FMV-UBA-COMUNIDAD-REVISTAS Y LIBROS DIGITALES

Considerations over the 1991 guidelines for the Council of International Organizations of Medical Science (CIOMS) for epidemiological studies.

por Delia Outomuro

Resumen

En esta comunicación analizamos críticamente las normas del CIOMS de 1991 con el propósito de sugerir algunas modificaciones. En primer lugar, proponemos actualizar los conceptos de Salud Pública y de Epidemiología y considerar a la investigación epidemiológica como una de las funciones esenciales de la primera. Entendemos que debe prevalecer el respeto por la dignidad de las personas y, por lo tanto, de las reglas de consentimiento informado, privacidad, confidencialidad y veracidad. Enfatizamos la justicia como equidad. Insistimos en que los riesgos de una investigación no son solo biológicos sino también psicológicos, socio-culturales, y legales, y que todos ellos deben ser tenidos en cuenta. Asimismo, destacamos la participación de los comités de ética en investigación, independientes y con presencia de miembros de la comunidad, para la evaluación de protocolos. Finalmente, sugerimos incluir dentro de la pobalción vulnerable a los trabajadores asalariados.

Abstract

In this paper, we analyse and criticize 1991 CIOMS norms. Our purpose is to suggest some changes. First of all, we propose to updateconcepts such as Public Health and Epidemiology, as well as consider epidemiological research as one of the public health essential functions. We think that dignity must be respected so we must have respect for informed consent, privacy, confidentiality and veracity rules. We emphasised justice as equity. We insist on the importance of considering not only biological risks but also psychological, social, cultural and legal risks. Also we point out the roll of the research ethic committees which must be independent and include community members. Finally, we suggest to consider salaried workers as a vulnerable group.

Palabras clave

Epidemiología – Bioética – Investigación

Key words

Public health – Epidemiology – Bioethics – Research

Exposition of the problem

Bioethics was originated during the seventies, due mostly to the abuses committed in the researches with human beings. The history of science and of medicine in particular shows the manner in which such abuses, in some way or another, have been present, and how the efforts to avoid them have turned out to be unsuccessful.

Facing this reality, the Council for International Organizations of Medical Science (CIOMS), decided to elaborate norms appointed to lead the research in human beings in an ethical way. This is the origin of the 1993 guidelines that were revised later in the year 2002.

In the same way, CIOMS considered the necessity of making norms for the epidemiological research and in order to do so, they elaborated the 1991 guidelines. Indubitably, this last step meant an advance in the field of ethic research, since, until then, the interest for the protection of persons subject to investigation was centred in the biomedical research. It is 1991 CIOMS merit to admit the fact that the persons and the communities may suffer risks because of the gathering, the storage, and the usage of data in the epidemiological studies.

Nevertheless, we consider that these guidelines should be revised, for they present some dark points, which are contradictory or incomplete. Next, with the purpose of suggesting some modifications, we will analyse each of the items that conform these norms. At the conclusion, we will make a synthesis of the proposed modifications.

We wish to clarify AD INITIO the following issue: the epidemiological studies show, perhaps more clearly than the biomedical studies, the conflict between individual rights against society rights. In our opinion and from a Kantian point of view, no person should be considered a mean for no end, not even for a laudable end as the scientific knowledge could be (1). Notwithstanding, we recognize the existence of situations where the autonomy principle should cede in the presence of commonweal; but these situations should be the exception and not the rule when dealing with researches which only purpose is to increase the knowledge flow. The respect for the dignity of the individual will then be the conducting thread of our analysis.

Analysis of the guidelines

A) Considerations regarding the “Introduction”

We think it appropriate to redefine the concepts of “public health” and of “epidemiology”

Public health is the collective intervention of the State and of the Civil Society orientated to protect and improve the individual health. Traditionally, the adjective “public” has assimilated itself with the governmental action, this is, the public sector. However, a wider interpretation is possible. This interpretation includes not only the participation of the government but also the participation of the organized community (the public). Public health has also been related to the so called “non personal health services”, those applied to the environment (sanitation for instance) or to the collectivity (for instance, the education for the health). Moreover, public health can be understood as the gathering of personal services of preventive nature led to vulnerable groups (the programs of maternal-infantile attention). Finally, the expression “problem of public health” is often used, especially in non-technical language, to refer to sufferings of high frequency and great danger.

Lately, a more integral viewpoint for the public health concept has arisen, arguing that the adjective “public” does not refer to the gathering of services in particular, or to a property way, not even to a type of problem. It refers to a specific analysis level, that is, a level for the population. This perspective involves two aspects: as a knowledge field and as an action atmosphere. Public health is then understood as the health for the public, and therefore, covers all the collective dimensions of health.

