clinical investigations. In the United States, more tiuui 500 research institutions have established permanent committees to review research in humans conducted by their institutions. Members of these connnittecs must represent a broad range of backgrounds, interests, and concerns. Prospective study subjects must be able to make an informed decision on whether to participate in any study, witliout any element of force, deceit, duress, or other form of constraint or coercion. Obtaining informed consent for studies of children, the mentally infirm, and persons with restricted civil freedom presents special problems. Ethical considerations also raise questions on appropriate designs for clinical studies, e.g. use of untreated controls. Dental studies, particularly those testing caries-preventive agents, raise special questions of design, informed consent, ethical procedures and the use of diagnostic radiographs. The director of a clinical study is responsible for the conduct of all personnel connected with the investigation. (Received for publieation 28 September, aecepted 11 Oetober 1975)
HISTORICAL BACKGROUND Medical experimentation on humans can be traced to the beginnings of recorded history. Until the late 19th century, however, research efforts were, for the most part, conducted in a nonsysteniatic manner and consisted primarily of clinically treating patients with agents or methods whose value had not been established in any scientific manner. Moreover, until recent times, the law contained little about the rights of human subjects who took part in research projects. In fact, before 1962, no specific U.S. federal or state statutes regulated institutions or investigators in their use of human subjects for experimental purposes. There have been several reasons for increasing interest in and concern for ethical considerations pertaining to the conduct of human biomedical research. Among them are the atrocious experiments conducted by Germati doctors during World War II, which led to condemnation of the perpetrators
of these "research" efforts as conducting crimes against humanity. The Nuremberg Code of Ethics for Medical Research emanated from the Nuremberg War Crimes Tribunal. One key provision of the Code states that the voluntary consent of the human subject is absolutely essential. Furthermore, each subject should be in a position to make a choice about participation, free from any clement of force, fraud, deceit, duress, or any form of constraint or coercion. In addition, he should have sufficient knowledge or understanding of the elements of the research project and the procedures to make an enlightened decision about participation. The Declaration of Helsinki, adopted by the World Medical Association in 1966, largely repeated the provisions of the Nuremberg Code. The Declaration does discuss the interesting concepts of distinguishing between research that may be expected to provide therapeutic benefit to the patient and clinical research, the objective of which is solely
44
HOROWITZ
scientific with no specific value to the participating subject. It provided that consent by the subject or his legally authorized representative should be voluntary, not coercive, that the research effort should have a solid scientific foundation and that the study should be expected to yield useful results for the good of society. Other organizations have developed their own regulations, policies, or ethical guidelines for the protection of human subjects who take part in clinical investigations. Among those that have received attention are the codes of the American Medical Association (1966) and the American Psychological Association (1972), and, in dentistry, the American Dental Association (1973) and the International Dental Federation (1973). The National Institutes of Health first developed procedures in 1953 to regulate clinical research done at its own Clinical Center. The requirement of prior review was extended in 1966 to cover all extramural research supported by United States Public Health Service (PHS) grants and awards. The review process has been extended and developed in the past few years so that, since 1971, all research agreements of the U.S. Department of Health, Education and Welfare (DHEW), both contracts and grants, that involve humans are subject minimally to a dual system of review. One of these reviews is made by the organization that will conduct the research under contract or grant. The other review is made by DHEW. Many studies done in the past that were considered legitimate or ethical at the time they were initiated or in which the protection of human subjects was not specifically considered would not pass the scrutiny of present-day reviewers nor meet current regulations. Among these are the study of untreated latent syphilis in black males in Macon County, Alabama (the highly publicized Tuskegee Study) begun in 1932; a study of the toxic effects of chloramphenicol in premature infants in Los Angeles County General Hospital (1958-1960) in which no parental consent was obtained; and a study of the rejection of cancer cell implants by non-cancer patients (1963). The classic studies by WALTER REED with Aedis aegypii vectors in producing yellow fever in healthy volunteers would be considered unethical today by most researchers. All these studies, of course, were done before adoption of current PITS policy in 1966.
Because of the Federal Government's dominant role in funding biomedical research in the United States, its regulations have had great impact on the conduct of research on humans. Largely in response to Federal regulations, more than 500 research institutes in the U.S. have estabUshed permanent committees to review research on human subjects conducted at their institutions. Another 500 agencies or institutions utilize ad hoc conmiittces. The membership of the institutional committees must represent a broad range of backgrounds, interests, and concerns. The members must be competent to review a proposed project scientifically and to determine its acceptability in terms of local law, standards of professional conduct, and community practice. In contrast, the review by DHEW is primarily technical, but it is also concerned with whether the study's design meets standards that do not interfere with or impair the rights and welfare of subjects. The dual reviews, thus, are complementary, the institutional review being concerned primarily with local standards, law, and practices and the Federal review considering broader risk-benefit probabilities. Before any research activity is approved and funded by DHEW, it must have undergone this dual review to establish that the rights and welfare of the subjects involved are adequately protected, that the risks to participants arc outweighed by potential benefits or by the importance of the knowledge to be gained, and that informed consent is to be obtained by methods that are adequate and appropriate. INFORMED CONSENT One of the principal concerns of those organizations that have developed regulations for the conduct of field studies has been the methods that have been used to obtain informed consent for each subject's participation. Development of the concept of "informed consent" dates from the late 195O's. Informed consent means that each individual (or his legally authorized representative) is able to exercise free choice in deciding whether to participate, without any clement of force, fraud, deceit, duress, or other form of constraint or coercion. The basic elements oi information necessary to obtain such consent include: (1) A fair explanation oi the procedures to he followed and their purposes, including identification of any that may be experimental;
