THE concern about ethical issues in social science research, unlike say in medical research, is relatively recent, particularly in India. The attention of the medical and scientific community was most dramatically drawn to this issue at the time of the Nuremberg trials, post the Second World War, alerting the world to the callous and inhuman treatment with malefic intentions meted out to prisoners in the name of conducting scientific and medical research in Nazi concentration camps. Subsequently, the international medical and scientific community moved to codify the ethical standards to be followed in medical and scientific research involving human subjects.
Efforts to improve and introduce more refined considerations to maintain the dignity of the subjects involved continues, specifying more rigorously the do’s and don’ts for researchers (Perspectives, EPW 2000). In India too, although with some lag, the Indian Council of Medical Research in 1980 adopted its first code of ethics entitled ‘Policy Statement on Ethical Considerations Involved in Research on Human Subjects.’ The guidelines have since been discussed, revised and formally announced (ICMR 2000).
In the social sciences, however, the scene is not encouraging although ethical issues and dilemmas have surfaced sharply in research dealing with health aspects and efforts to evolve and incorporate the necessary ethical codes by some researchers and research organizations. CEHAT, a Mumbai based organization conducting action and policy research in the health sector, has set up an Institutional Ethics Committee to wet all its research projects. In addition, all ethical issues and dilemmas experienced in the field are brought back to the committee for discussion and, if possible, resolution. Such a comprehensive and informed attempt is unfortunately absent in most organizations involved in health related social science research.
Interestingly, institutions such as the Indian Council of Social Sciences Research (ICSSR) the apex council for social science research in the country, and the University Grants Commission (UGC), responsible for guiding and regulating higher education and academic research in the country, have still to evolve the required policy guidelines. CEHAT was instrumental in initiating and organizing structured discussions around the subject at the national level, widely circulating a draft code of ethics in social sciences and health research. The code has the potential of being accepted by both the wider social science community and apex agencies such as the ICSSR and the UGC.
The evolution of ethical guidelines in the social sciences has followed different trajectories among particular disciplines in different countries. Associations of anthropologists, sociologists, political scientists and psychologists among others, have attempted to introduce ethical guidelines. Economists, for some reason, have not taken any specific initiative in this regard. Efforts are underway in different countries to evolve common ethical guidelines for medical, natural and social science disciplines, emphasizing the commonality in the ethical principles governing research on human subjects and that researchers must recognize and respect the rights of the participants of research.
Most discussion on ethics focus on the actions of the researcher with respect to the participants. Ethical considerations from the point of view of non-participants in research are, therefore, largely ignored. This is inequitable. Where research has practical implications, a more equitable perspective implies that issues of quality take precedence over those of ethics. Such a conclusion, if accepted, favours the use of more rigorous designs, such as experimental approaches, which often create difficult ethical dilemmas when considered solely from the point of view of participants.
For instance, research may be justified even where it risks ‘endangering’ participants than another less appropriate design. Pure research, on the other hand, remains pointless, however ‘ethically’ it appears to be conducted. Good intentions, however, do not guarantee good outcomes (for a detailed treatment of the subject, see Stephen Gorard 2002). Within this framework, the problem as perceived by the researcher becomes more important. The researcher may claim that the research problem is more relevant for an abstract community and hence compromise with ethical considerations.
This article does not intend to further this debate. Instead, it presents a brief review of the ethical considerations governing the all India coordinated research undertaken under the Abortion Assessment Project, India (AAP-I). More specifically, it critically reviews the efforts made to incorporate ethical concerns. Within the perspective of the larger debate, the project attached greater importance to the human subject being researched rather than the research problem. In essence, the AAP-I categorically tried to uphold at least three moral principles of ethics in research – non-malfeasance, beneficence and autonomy.
Abortion is a sensitive issue impacting crucial aspects of women’s health and reproductive rights. In India, medical termination of pregnancy is legal under specified circumstances. However, there are serious problems with both the policy and its implementation. In the last decade, women’s health advocates have tried to draw the attention of policy-makers and administrators to a range of issues related to abortion in order to improve the availability, safety and use of services. The Abortion Assessment Project-India, aimed to undertake a comprehensive review of government policy, availability of funds and the programme environment in the country relating to family planning and abortion care. The project also undertook a survey of public and private facilities in different states, focusing on organization, management, facilities, technology, registration, training, certification and utilization.
Also studied, using qualitative research methodologies, was the user perspective, focusing on women’s perceptions of accessibility, availability and quality of services and confidentiality, consent, post-abortion contraception and the attitudes of service providers. Aspects such as socio-economic and cultural factors, male responsibility, family dynamics and women’s decision-making capacities and processes were also examined in different studies under the project. Community based studies in two states were undertaken for estimating the rate of abortion, resultant morbidity and mortality, and understanding factors leading to spontaneous abortion and reasons for induced abortions. Thus, the research was vast and involved many research subjects. Formal and informal and registered and unregistered service providers, administrators, service users, i.e. women, family members, community members all formed research subjects. Each individual subject and group had to be placed in a context and ethical considerations addressed.