The knowledge that public health requires is provided by diverse disciplines, articulated in function of its objective and its objects. This knowledge articulation is its epistemological substance, covering an interdisciplinary dimension. In this way, public health can make use of the methods of those different disciplines that nourish it. However, there is a discipline that seems to possess the greatest presence and potential: the epidemiology. This discipline supplies the public health method par excellence. We also consider that the epidemiological research is one of the essential functions of public health (2, 3).

The epidemiological studies are classified in observational and experimental. The observational studies contemplate cross - sectorial, case-control, and cohort studies. It is maintained that observational studies “involve no intervention other than asking questions” and that “these types of study carry minimal risk to study subjects”. In our opinion, these considerations minimize the risks of these studies.

We believe that the damages a person subject to investigation may suffer, exceed the biological plane; and that the psychological, social, and legal risks, should also be taken into account. According to the predominant social or legal frame, these risks can even be more important than the physical risks. An example of this could be a study about abortion in countries where abortion is illegal.

In the same way, several diseases are stigmatizing. Perhaps AIDS represents a paradigmatic case in this point, but each society or culture builds different meanings in connection with affections; therefore, the discriminatory or stigmatizing potentiality of a disease should be understood culturally, and not in an universal manner.

It is acknowledged that in the prospective studies “the informed consent of subjects is normally required” but it is said immediately that “although there are some exceptions - for example, very large cohort studies conducted exclusively by examining medical records”. Expressed in this way, it seems that the volume of that which has to be examined justifies the consent omission. The difficulty in the achievement of a study does not seem to be a valid reason for such an omission.
It is supposed that the retrospective studies “generally do not involve invasion of privacy” . The concept of privacy recognizes several definitions, but in a wider meaning, it is safe to say that all the information referred to a person corresponds to the privacy ambit and, prima facie, nobody has the right of acceding to this information without previous consent (4, p. 407-409). We say prima facie because there could be situations in which society rights may justify the infringement of the individual right to privacy.

It is said that the informed consent will be requested for this kind of studies “if a case-control study requires direct contact between research workers and study subjects” but it is not necessary “if it entails only a review of medical records”. Again, the criterion used to decide whether the consent should be requested or not, proves to be unjustified; it is not understood why the “direct contact” with the investigators is the demarcation criterion by itself.

A precise identification of every investigated person is claimed, at least in a first stage: “It is essential to identify precisely every individual studied [...]. After identities have been established to compile the statistical tables, all personal identifying information is obliterated, and therefore privacy and confidentiality are safeguarded”. This point also turns out to be very controversial and the fact that this requirement is truly unavoidable should be analysed. If it were in fact unavoidable, it would oblige to obtain the person’s authorization since, as we said before, all the information about someone belongs to that someone, that is, to the scope of that person’s privacy.
Finally, the “unethical character” of randomized studies, is recognised: “Such experiments involving human subjects should be regarded as unethical unless there is genuine uncertainty about the regimen or procedure and this uncertainty can be clarified by research”. We share this consideration. However, the usage of placebos is not mentioned. The randomized studies in which the control group receives the placebo are still frequent. In these studies, there are several principles and bioethical rules that can be transgressed; therefore, these studies should be exceptional and they are only justified if certain conditions are considered. The conditions are the following: the knowledge requested must be relevant, this knowledge cannot be obtained in any other way, the subjects involved should be aware that the information they receive will be partial (and they should agree to be deceived), and the subjects involved must not be exposed to considerable risks (9)

B) Considerations regarding the “Preamble”

It is manifested that, since the epidemiological research is carried out in “groups of people”, the regular codes (Nuremberg, Helsinki, etc) “do not adequately cover its special features” and that the protocols “proposals for epidemiological studies should be reviewed independently on ethical grounds”. It would probably be more adequate to say that such codes “turn out to be insufficient” to broach the ethical problematic of epidemiology, but they should not be ignored; an analysis from “independent” or “special” ethical principles is neither convincing, if this is what it is intended to be said.
Regarding the studies externally sponsored and carried out in developing countries, it is sustained that “a framework for the application of these guidelines is set by the laws and practices in each jurisdiction in which it is proposed to undertake studies”. This point should be expanded and clarified. As it is known, the practice of investigating in emerging countries is carried out because the permissive legislation that they possess allows these researches (5). It is suitable to remember the classic distinction between “legality” (conformability with the positive law) and “legitimacy” (conformability with ethics) and that not everything which is legal is legitimate. The fact that the legal frame of an outlying country allowed to carry out a research, is not condition enough to go through it if it lacks of legitimacy.
It is affirmed that “not all ethical principles weigh equally. A study may be assessed as ethical even if a usual ethical expectation, such as confidentiality of data, has not been comprehensively met, provided the potential benefits clearly outweigh the risks and the investigators give assurances of minimizing risks”. We can make here two considerations: firstly, the principialism, explicitly sustains that the four ethical principles are all at the same level; to sustain the contrary would imply to go through a new philosophical basis, and we believe this exceeds the purpose of these guidelines. So, if the four principles are accepted, there should be accepted like their authors (Beauchamp and Childress) propose them (4). When these principles enter in conflict, it is attempted to dissolve the dilemma opting for one or another under the circumstances and the relevant variables of each case. It can not be established a priori that the violation of confidentiality is justified when the benefits overcome the risks; what can be understood of benefit and what can be understood of risk, is something that must be decided in a particular way, and not in an universal way.