Ethics in clinical trials (2) A description of any associated discomforts or risks that are reasonable to expect; (3) A description of any benefits that are reasonable to expect; (4) A disclosure of appropriate alternative procedures that might be advantageous for the subjects; (5) An offer to answer any inquiries concerning the procedures; and (6) An instruction that the person is free to withdraw his consent and to discontinue participation in the project or activity at any time without prejudice to the subject. Certain special provisions of informed consent have been or are being considered for incorporation in existing regulatory doctrines. For example, a provision of the DHEW regulations states that no request for participation shall include any exculpatory language through which the subject is made to believe that he has waived his legal rights, including the release of the institution sponsoring or conducting the research from liability for negligence. It is essential that a recruited subject be able to weigh the risks to benefits associated with his participation in a study. Special population groups There are special concerns about informed consent for studies that involve children, the mentally ill, or mentally defective. Such studies should be conducted only when there is no significant risk of physical or mental harm to the subject or when direct benefit to the subject is anticipated. There currently is also special concern for studies of individuals with limited civil freedom, such as prisoners. Institutionalized populations, particularly prisoners, are frequently sought as subjects for certain investigations in which the investigators would like all participants to live under near-identical conditions, i.e. exposed to the same environmental factors. Moreover, the use of prisoners as study subjects offers the advantage of being able to monitor them easily for any untoward side effects. Because prisoners are a captive group, there is always the danger that their consent to participate in research will be obtained because they arc under duress. Many aspects of institutional life may influence a prisoner to participate in a research activity. These influences may constitute coercion, even if unintended. Because prisons frequently offer little or no opportunity for productive activity, research projects
45
may appear to the prisoners to offer relief from boredom. Research projects may offer a source of income, an opportunity for better food, comfortable bedding and medical attention for prisoners. Moreover, a prisoner may believe that his participation will be viewed favorably by prison authorities or by the parole board that will determine the time of his eventual release. Thus, the inducement on the part of prisoners to participate in biomedical research projects may be difficult to resist. Although the prisoner-subject being recruited may be adequately informed about the proposed investigation and be competent to make judgments, he may also lack the capacity to give truly voluntary consent because his liberty is limited. Although not institutionalized, one must carefully evaluate the availability of free power of choice on the part of military personnel to participate in clinical studies, or, for that matter, of medical or dental students who may feel coerced into participating in investigations being conducted by researchers at their schools. There are circumstances that not only justify, but require, the participation of children in biomedical research. The disease or condition under consideration may occur only infrequently in adults, being confined largely to children. In such instances, children are the only subjects in whom the disease process and possible modes of therapy can be studied. In the United States, current Food and Drug Administration (FDA) acceptance of a new drug application includes a review of its proposed labeling, which must include "adequate directions for use" and "adequate warnings" as to unapproved uses. For a drug to be distributed in interstate commerce and approved for use by children (or pregnant women), it must have been tested adequately for safety, efficacy, and dosage in those groups. If such testing has not been done, the label must state this fact explicitly. Thus, participation of children in drug research is the only means of licensing new drugs for their use and benefit. An investigator who proposes research activities among children must document that the risk to health is either minimal or that, although some risk exists, the benefit is significant and far outweighs that risk. Special provisions of proposed DHEW regulations on research in children stipulate that consent of bolli parents should be obtained before any child may participate in research activities and
46
HOROWITZ
that children, themselves, must be given the opportunity to refuse to participate. Although it is unlawful and morally wrong to proceed with consent of a child-subject but without the consent of his parents or legal guardian, it is believed by some that the reverse is also true, and that the consent of the childsubject must also be obtained when the child has reached an appropriate "age of discretion." Current regulations of DHEW state that only the "assent," or passive agreement, of the child be obtained, rather than his active consent. The institutionalized mentally infirm arc doubly incapacitated and disadvantaged with regard to deciding whether to participate in research activities: like children, they may lack the capacity to understand relevant information and to make judgments and, like prisoners, they are confined under conditions that limit their civil freedom. They thus require consideration for protection that pertains both to children and to prisoners. Because of their special problems related to obtaining informed consent, research projects among institutionalized mentally infirm persons should be limited to studies of the etiology, pathogenesis, diagnosis, treatment, prevention, management or prevention of the disability from which they suffer. "Informed consent" and "the ethical justification for human experimentation" are presently accepted principles for the conduct of research in humans. The application of these principles, however, leads inevitably to serious differences in interpretation of acceptable practices, e.g. the use of placebos, and benefit to the individual versus benefit to mankind. ETHICAL CONSIDERATIONS Because ethics is the study of the general nature of morals, specifically moral choices, there arc no correct answers as to how a study should be designed or conducted. In this context, ethical considerations must be distinguished from legal considerations. Laws and regulations may demand that certain rules be followed, but because of the diverse possibilities for clinical research, these rules, by necessity, must be general. The interpretation of these rules, as governed by acting in accordance with standards and precepts of behavior usually practiced in the proposed study's locale, constitutes the ethical and moral considerations associated with research projects. There have been few papers in which specific
ethical considerations associated with dental investigations have been discussed. Although some dental organizations have developed ethical guidelines for the conduct of dental research projects, these guidelines have all tended to be similar to one another and to correspond closely to guidelines for medical research, even though some considerations are unique or largely confined to dental studies. For example, most medical studies or evaluations of drugs are done on patients who have a particular disease, whereas most dental studies, particularly those on dental caries, are done to prevent the occurrence, or additional occurrences, of a disease process. Ih'^ divergency raises questions of whether subjects m dental field trials are "normal volunteers" or "p^' tients", the consideration of which has its own set ol ethical concerns. As an example, it is not ethical to withhold an effective treatnient from patients \vho are afflicted by a disease, but it is generally accepted that various preventive agents may be evaluated on subjects who do not (yet) have a disease to which they may be susceptible, or at risk. Subjects in studies of dental caries preventives are "normal," at least with regard to their teeth that have not yet developed dental caries and for which prevention oi dental caries is sought. ETHICS AND CONSENT IN DENTAL STUDIES The remainder of this paper is devoted to discussing some special eonsiderations in the conduct of studies of caries-preventive agents as they relate to the design of studies and to various aspects of obtaining informed consent. General reference will be made to studies that contain questionable ethical aspects, but individual studies will not be identified by name or by reference. Design oj dental caries trials In most studies of agents or methods that may reduce the incidence of dental caries, one or more test group? are given or are asked to use a test product. The investigators in these studies usually wish to measure the caries-preventive effect of each test product. At times, one of the test products may be a positiveeontrol, or an agent that has been tested widely and whose cariostatic effect is established and, ideally, known precisely or at least within a restricted range of efficacy. The normal incidence of dental carief in a clinical trial is measured among subjects in a control group. In many studies, the subjects in a