In the context of social science research in reproductive health, unlike biomedical research, the potential risks to research subjects are not usually physiological. Rather, researchers need to be aware of the dangers of exposing the subject to potential psychological, social harm or inconvenience (Campbell et al. 1999, p. 77). The AAP had to ensure that the subjects were not exposed to any psychological stress or social harm. Another concern was that the willingness of subjects to participate should not jeopardize their prospects of future health care. Two other major ethical issues were informed consent and confidentiality and privacy. Finally, ethical issues related to incentives and rewards, public domain, accountability and responsibility were also considered in the project.
In the research study, ethical considerations differed for different subjects. Abortion service providers and administrators were one major category. Under the law, medical termination of pregnancy can be carried out only in registered centres and by certified medical practitioners. In reality a vast number of service providers are neither registered nor certified. Even well-qualified and trained practitioners are often reluctant to get themselves registered given the many complicated bureaucratic blocks and legal and administrative obligations. It is known that a large proportion of the total abortions that take place in the country are performed by unregistered qualified and not fully qualified medical practitioners. The researchers had to deal mostly with unregistered service providers. Therefore, gaining their confidence was crucial since they would not be willing to talk unless the researchers ensured confidentiality. It became necessary not only to gain their confidence but also create a non-threatening environment for inquiry.
A related dilemma with respect to abortion service providers was that researchers had to undertake quality assessment in an invasive manner. The assessment involved entering the labour room, operation theatre and other related facilities to physically verify whether the appropriate standard instruments existed and were being used by the service provider. This was part of the agreed methodology in which a checklist was constructed to verify the availability of equipment and instruments. Personal interviews with the administrator and service providers in each facility was an integral part of the research design. While in the case of facilities in the public domain, the issue was one of obtaining permission from appropriate authorities including the administrator, in the case of private facilities such an assessment implied invading privacy. Informed consent thus became important for this group of subjects.
Service users, i.e. women, were the next most important subjects in this research. Social workers and health activists in the country point out that given the secrecy and often ‘shame’ associated with abortion, women rarely speak freely. However, local women’s groups and health workers do gain the confidence of the community and women may talk about their abortion experience to such persons. Any study on abortion and related issues should, therefore, be designed such that important information is gathered in a gentle and non-judgemental manner.
For any woman, abortion, willing or unwilling, is invariably a traumatic experience. Research, which tries to capture the woman’s perception about the abortion service, forces her to relive the experience. This is a sensitive matter and hence creates ethical dilemmas whether or not to assess and record the experience. However, having established the principle of essentiality, the implied ethical requirement in studies involving the women service users, was to ensure community and local organizations’ partnership. Additionally, the principles of non-exploitation, privacy, anonymity and confidentiality and risk minimization had to be considered and established. Thus, ethical guidelines in the abortion assessment project were developed with deep appreciation of the subjects’ situation and context.
The structure of evolving and maintaining ethical concerns in the AAP-I went through an important change right at the beginning. In the initial stages of the project, ethics related issues were to be managed through a National Ethics Committee. But the Technical Advisory Committee (TAC), after long deliberations, decided that the project management would work, not as a controlling and regulating agent with regard to the management of ethical issues by researchers, but as a guide and facilitator. Consequently, the decision was to constitute a Ethical Consultative Group (ECG) instead of the National Ethics Committee.
In its earlier incarnation, it would have been imperative for the committee to assume responsibility for all the functions of examining, approving, monitoring and evaluating the ethical concerns related to design, protocols and field conduct of the research in the case of each project partner. But not only did the AAP-I cover a large number of projects under the facility surveys, qualitative studies and community-based assessments, the geographical spread was too huge to ensure complete coverage. It was also unclear whether the proposed national committee was adequately equipped to undertake the task.
Finally, it was resolved that a consultative group would be more appropriate to handle ethical issues in the project. The ECG was expected to review all proposals and protocols from the ethics standpoint, provide assistance and training on ethical aspects to IEC of partner institutions and researchers wherever needed, monitor ethical concerns, review of reports submitted by researchers and through field visits as and when required, provide inputs on ethics in project workshops and meetings, address ethical dilemmas faced by researchers and act as a sounding board.
Aworkshop was organized with the project partners to discuss the role of the ECG. It was agreed that the ‘Ethical Guidelines for Social Science Research in Health’ brought out by CEHAT in November 2000 would serve as the standard guideline for AAP-I and its partners. For the management of ethical issues the following guidelines were agreed upon.