Secondly, the responsibility of deciding what are minimal risks falls back over the researchers. In some way, the previous quotation confides in their prudence. But researchers can neither evaluate nor take decisions with objectivity because they are interested part. The scientific knowledge is not neutral, and much less neutral are their social actors (6, 7, 8). The independent committees of ethic in research should be in charge of taking these decisions. (9,10,11).

In the preamble there should be a specification of two preconditions for the approval of any research protocol, namely:
- Competence of the main researcher and of his group.
- Methodological quality of the study. A badly designed study, without clear objectives or superfluous objectives is not ethical ipso facto, because there is no justification for the submission of the persons to risk when the conclusions will be fallacious, even though the risk is minimal. The inverse is not valid; that is to say, a protocol may be methodologically correct and, however, may have ethical problems. The scientifity does not guarantee the ethicity. (1,11).

C) Considerations about the “General ethical principles”

It is said that the principle about respect for persons“incorporates at least two other fundamental ethical principles, namely: a) autonomy [...] and b) protection of persons with impaired or diminished autonomy [...]”. Actually, it is not about two principles inside another principle. It is one and only principle, autonomy, that some authors call principle of respect for persons. The diminishment of autonomy does not generate a new principle, though it implies a different treatment of the person whose autonomy is absent or compromised (4).
Appreciation about the principle of non - maleficence should be ampled. Not only “avoidable” damages have to be considered, but the “predictable” ones (whether these are or not avoidable). Likewise it should be clarified that the damage caused does not always have to do with “actions”, but often with “omissions”. Moreover, an omission can sometimes bring more damages that an action (4). We believe that this is relevant in case of prospective epidemiological studies. Simple observation, not doing anything else but observing, does not save from damage. The Taskegee case has to be remembered, in which investigators simply observed syphilis’ natural evolution, without any intervention, even when penicillin could inject some hope.
Referring to justice principle, we consider convenient to reinforce the concept of equity in the sample selection: subjects selected from all the human groups that can be potentially benefited have to participate. The idea is not only to share the benefits, but also the risks (9). We think this point is very important and usually it is not taken in account.
When the research takes place in population of emerging countries it has to be considered if what is being researched is or not of interest for such population and, in the case of being so, if the population will have access to the potential benefits. Studies in which, for example, oncological drugs are proved in communities where that pathology is not the main health issue are well known. The conflictivity increases when the study pretends to be randomized. In this case, Helsinki’s Declaration attempts the subjects protection by sustaining that the control group must receive the best treatment so far. Though this does not solve the issue, because which “best treatment” is it talking about? To the one applied currently by the emergent community? Or to the one applied by the central country acting as sponsor? (9). The discussion generated around pharmacological research in pregnant women with aids made in Africa and Asia must be remembered.

D) Considerations about “Ethical principles applied to epidemiology”

Related to informed consent, it is said that it could be omitted when: “the purpose of some studies would be frustrated - for example, prospective subjects on being informed would change the behaviour that it is proposed to study, or might feel needlessly anxious about why they were subjects or study. The investigator will provide assurances that strict safeguards will be maintained to protect confidentiality and that the study is aimed at protecting or advancing health”. Again, the confidence is put on the investigator’s prudence and, at the same time, a paternalist attitude is assumed when it is pretended to justify the hiding of information in order to not disturb the patient. Again, Taskegee case is illustrative here.

We insist on the need of obtaining inform consent in experimental studies and also in prospective ones. In case of retrospective and cross-sectorial studies, when they are based on clinical reports, the patients should be inquired at the time of their confection if they authorize the use of their data with investigative finality in the future. It is necessary to remember that, when the patient goes to a medical treatment, he consents information about himself to be obtained with the purpose of it redounding in his personal benefit. On the contrary, in a research, the main purpose is not to benefit the patient but to obtain information. The data which acquisition was authorized by the patient with one purpose cannot be used for a different one.
We do not consider either the acquisition of consent by the group’s leader as enough condition to make the study. It can be a pre requisite, but never to replace the permission each person has to give. This is considered in the guidelines, but it would be convenient to emphasize it.
The selective handling of information, not divulgating the one that can induce conducts differing the ones we wish to observe, is also a dark point that violates the rule of veracity. We have commented the justification for not respecting this rule when we talked about placebos.