Ethics in clinical trials control group are given a placebo so that they believe they are being treated. Ethical considerations are readily apparent with this design. For example, is it ethical or moral to give a placebo to volunteers who have agreed to participate in a study of cariespreventive agents? Such studies must be done if we wish to learn the effects of therapeutic or preventive agents that may be developed. If caries-preventive agents whose exact effectiveness was known under all conditions of use were available to use as positive controls, then untreated control groups might not be necessary. Unfortunately, this is not the case, and thus control groups, either untreated or placebo-treated, will continue to be needed for some time in field trials of caries-preventive agents. Subjects who are being recruited for studies, however, should be informed of the likelihood of their receiving a placebo. For example, in a study with four groups, three of which will receive a thoughtto-be preventive regimen and one which will serve as a control, it should be clearly stated in the explanatory letter that accompanies the consent form that there is a one in four, or a 25 %, chance that a participant will receive an ineffective treatment. To compensate for the possible disadvantage of being a subject in a study's control group, the investigators may arrange to provide the preventive treatments used in the study to participants in the control group following the termination of the clinical phase of the investigation or, if the study is one in which a therapeutic dentifrice is being evaluated, perhaps control group subjects could be given a generous supply of dentifrice, toothbrushes and dental floss, with instruction for their use, when the study ends. One obvious ethical consideration of studies that include a control group deserves mentioning. The control group should not be asked to engage in any practice that may have a harmful effect on the dentition. For example, subjects in a control group should not be given or asked to use chewing gum that is sweetened with sucrose, nor should they be asked to use sucrose rinses or other agents to produce decalcification of enamel or initial carious lesions, unless the investigators are certain that they can "reverse" the defects, e.g. with fluoride. Studies have been done in the past in which institutionalized populations have been given candies and other sucrose-rich, oral retentive snacks at var^dng, sometimes frequent, intervals during each day. Although
47
results from these studies have greatly increased our knowledge of the effects of ingesting various types of refined carbohydrates on dental caries, the regimen may be deemed to be unacceptable by today's ethical standards. There are, of course, exceptions to every rule. It may be eonsidered to be ethical to conduct such studies if provision is made to give complete restorative dental care during or at the termination of the studies to subjects in an institutionalized population who do not ordinarily have care available to them and who ordinarily receive a diet conducive to the development of dental earies. Neither should control subjects be asked to refrain from their usual practices that may have a beneficial effect in reducing the incidence of dental caries. Thus, it is not appropriate to request that subjects in a study postpone visiting their dentists for routine dental care or forego engaging in their usual oral hygiene practices for more than just a few days. Sometimes the ready availability of effective caries-preventive products from commercial sources may raise questions on the ethics of a particular study design. For example, in the United States, fluoride-containing dentifrices currently constitute about two-thirds of the total sales of commercial dentifrice. Is it ethical to ask a control group to use a placebo or inert dentifrice when most of them are routinely using a product that can reduce the incidence of dental caries by 15 to 30 % when used ad libitum at home? Probably not. In fact, the reality of a study design is questionable in which a control group is asked to use a placebo dentifrice when much of the true control population (of the United States) is using a therapeutic dentifrice. Under present-day circumstances a study group using a placebo dentifrice constitutes an artificial situation that is not representative of the population at large. Several studies done in Scandinavian countries in the past several years have not included a traditional control group. Because many children in some Scandinavian countries routinely take part in school-based, fortnightiy mouthrinsings with 0.2 % neutral sodium fluoride solutions, these studies have used children on this regimen as the benchmark for comparison in their evaluations of alternative preventive tnethods. Again, this design is entirely realistic and appropriate for the situation. Ethical idealists have claimed that it is morally wrong to withhold effective agents or methods from
48
HOROWITZ
any participants who take part in a study. In essence, they say that all participants should be given topically applied fluorides in various forms, restorative dental care, oral hygiene instruction, etc. Those who propound this design, however, have not been able to explain satisfactorily how new informatioji on dental caries prevention can be learned when all study subjects receive preventive attention. Clearly, there are conflicts at titnes between the demands and needs of society and the needs and, perhaps, rights of individuals. Another dilemma that faces the director of a research project in which an untreated or placebotreated control group is used is what to do about children assigned to the control group who are highly susceptible to dental caries. Sound research design stipulates that such children must not receive preventive treatments, yet compa-ssion dictates that preventive action be taken. There is no correct answer to this problem. Although there is no legal obligation to treat subjects, there are legal considerations about informing subjects of their condition. One possible solution to the problem is initially to eliminate children with rampant caries from an investigation and place all stich children in the study's treatment group thought-to-be most effective for the prevention of decay. Of course, findings of these children are not included in the analyses. There arc alternatives in design to the controlled clinical investigation (which implies the use of a control group) that may be used for some studies and that avoid having to withhold treatments from any participants. For example, the effects of a preventive regimen may be evaluated by determining the dental caries prevalence in a population or sample that has participated in a preventive program and comparing the findings with a baseline caries prevalence observed in the same population before the measure or procedure was instituted. This design has been used for most trials of community water fluoridation and school water fluoridation. The evaluation of dental health education programs may also utilize a cotnparisoii of cross-sectional data before and after the programs M'ere implemented. However, it should be borne in mind that inherent in studies of this design are possible effects of changes with time in the natural occurrence of dental caries in a community together with the effects of the method under investigation, which may confound interpretation of the study.