* On commencement of the project, all the collaborating institutions will mutually agree on a code of ethics to be followed during the course of the study and in dissemination of study findings.
* Taking informed consent of the respondent by communicating (verbally or in writing as the case may be) the objective of the study, how the information collected will be used and commitment to protect the identity of respondents.
* Protect the identity of the provider/respondent by using code numbers instead of name of the facility surveyed, name of the service provider and identity of the respondent.
* Data generated in this project will be presented only in a tabulated form. Individual case studies/information will be stored carefully, protected and not made public.
* Data collected through this study will be available for use of researchers, women’s health advocates, the government and any other stake-holder. To this end, efforts will be made to ensure the tabulated data and the reports are available with all the collaborating institutions, donors, the government and if possible, on the internet website.
In all studies supported under the AAP-I, ethical review was mandatory. A separate checklist was developed addressing the following issues: formation of an institutional/project ethics committee; identification of social, economic and psychological risks involved; steps taken to mitigate risks, if any; maintaining confidentiality; mode and process of obtaining consent; sampling design and selection of subject; means proposed for voluntary participation; whether remuneration was proposed for participants, if yes, justification and details; methods of data sharing and credit assignments for the research; and sensitizing the field investigators on ethical concerns.
Fortunately, the project did not include any component that might involve bodily harm to the research subject, thus ruling out any physiological risk. Psychological pressure was envisaged only in cases where the studies involved intensive interaction with women, which as stated earlier, would have implied reliving the experience. Since the essentiality of research was established, all project partners were advised to approach the women respondents with full understanding and empathy rather than give an impression of a superior person demanding information. The principle of knowledgeable and professional researcher was also adhered to rather strictly, alongside maintaining the principle of privacy, anonymity and confidentiality. This helped significantly reduce the psychological risk.
There was some discussion on who is best suited to do research on such sensitive subjects. Persons from research institutions without a background of action research would face a quandary if the respondents shared immediate or related health problems and requested for or expected health service. Should research thus be conducted only in areas where some organization was providing health services? The committee resolved that as far as possible research should be taken up only in areas where some partner agency was active and could ensure health related services or proper guidance if some services were sought or expected in the course of the research inquiry.
The most important ethical consideration in health research involving human subjects relates to informed consent; whether the participant/ respondent actually understands what is expected of her and has the freedom to refuse, even after agreeing to participate in the study. How should one go about taking written consent because there are serious ethical questions regarding identity? In the case of tribal populations, the process of taking informed consent is different since they constitute a separate group with different background and socio-cultural practices. This was particularly relevant for Mizoram. Similarly, in rural societies, getting women to sign papers is a near impossible requirement.
The ECG decided that the subjects must be fully informed about the research and what the study expects from them and then seek oral consent if written consent was not possible. A related concern was how to ensure that oral consent was taken because the researcher may simply claim that the consent had been secured. Hence, it was decided that the oral consent should be taken in the presence of another member of the research team. This requirement was also expected to improve compliance. It was further decided that the researcher should ensure that the subject remained free to withdraw at any time.
The ECG deliberated one more important ethical consideration – about incentives, rewards and deception. The discussion was mainly around the issue of subject’s time and opportunity cost. Should the subject be compensated? Or given an a priori incentive? The ECG resolved that the matter be left to the agency conducting research, but a separate question was incorporated in the ethics review in which the agency had to justify if any incentive and rewards were considered. As for to deception or decoy, it was decided that deception may be used with caution while protocols were being pre-tested and sample selections made.
The ECG reviewed the submissions made by all researchers involved in facility surveys, qualitative and community studies and provided feedback on ethical aspects to help improve ethics management in the project. ECG members also attended training workshops conducted by some partners. The ethical dilemmas arising in the field had to be mostly resolved by researchers and the project ethics committee. The researchers were expected to document the entire ethics related experience in the project report.
In the context of the ethical guidelines brought out by the National Committee, Sunita Bandewar has an interesting comment. She says: ‘This neither means that ethical principles were disregarded in earlier research, nor does it mean that all research hereafter will be ethical’ (S. Bandewar 2000). While true as a general statement, in the specific context of AAP-I, the question whether ethics could be maintained needs scrutiny. There are two ways in which one can respond. One, social science research in health is yet to come to grips with the handling of ethical issues in a research project. As noted at the beginning, this need is still not formally reflected in the apex social science research institutions. For instance, most social science research organizations – non-profit and for profit (consultancy organizations) – do not have a institutional ethics committee for ethical review. The IECs are found mostly in organizations conducting only medical research. At least in AAP-I, a beginning was made to foreground the need for institutional arrangements to address ethical concerns. Project ethics committees were formed in all such cases where the IECs did not exist. Thus, an important accomplishment of the project has been that a section of social science researchers and the institutions with which they are affiliated have been sensitized to such concerns.