E) Considerations about the “Ethical review procedures”.

We consider that “salaried workers” should be included inside the vulnerable and depending groups, mainly in those countries where there is high unemployment. During pre-occupational and periodical health tests, as in researches made in fabrics, the workers can be forced to consent practices that they would not authorize in other context.
Finally, we understand that investigators must provide the following information:

Conclusions

Viewing the previous observations we offer, as a suggestion, the following conclusions to be introduced as modifications of current guidelines:

Public health understood as population’s health, constitutes the fundamental referent of every effort to improve healthcare, from which it constitutes its most complete manifestation.
Public health is a field of interdisciplinary growth.
The epidemiological method is the most potent and general instrument – although not the only one – to articulate the contribution to public health from multiple disciplines.
One of public health’s essential functions is research, and therefore, epidemiological research.
Possible damages in epidemiological studies are: biological, psychological, socio-cultural, and legal.
Any of them can be important enough as to justify obtaining informed consent.
The authorization for a study by the communal leader may be a necessary condition, but it is not condition enough to make such study. Researchers must obtain permission from each person involved in it.
Deceit is not justified unless: knowledge to be obtained is relevant, it cannot be obtained in other way, the involved subjects are aware of the fact that information given is partial (and consent to be deceived), and that they are not exposed to considerable risks.
The legal context of emerging countries where research sponsored by central countries takes place must be kept in mind. Nevertheless, if the legal context allows the study to take place it does not mean ipso facto that it is a legitimate research. Ethical analysis can never be omitted.
We remark the participation of ethics investigation committees in the analysis of every epidemiological protocols. With the purpose of avoiding interests’ conflicts, we understand that these committees must be independent from researchers, sponsors or institutions where the study takes place. We also remark the importance of the presence of community members in such committees.
The following are pre-conditions for ethical approval of any protocol: aptitude of the main investigator and his group, and the methodological-scientific quality of the study.
In general lines, CIOMS 91 rests on the investigator’s ethical conduct and prudence. But the taking of decisions if, in a particular case, some rule or ethical principle is justified to be violated cannot be left to investigators because they are an interested part.
When data is picked up for assistence medical case histories the patient should be asked if he/she authorizes that such information can be used for scientific researches, leaving a clear register of the authorization or negative.
Salaried workers should be included in the category of vulnerable groups, mainly in countries with high unemployment.
The information that investigators will have to present to the independent committees of ethic in research was expressed in point E).

Bibliography

Lolas, F. Temas de Bioética. Santiago de Chile: Editorial Universitaria; 2002. p. 75-79
Lemus, J D. Epidemiología. Buenos Aires: Editorial Kohan; 1996. p. 197-200
Lemus, J D. Salud Pública; marco conceptual e instrumentos operativos. Buenos Aires: Editorial Kohan; 2001. p. 7-21
Beauchamp, T L.; Childress, J F. Principles of Biomedical Ethics. New York: Oxford University Press; 1994. p. 170-225
Mainetti, J A.. Bioética de la experimentación humana. Experiencia de la Escuela Latinoamericana de Bioética (ELABE). En: Pellegrino Filho, A; Macklin, R, editores. Investigación en Sujetos Humanos: experiencia internacional. Santiago de Chile: OPS/OMS; 1999.
Heler, M. Etica y Ciencia: la responsabilidad del martillo. Buenos Aires: Biblos; 1996. p. 32
Barnes, B. El Problema del Conocimiento, en León Olivé, editores. La Explicación Social del Conocimiento. México: Dirección General de Publicaciones de la Universidad Nacional Autónoma de México; 1985. p. 53-99
Gracia, D. Profesión Médica, Investigación y Justicia Sanitaria. Bogotá: El Búho; 2002. p. 68-76
Lolas, F. Bioética y Medicina. Santiago de Chile: Biblioteca americana; 2002. p. 84-153
Bortz J, Outomuro D, Sanchez N, Kohn Loncarica A. Certificados de conformidad ética. Revista de la AMA 2001; 114(3): 20-24.
Emanuel, E. ¿Qué hace que la investigación clínica sea ética? Siete requisitos. En: Pellegrino Filho, A; Macklin, R, editores. Investigación en Sujetos Humanos: experiencia internacional. Santiago de Chile: OPS/OMS; 1999. p. 33-40
Arboleda – Florez, J. La investigación en sujetos humanos: poblaciones vulnerables. En: Pellegrino Filho, A; Macklin, R, editores. Investigación en Sujetos Humanos: experiencia internacional. Santiago de Chile: OPS/OMS; 1999. p. 90-91

Acknowledgments: We thank Prof. María Fernanda Sabio for critical reading of the manuscript.