Radiographs in dental trials The use of radiographs in clinical trials to supplement findings of visual-tactile examinations is currently another controversial subject, particularly because the effects of ionizing radiation are cumulative and can be deleterious. Not all investigators agree on whether radiographs are necessary to all clinical field trials. This paper is not the proper forum to discuss the need for radiographs in clinical trials. It is sufficient to say that there may be studies in which it is important to have radiographic information, e.g. cost-effectiveness studies where total dental caries incidence is needed, studies that include restorative care, or short-term studies of agents that are known to have a selective effect on approximal surfaces. Parents should be so informed if dental radiographs will be taken of their children during a clinical field trial. The number and frequency of taking radiographs should be told to the parent and a statement should be included, similar to the following: "[U.S.] Federal regulations require that you be informed that X-rays are potentially hazardous. However, the amount of radiation employed in this study is consistent with levels routinely used by your dentist for diagnostic purposes." If radiographs are to be used in a study, the safest available equipment should be used. Before a machine is used for each survey, it should be serviced and checked completely to ensure that it is operating at maximal efficiency and that there is no leakage of radiation. Maximal protection should be afforded to participating subjects and to the personnel who will be exposing the radiographs. Subjects must be provided with lead-shielded aprons during the radiographie procedures. The "fastest" available film and minimal milliamperage should be used. A lead diaphragm should be used to collimate the X-ray beam. To avoid the need for "re-takes", operators must be well trained and should use a film-positioning device to enhance reproducibility of radiographic images. Operators must wear film safety badges to record possible stray radiation. Large, portable lead shields with a leaded window should be provided for the operator, who must step behind the shield in order to expose each film. Children awaiting examination
Ethics in clinical trials 49 must be kept at a safe distance from the radiographic examination area. Dual-film packets should be used for caries-preventive studies. Parents should be informed that a set of films will be sent to their child's dentist, if requested. This arrangement may avoid unnecessary repetition of radiographic procedures. Parents should also be informed that they can request the results of the visual-tactile examinations. It is probably best not to quantify the number of carious lesions detected by the study's examiners, because this number may not agree with the diagnosis of the child's own dentist, but parents certainly have the right to be informed that the study-examiners believe that the child should seek immediate dental care. The director of a field study in which a dental caries-preventive agent is being evaluated is responsible for the conduct of all personnel connected with the investigation. If the caries-preventive agent is professionally applied, the operators who make the applications should be told to look for any untoward reactions that may occur with use of the agent(s) under investigation. A sampling of treated children should be recalled for observation at various intervals following treatment to check for any adverse reactions; not all reactions occur immediately. Treatments should be stopped immediately if unexpected, adverse reactions occur. The increasing attention being given to the need for informed consent of subjects has resulted in emphasis on a study's consent form being used as the document in which to present the essential elements of necessary information to prospective study-subjects. The consent form should provide space both for granting permission and for withholding permission, and, if possible, all forms distributed to children to take home for their parents' consideration or mailed to parents should be returned. It is not acceptable to word permission forms in a format that asks parents to return and sign the form only if they do not wish to have their child participate, and then to assume that all non-returned forms give tacit approval. A good procedure to follow is to have forms signed by a parent returned in all instances, indicating whether or not a child may participate. An explanatory letter to parents should aecompany the consent form. In addition to the other items required for the parent to give informed con-
sent, the letter should contain the name, title, address and telephone number of someone in authority in connection with the study. It should be stated that this person should be contacted if there are any questions about the study before permission is granted or if questions arise during the course of the study.
REFERENCES SUGGESTIONS FOR ADDITIONAL READING 1. AMERICAN
PSYCIIOLOCICAL
ASSOCIATION,
INC.: Ethical
principles in the conduct of research with human participants. Am. Psych. Assoc, Washington 1973. 2. AsTLEY HOPE, H . D . : The ethics of clinical trials in England today. Letter to Editor. Br. Dent. J. 1975: 138: 43. 3. CouNctL ON DENTAL RESEARCH, AMERICAN DENTAL AS-
SOCIATION: Ethical guidelines for clinical investigation: ethical policy of the American Dental Association regarding the use of human stibjects in clinical research. /. Am. Dent. A.^soe. 1973: 86.- 687-689. 4. Dental ethics and clinical trials. Editorial. Br. Dent. J. 1975: 138: 41-42. 5. Dl';PARTMEN'T OF H E A L T I I ,
EDUCATION AND WELFARE,
OFFICE OF THE SECRETARY: Protection of human sub-
jects. Federal Register 39, No. 105, 1974: 18914-18920. 6. DEPARTMENT
OF HEALTH,
EDUCATION
AND WELFARE,
OFFICE OF 'riiE SECRETARY: Protection of human subjects, proposed pohcy. Federal Register 39, No. 165, 1974: 30648-30657. 7. DEPARTMENT
OF HEALTH,
EDUCATION
AND WELFARE,
OFFICE OF THE SECRErARY: Protection of human subjects, techtiical amendments. Federal Register 40, No. 50, 1975: 1185