The second way of responding to the question of maintaining ethical guidelines in the project is to analyze the role actually played by the ECG constituted under the AAP-I, and whether the project/institutional ethics committees performed adequately. In this context, it needs to be admitted that the handling of the ethical aspects in the project was perhaps inadequate, though not because of any lack of understanding about the issues involved in abortion research.
CEHAT and other researchers had rich knowledge and experience about the ethical issues and concerns in abortion research and could agree upon a concise framework and evolve a fairly comprehensive checklist out of this knowledge. Each partner agency also tried to address the ethical concerns by incorporating them into their design and instruments of research. The inadequacy at the AAP-I level can be explained largely because of the inability of the ECG to undertake a comprehensive review and obtain adequate field level feedback.
Nevertheless, all project partners did seek review of the ethical considerations that had been done under the project. Both the facility studies spread across six states and the two community studies obtained informed consent from the research subjects. Three partners obtained written and five oral consent. In the qualitative studies involving women service users and their family members, five partners took oral consent, two obtained written consent from the participants and one partner obtained both oral and written consent.
More serious ethical dilemmas arose in case of facility studies where the service providing doctors had to be interviewed. Privacy could not be maintained in all cases. Most often doctors answered queries from researchers while examining patients. The researcher could choose to revisit, but during each revisit faced the same situation. It was thus practically impossible to interview the doctor in private. In some cases the doctors were unwilling to admit that they provided abortion services. This was mainly in the case of private practitioners who though formally qualified, were not registered. The research inquiry had thus to rely on information from secondary sources and service users.
The qualitative studies involving women subjects gave rise to a different problem. Though the women gave consent and were also convinced about their freedom to withdraw, they were unable to comprehend how and why they were identified for case interviews. They were also apprehensive about the use of information collected. The most common dilemma was that the subjects demanded or expected that the researchers would arrange for health services, such as diagnostic and treatment camps, for their problems. The pure researchers too faced a peculiar dilemma. Women subjects, in a group discussion dealing with the subject, would often respond in full detail. However, when interviewed separately, often denied any personal experience of abortion. This sometimes pushed the researchers into using more aggressive, and thereby intrusive interview techniques.
The ECG was unable to provide adequate support to resolve the dilemmas arising in the field. This was partly because the model of a National Ethics Committee was given up right at the beginning. Under the ECG model, it was agreed to evolve a commonly agreed checklist and undertake a review. The ECG, upon reviewing the submissions made by the researchers for ethical issues under the commonly agreed format, found that there was often a need for detailed information to understand the processes. However, given the time constraints, it was not possible for the ECG to continue with a comprehensive review of the submissions, or attempt a resolution of dilemmas in the field, despite being approached by the researchers. Thus at the project level, the ECG cannot confidently claim that all the reported processes and steps were carried out in each project as required and mutually agreed.
It must also be admitted that the modalities for ethical review by the ECG could have been laid down more precisely at the beginning when protocols were being finalized and detailed proposals along with the ethics checklist were coming in. In fact, it was envisaged that some guidance and review would take place in the training programme for field investigators. However, even such participation was limited. Overall, since the submissions made by researchers did not contain all the necessary details and ECG members could not visit all the locations, the ethics related review remained somewhat incomplete.
This had implications for the role of ECG members at the time of presentation of draft results and discussions in the subsequent workshops. The ECG was in no position to verify whether ethical guidelines were adhered to in actual practice because it did not have complete initial data. Thus, both for the purposes of initial and post implementation review of ethical concerns in the project, the ECG could at best play a limited role.
What are the lessons from this experience? While a National Ethics Committee model may prove impractical because no single committee can realistically guide, oversee and introduce corrections in ethics management of a large project, even the less ambitious ECG model is likely to fail if members are unwilling or unable to commit necessary time.
Sunita Bandewar, 2000. ‘Abortion Research: Ethics in Practice’. Mimeograph, CEHAT, Pune.
Oona Campbell, et al., 1999. Social Science Methods for Research on Reproductive Health, World Health Organization, Geneva.
ICMR, 2000. ‘Ethical Guidelines for Biomedical Research on Human Subjects’. Indian Council of Medical Research, New Delhi.
National Committee, 2000. ‘Ethics in Social Sciences and Health Research’, Economic and Political Weekly, 19-26 February 2000. The final version titled ‘Ethical Guidelines for Social Science Research in Health’, was published by CEHAT, Mumbai, November 2000.
Stephen Gorard, 2002. ‘Ethics and Equity: pursuing the perspective of non-participants’, Social Research Update 39, Department of Sociology, University of Surrey, United Kingdom.