Ethics
(Key words: dcntat caries; ethics, dental)
Ethical considerations of study participants in dental caries clinical trials HERSCHEL S. HOROV/ITZ
Caries Prevention and Research Branch, National Caries Prograin, National Institute of Dental Researeh, NIH, Bethesda, Maryland, U. S. A. AiisTRACT - During the past 30 years there has been increasing concern for ethical considerations that pertain to the conduct of human biomedical research. Consequently, many national and international medical and dental organizations and agencies have developed reguhitions, policies or ethical guidelines for the protection of human subjects who take part in clinical investigations. In the United States, more tiuui 500 research institutions have established permanent committees to review research in humans conducted by their institutions. Members of these connnittecs must represent a broad range of backgrounds, interests, and concerns. Prospective study subjects must be able to make an informed decision on whether to participate in any study, witliout any element of force, deceit, duress, or other form of constraint or coercion. Obtaining informed consent for studies of children, the mentally infirm, and persons with restricted civil freedom presents special problems. Ethical considerations also raise questions on appropriate designs for clinical studies, e.g. use of untreated controls. Dental studies, particularly those testing caries-preventive agents, raise special questions of design, informed consent, ethical procedures and the use of diagnostic radiographs. The director of a clinical study is responsible for the conduct of all personnel connected with the investigation. (Received for publieation 28 September, aecepted 11 Oetober 1975)
HISTORICAL BACKGROUND Medical experimentation on humans can be traced to the beginnings of recorded history. Until the late 19th century, however, research efforts were, for the most part, conducted in a nonsysteniatic manner and consisted primarily of clinically treating patients with agents or methods whose value had not been established in any scientific manner. Moreover, until recent times, the law contained little about the rights of human subjects who took part in research projects. In fact, before 1962, no specific U.S. federal or state statutes regulated institutions or investigators in their use of human subjects for experimental purposes. There have been several reasons for increasing interest in and concern for ethical considerations pertaining to the conduct of human biomedical research. Among them are the atrocious experiments conducted by Germati doctors during World War II, which led to condemnation of the perpetrators
of these "research" efforts as conducting crimes against humanity. The Nuremberg Code of Ethics for Medical Research emanated from the Nuremberg War Crimes Tribunal. One key provision of the Code states that the voluntary consent of the human subject is absolutely essential. Furthermore, each subject should be in a position to make a choice about participation, free from any clement of force, fraud, deceit, duress, or any form of constraint or coercion. In addition, he should have sufficient knowledge or understanding of the elements of the research project and the procedures to make an enlightened decision about participation. The Declaration of Helsinki, adopted by the World Medical Association in 1966, largely repeated the provisions of the Nuremberg Code. The Declaration does discuss the interesting concepts of distinguishing between research that may be expected to provide therapeutic benefit to the patient and clinical research, the objective of which is solely
44
HOROWITZ
scientific with no specific value to the participating subject. It provided that consent by the subject or his legally authorized representative should be voluntary, not coercive, that the research effort should have a solid scientific foundation and that the study should be expected to yield useful results for the good of society. Other organizations have developed their own regulations, policies, or ethical guidelines for the protection of human subjects who take part in clinical investigations. Among those that have received attention are the codes of the American Medical Association (1966) and the American Psychological Association (1972), and, in dentistry, the American Dental Association (1973) and the International Dental Federation (1973). The National Institutes of Health first developed procedures in 1953 to regulate clinical research done at its own Clinical Center. The requirement of prior review was extended in 1966 to cover all extramural research supported by United States Public Health Service (PHS) grants and awards. The review process has been extended and developed in the past few years so that, since 1971, all research agreements of the U.S. Department of Health, Education and Welfare (DHEW), both contracts and grants, that involve humans are subject minimally to a dual system of review. One of these reviews is made by the organization that will conduct the research under contract or grant. The other review is made by DHEW. Many studies done in the past that were considered legitimate or ethical at the time they were initiated or in which the protection of human subjects was not specifically considered would not pass the scrutiny of present-day reviewers nor meet current regulations. Among these are the study of untreated latent syphilis in black males in Macon County, Alabama (the highly publicized Tuskegee Study) begun in 1932; a study of the toxic effects of chloramphenicol in premature infants in Los Angeles County General Hospital (1958-1960) in which no parental consent was obtained; and a study of the rejection of cancer cell implants by non-cancer patients (1963). The classic studies by WALTER REED with Aedis aegypii vectors in producing yellow fever in healthy volunteers would be considered unethical today by most researchers. All these studies, of course, were done before adoption of current PITS policy in 1966.
Because of the Federal Government's dominant role in funding biomedical research in the United States, its regulations have had great impact on the conduct of research on humans. Largely in response to Federal regulations, more than 500 research institutes in the U.S. have estabUshed permanent committees to review research on human subjects conducted at their institutions. Another 500 agencies or institutions utilize ad hoc conmiittces. The membership of the institutional committees must represent a broad range of backgrounds, interests, and concerns. The members must be competent to review a proposed project scientifically and to determine its acceptability in terms of local law, standards of professional conduct, and community practice. In contrast, the review by DHEW is primarily technical, but it is also concerned with whether the study's design meets standards that do not interfere with or impair the rights and welfare of subjects. The dual reviews, thus, are complementary, the institutional review being concerned primarily with local standards, law, and practices and the Federal review considering broader risk-benefit probabilities. Before any research activity is approved and funded by DHEW, it must have undergone this dual review to establish that the rights and welfare of the subjects involved are adequately protected, that the risks to participants arc outweighed by potential benefits or by the importance of the knowledge to be gained, and that informed consent is to be obtained by methods that are adequate and appropriate. INFORMED CONSENT One of the principal concerns of those organizations that have developed regulations for the conduct of field studies has been the methods that have been used to obtain informed consent for each subject's participation. Development of the concept of "informed consent" dates from the late 195O's. Informed consent means that each individual (or his legally authorized representative) is able to exercise free choice in deciding whether to participate, without any clement of force, fraud, deceit, duress, or other form of constraint or coercion. The basic elements oi information necessary to obtain such consent include: (1) A fair explanation oi the procedures to he followed and their purposes, including identification of any that may be experimental;
Ethics in clinical trials (2) A description of any associated discomforts or risks that are reasonable to expect; (3) A description of any benefits that are reasonable to expect; (4) A disclosure of appropriate alternative procedures that might be advantageous for the subjects; (5) An offer to answer any inquiries concerning the procedures; and (6) An instruction that the person is free to withdraw his consent and to discontinue participation in the project or activity at any time without prejudice to the subject. Certain special provisions of informed consent have been or are being considered for incorporation in existing regulatory doctrines. For example, a provision of the DHEW regulations states that no request for participation shall include any exculpatory language through which the subject is made to believe that he has waived his legal rights, including the release of the institution sponsoring or conducting the research from liability for negligence. It is essential that a recruited subject be able to weigh the risks to benefits associated with his participation in a study. Special population groups There are special concerns about informed consent for studies that involve children, the mentally ill, or mentally defective. Such studies should be conducted only when there is no significant risk of physical or mental harm to the subject or when direct benefit to the subject is anticipated. There currently is also special concern for studies of individuals with limited civil freedom, such as prisoners. Institutionalized populations, particularly prisoners, are frequently sought as subjects for certain investigations in which the investigators would like all participants to live under near-identical conditions, i.e. exposed to the same environmental factors. Moreover, the use of prisoners as study subjects offers the advantage of being able to monitor them easily for any untoward side effects. Because prisoners are a captive group, there is always the danger that their consent to participate in research will be obtained because they arc under duress. Many aspects of institutional life may influence a prisoner to participate in a research activity. These influences may constitute coercion, even if unintended. Because prisons frequently offer little or no opportunity for productive activity, research projects
45
may appear to the prisoners to offer relief from boredom. Research projects may offer a source of income, an opportunity for better food, comfortable bedding and medical attention for prisoners. Moreover, a prisoner may believe that his participation will be viewed favorably by prison authorities or by the parole board that will determine the time of his eventual release. Thus, the inducement on the part of prisoners to participate in biomedical research projects may be difficult to resist. Although the prisoner-subject being recruited may be adequately informed about the proposed investigation and be competent to make judgments, he may also lack the capacity to give truly voluntary consent because his liberty is limited. Although not institutionalized, one must carefully evaluate the availability of free power of choice on the part of military personnel to participate in clinical studies, or, for that matter, of medical or dental students who may feel coerced into participating in investigations being conducted by researchers at their schools. There are circumstances that not only justify, but require, the participation of children in biomedical research. The disease or condition under consideration may occur only infrequently in adults, being confined largely to children. In such instances, children are the only subjects in whom the disease process and possible modes of therapy can be studied. In the United States, current Food and Drug Administration (FDA) acceptance of a new drug application includes a review of its proposed labeling, which must include "adequate directions for use" and "adequate warnings" as to unapproved uses. For a drug to be distributed in interstate commerce and approved for use by children (or pregnant women), it must have been tested adequately for safety, efficacy, and dosage in those groups. If such testing has not been done, the label must state this fact explicitly. Thus, participation of children in drug research is the only means of licensing new drugs for their use and benefit. An investigator who proposes research activities among children must document that the risk to health is either minimal or that, although some risk exists, the benefit is significant and far outweighs that risk. Special provisions of proposed DHEW regulations on research in children stipulate that consent of bolli parents should be obtained before any child may participate in research activities and
46
HOROWITZ
that children, themselves, must be given the opportunity to refuse to participate. Although it is unlawful and morally wrong to proceed with consent of a child-subject but without the consent of his parents or legal guardian, it is believed by some that the reverse is also true, and that the consent of the childsubject must also be obtained when the child has reached an appropriate "age of discretion." Current regulations of DHEW state that only the "assent," or passive agreement, of the child be obtained, rather than his active consent. The institutionalized mentally infirm arc doubly incapacitated and disadvantaged with regard to deciding whether to participate in research activities: like children, they may lack the capacity to understand relevant information and to make judgments and, like prisoners, they are confined under conditions that limit their civil freedom. They thus require consideration for protection that pertains both to children and to prisoners. Because of their special problems related to obtaining informed consent, research projects among institutionalized mentally infirm persons should be limited to studies of the etiology, pathogenesis, diagnosis, treatment, prevention, management or prevention of the disability from which they suffer. "Informed consent" and "the ethical justification for human experimentation" are presently accepted principles for the conduct of research in humans. The application of these principles, however, leads inevitably to serious differences in interpretation of acceptable practices, e.g. the use of placebos, and benefit to the individual versus benefit to mankind. ETHICAL CONSIDERATIONS Because ethics is the study of the general nature of morals, specifically moral choices, there arc no correct answers as to how a study should be designed or conducted. In this context, ethical considerations must be distinguished from legal considerations. Laws and regulations may demand that certain rules be followed, but because of the diverse possibilities for clinical research, these rules, by necessity, must be general. The interpretation of these rules, as governed by acting in accordance with standards and precepts of behavior usually practiced in the proposed study's locale, constitutes the ethical and moral considerations associated with research projects. There have been few papers in which specific
ethical considerations associated with dental investigations have been discussed. Although some dental organizations have developed ethical guidelines for the conduct of dental research projects, these guidelines have all tended to be similar to one another and to correspond closely to guidelines for medical research, even though some considerations are unique or largely confined to dental studies. For example, most medical studies or evaluations of drugs are done on patients who have a particular disease, whereas most dental studies, particularly those on dental caries, are done to prevent the occurrence, or additional occurrences, of a disease process. Ih'^ divergency raises questions of whether subjects m dental field trials are "normal volunteers" or "p^' tients", the consideration of which has its own set ol ethical concerns. As an example, it is not ethical to withhold an effective treatnient from patients \vho are afflicted by a disease, but it is generally accepted that various preventive agents may be evaluated on subjects who do not (yet) have a disease to which they may be susceptible, or at risk. Subjects in studies of dental caries preventives are "normal," at least with regard to their teeth that have not yet developed dental caries and for which prevention oi dental caries is sought. ETHICS AND CONSENT IN DENTAL STUDIES The remainder of this paper is devoted to discussing some special eonsiderations in the conduct of studies of caries-preventive agents as they relate to the design of studies and to various aspects of obtaining informed consent. General reference will be made to studies that contain questionable ethical aspects, but individual studies will not be identified by name or by reference. Design oj dental caries trials In most studies of agents or methods that may reduce the incidence of dental caries, one or more test group? are given or are asked to use a test product. The investigators in these studies usually wish to measure the caries-preventive effect of each test product. At times, one of the test products may be a positiveeontrol, or an agent that has been tested widely and whose cariostatic effect is established and, ideally, known precisely or at least within a restricted range of efficacy. The normal incidence of dental carief in a clinical trial is measured among subjects in a control group. In many studies, the subjects in a
Ethics in clinical trials control group are given a placebo so that they believe they are being treated. Ethical considerations are readily apparent with this design. For example, is it ethical or moral to give a placebo to volunteers who have agreed to participate in a study of cariespreventive agents? Such studies must be done if we wish to learn the effects of therapeutic or preventive agents that may be developed. If caries-preventive agents whose exact effectiveness was known under all conditions of use were available to use as positive controls, then untreated control groups might not be necessary. Unfortunately, this is not the case, and thus control groups, either untreated or placebo-treated, will continue to be needed for some time in field trials of caries-preventive agents. Subjects who are being recruited for studies, however, should be informed of the likelihood of their receiving a placebo. For example, in a study with four groups, three of which will receive a thoughtto-be preventive regimen and one which will serve as a control, it should be clearly stated in the explanatory letter that accompanies the consent form that there is a one in four, or a 25 %, chance that a participant will receive an ineffective treatment. To compensate for the possible disadvantage of being a subject in a study's control group, the investigators may arrange to provide the preventive treatments used in the study to participants in the control group following the termination of the clinical phase of the investigation or, if the study is one in which a therapeutic dentifrice is being evaluated, perhaps control group subjects could be given a generous supply of dentifrice, toothbrushes and dental floss, with instruction for their use, when the study ends. One obvious ethical consideration of studies that include a control group deserves mentioning. The control group should not be asked to engage in any practice that may have a harmful effect on the dentition. For example, subjects in a control group should not be given or asked to use chewing gum that is sweetened with sucrose, nor should they be asked to use sucrose rinses or other agents to produce decalcification of enamel or initial carious lesions, unless the investigators are certain that they can "reverse" the defects, e.g. with fluoride. Studies have been done in the past in which institutionalized populations have been given candies and other sucrose-rich, oral retentive snacks at var^dng, sometimes frequent, intervals during each day. Although
47
results from these studies have greatly increased our knowledge of the effects of ingesting various types of refined carbohydrates on dental caries, the regimen may be deemed to be unacceptable by today's ethical standards. There are, of course, exceptions to every rule. It may be eonsidered to be ethical to conduct such studies if provision is made to give complete restorative dental care during or at the termination of the studies to subjects in an institutionalized population who do not ordinarily have care available to them and who ordinarily receive a diet conducive to the development of dental earies. Neither should control subjects be asked to refrain from their usual practices that may have a beneficial effect in reducing the incidence of dental caries. Thus, it is not appropriate to request that subjects in a study postpone visiting their dentists for routine dental care or forego engaging in their usual oral hygiene practices for more than just a few days. Sometimes the ready availability of effective caries-preventive products from commercial sources may raise questions on the ethics of a particular study design. For example, in the United States, fluoride-containing dentifrices currently constitute about two-thirds of the total sales of commercial dentifrice. Is it ethical to ask a control group to use a placebo or inert dentifrice when most of them are routinely using a product that can reduce the incidence of dental caries by 15 to 30 % when used ad libitum at home? Probably not. In fact, the reality of a study design is questionable in which a control group is asked to use a placebo dentifrice when much of the true control population (of the United States) is using a therapeutic dentifrice. Under present-day circumstances a study group using a placebo dentifrice constitutes an artificial situation that is not representative of the population at large. Several studies done in Scandinavian countries in the past several years have not included a traditional control group. Because many children in some Scandinavian countries routinely take part in school-based, fortnightiy mouthrinsings with 0.2 % neutral sodium fluoride solutions, these studies have used children on this regimen as the benchmark for comparison in their evaluations of alternative preventive tnethods. Again, this design is entirely realistic and appropriate for the situation. Ethical idealists have claimed that it is morally wrong to withhold effective agents or methods from
48
HOROWITZ
any participants who take part in a study. In essence, they say that all participants should be given topically applied fluorides in various forms, restorative dental care, oral hygiene instruction, etc. Those who propound this design, however, have not been able to explain satisfactorily how new informatioji on dental caries prevention can be learned when all study subjects receive preventive attention. Clearly, there are conflicts at titnes between the demands and needs of society and the needs and, perhaps, rights of individuals. Another dilemma that faces the director of a research project in which an untreated or placebotreated control group is used is what to do about children assigned to the control group who are highly susceptible to dental caries. Sound research design stipulates that such children must not receive preventive treatments, yet compa-ssion dictates that preventive action be taken. There is no correct answer to this problem. Although there is no legal obligation to treat subjects, there are legal considerations about informing subjects of their condition. One possible solution to the problem is initially to eliminate children with rampant caries from an investigation and place all stich children in the study's treatment group thought-to-be most effective for the prevention of decay. Of course, findings of these children are not included in the analyses. There arc alternatives in design to the controlled clinical investigation (which implies the use of a control group) that may be used for some studies and that avoid having to withhold treatments from any participants. For example, the effects of a preventive regimen may be evaluated by determining the dental caries prevalence in a population or sample that has participated in a preventive program and comparing the findings with a baseline caries prevalence observed in the same population before the measure or procedure was instituted. This design has been used for most trials of community water fluoridation and school water fluoridation. The evaluation of dental health education programs may also utilize a cotnparisoii of cross-sectional data before and after the programs M'ere implemented. However, it should be borne in mind that inherent in studies of this design are possible effects of changes with time in the natural occurrence of dental caries in a community together with the effects of the method under investigation, which may confound interpretation of the study.
Radiographs in dental trials The use of radiographs in clinical trials to supplement findings of visual-tactile examinations is currently another controversial subject, particularly because the effects of ionizing radiation are cumulative and can be deleterious. Not all investigators agree on whether radiographs are necessary to all clinical field trials. This paper is not the proper forum to discuss the need for radiographs in clinical trials. It is sufficient to say that there may be studies in which it is important to have radiographic information, e.g. cost-effectiveness studies where total dental caries incidence is needed, studies that include restorative care, or short-term studies of agents that are known to have a selective effect on approximal surfaces. Parents should be so informed if dental radiographs will be taken of their children during a clinical field trial. The number and frequency of taking radiographs should be told to the parent and a statement should be included, similar to the following: "[U.S.] Federal regulations require that you be informed that X-rays are potentially hazardous. However, the amount of radiation employed in this study is consistent with levels routinely used by your dentist for diagnostic purposes." If radiographs are to be used in a study, the safest available equipment should be used. Before a machine is used for each survey, it should be serviced and checked completely to ensure that it is operating at maximal efficiency and that there is no leakage of radiation. Maximal protection should be afforded to participating subjects and to the personnel who will be exposing the radiographs. Subjects must be provided with lead-shielded aprons during the radiographie procedures. The "fastest" available film and minimal milliamperage should be used. A lead diaphragm should be used to collimate the X-ray beam. To avoid the need for "re-takes", operators must be well trained and should use a film-positioning device to enhance reproducibility of radiographic images. Operators must wear film safety badges to record possible stray radiation. Large, portable lead shields with a leaded window should be provided for the operator, who must step behind the shield in order to expose each film. Children awaiting examination
Ethics in clinical trials 49 must be kept at a safe distance from the radiographic examination area. Dual-film packets should be used for caries-preventive studies. Parents should be informed that a set of films will be sent to their child's dentist, if requested. This arrangement may avoid unnecessary repetition of radiographic procedures. Parents should also be informed that they can request the results of the visual-tactile examinations. It is probably best not to quantify the number of carious lesions detected by the study's examiners, because this number may not agree with the diagnosis of the child's own dentist, but parents certainly have the right to be informed that the study-examiners believe that the child should seek immediate dental care. The director of a field study in which a dental caries-preventive agent is being evaluated is responsible for the conduct of all personnel connected with the investigation. If the caries-preventive agent is professionally applied, the operators who make the applications should be told to look for any untoward reactions that may occur with use of the agent(s) under investigation. A sampling of treated children should be recalled for observation at various intervals following treatment to check for any adverse reactions; not all reactions occur immediately. Treatments should be stopped immediately if unexpected, adverse reactions occur. The increasing attention being given to the need for informed consent of subjects has resulted in emphasis on a study's consent form being used as the document in which to present the essential elements of necessary information to prospective study-subjects. The consent form should provide space both for granting permission and for withholding permission, and, if possible, all forms distributed to children to take home for their parents' consideration or mailed to parents should be returned. It is not acceptable to word permission forms in a format that asks parents to return and sign the form only if they do not wish to have their child participate, and then to assume that all non-returned forms give tacit approval. A good procedure to follow is to have forms signed by a parent returned in all instances, indicating whether or not a child may participate. An explanatory letter to parents should aecompany the consent form. In addition to the other items required for the parent to give informed con-
sent, the letter should contain the name, title, address and telephone number of someone in authority in connection with the study. It should be stated that this person should be contacted if there are any questions about the study before permission is granted or if questions arise during the course of the study.
REFERENCES SUGGESTIONS FOR ADDITIONAL READING 1. AMERICAN
PSYCIIOLOCICAL
ASSOCIATION,
INC.: Ethical
principles in the conduct of research with human participants. Am. Psych. Assoc, Washington 1973. 2. AsTLEY HOPE, H . D . : The ethics of clinical trials in England today. Letter to Editor. Br. Dent. J. 1975: 138: 43. 3. CouNctL ON DENTAL RESEARCH, AMERICAN DENTAL AS-
SOCIATION: Ethical guidelines for clinical investigation: ethical policy of the American Dental Association regarding the use of human stibjects in clinical research. /. Am. Dent. A.^soe. 1973: 86.- 687-689. 4. Dental ethics and clinical trials. Editorial. Br. Dent. J. 1975: 138: 41-42. 5. Dl';PARTMEN'T OF H E A L T I I ,
EDUCATION AND WELFARE,
OFFICE OF THE SECRETARY: Protection of human sub-
jects. Federal Register 39, No. 105, 1974: 18914-18920. 6. DEPARTMENT
OF HEALTH,
EDUCATION
AND WELFARE,
OFFICE OF 'riiE SECRETARY: Protection of human subjects, proposed pohcy. Federal Register 39, No. 165, 1974: 30648-30657. 7. DEPARTMENT
OF HEALTH,
EDUCATION
AND WELFARE,
OFFICE OF THE SECRErARY: Protection of human subjects, techtiical amendments. Federal Register 40, No. 50, 1975: 1185
