- Atsushi Asai, MD, Mbioeth, PhD.
Associate Professor, Department of Biomedical Ethics, School of Public Health, Kyoto University Graduate School of Medicine,
Kyoto, Konoe-cho, Yoshida, Sakyo-ku, Kyoto, 606-8501, Japan
- Takeo Nakayama, MD, PhD.
Associate Professor, Department of Medical System Informatics, School of Public Health, Kyoto University Graduate School of Medicine, Kyoto
- Mariko Naito, D.D.S., PhD.
Research Fellow, Department of Medical System Informatics, School of Public Health, Kyoto University Graduate School of Medicine, Kyoto
Eubios Journal of Asian and International Bioethics 13 (2003), 147-151.
This study is aimed to address the problems associated with questionnaire-based research. Twelve hypothetical cases are presented and checked for ethical validity. The problems are categorized under seven headings: Whether the participation of the subject is truly voluntary; whether consent to participate in the study has been obtained or proxy consent is required; whether consent to participate has been obtained from "authentic" participants and their privacy is protected; whether participants are fully informed before they consent to participate; whether the validity of questionnaire-based research has been determined in terms of its possible harm; how the roles of the research ethics committee are defined; and what content should be presented in reports and articles. It is important that the ethics of questionnaire-based research should be reviewed with consideration to maintaining a balance between the protection of a subject's human rights and the promotion of research for the improvement of medical science and health care.
Keywords: Questionnaire, ethics, research, human rights, harm, ethics committee
Ethical standards for medical research involving humans have been established based on the ethical consideration of wartime atrocities committed by the Nazis, a historical article published by Henry Beecher, and criticism of many unethical medical studies conducted on humans over the past 50 years [1-5]. Although it seems some disagreements remain, basic principles regarding due process and procedure for implementation of medical research do exist today. The accumulated information and knowledge in this regard are summarized, disclosed, published, and used in the form of ethical principles and guidelines for research [1, 6].
There are also ethical guidelines for epidemiological research [7-10]. The guidelines are in principle based on the same ethical principles that require in medical research involving humans. However, according to these guidelines, epidemiological research using existing medical information or subject data can be implemented without informed consent if a subject's anonymity is guaranteed and, at the same time, the research ethics committee has exempted the researchers from their obligation to obtain prior informed consent from the subjects based on its judgment that no harm of any kind would be inflicted upon them as a result of the research and that their privacy would be fully protected [7-10]. This view is probably prevalent among medical care providers and medical researchers throughout the world. However, some argue that epidemiological research without informed consent cannot be justified solely because it has been approved by a research ethics committee and because a subject's right to privacy is fully protected and there is no possible physical harm to the subjects [11-13]. In addition, the ethical validity of the research cannot be judged solely by the degree of physical harm to the participants. Wrongdoing in research or unjust treatment of subjects is a controversial ethical issue as well .
In recent years, besides the ethical issues concerning medical and epidemiological research involving humans, more subtle issues have come to our attention. One is the theme of this study, namely, the ethical validity of questionnaire-based research. For example, at the end of 2000, a researcher in Japan conducted a questionnaire survey supported by a municipality. Clear informed consent had been obtained from the respondents. However, questions arose as to whether the items in the questionnaire were appropriate (whether or not certain questions could have violated a respondent's privacy) and whether or not the obtained informed consent came from a respondent voluntarily (the possible influence of municipal support to the research) . Another example of such a controversial questionnaire was one conducted on subjects aged eight to 18 to identify the relationship between suicide, family relationships, depression, and childcare. One of the questions asked the respondents to comment on the statement "I am not thinking about committing suicide." It was suggested that asking such a straight question about suicide to underage subjects could be invasive and harmful . It can be argued that the likely impact of questionnaires upon research participants is not often considered and therefore, the balance of benefit and harm not fully explored .
Under these circumstances, this paper addresses ethical issues in questionnaire-based research. A questionnaire is a common research method in the field of medicine, especially in the disciplines of medical ethics, quality of life (QOL), epidemiology, and social psychology. The success of a research project depends on the trust and cooperation that researchers gain from potential participants, such as the general public and those who are receiving medical care. Therefore, in order to implement efficient and ethically appropriate questionnaire-based research, it is essential to establish ethical standards that would govern it. In this study, we categorized the ethical issues in this regard into seven headings as follows:
1) Whether the participation of the subject is truly voluntary; 2) Whether consent to participate in the study has been obtained or proxy consent is required; 3) Whether consent to participate has been obtained from "authentic" participants and their privacy is protected; 4) Whether participants are fully informed before they consent to participate; 5) Whether the validity of questionnaire-based research has been determined in terms of its possible harm to the participants; 6) How the roles of the research ethics committee are defined; and 7) What content should be presented in reports and articles.
Because of many diverse issues identified, it would be difficult to address each of the various issues identified in this study using normative ethics. Therefore, the aim of this paper is basically to present the problems.
2. A Study of the Ethical Issues in Hypothetical Cases
In order to consider the ethical validity of questionnaire-based research, several hypothetical cases are presented as research models. All of these cases have been created by the authors and are not based on any particular existing studies.
2-1 Whether the participation of the subject is truly voluntary
Case 1: In research to verify the effects of a high school antismoking education, an anonymous self-administered questionnaire was used on students who had received antismoking advice. The first questionnaire survey was conducted during the antismoking class with support from the teaching staff.
Case 2: In order to identify the students' eating habits, an anonymous self-administered questionnaire was conducted among students aged 10 to 12. Classroom teachers gave the questionnaires to the students at the school. The questionnaire included questions regarding smoking and the drinking of alcohol.
Case 3: As a part of activities in a classroom or seminar, instructors used a questionnaire to obtain information about students' perceptions of terminal care. The results were later reported.
Case 4: A researcher studying smoking and sexual behavior of medical students and graduates of a medical school administered a questionnaire to the students and new medical staff members. The questionnaires were handed to the students and staff by their instructors and supervisors, respectively. Procedures were followed to ensure anonymity of the responses; however, the respondents' supervisors were able to guess who had or had not answered the questions.
Did all of the respondents in these four cases participate in the study voluntarily? Do potential respondents truly have the freedom to choose not to participate when the researchers in charge of the survey are their instructors or supervisors? It would be even more difficult to choose not to participate if the respondents' supervisors could find out who did or did not respond to the questionnaire. Similar problems are possible when doctors ask their subjects within the medical facilities to participate in a questionnaire study. When doctors or teachers ask their patients or students to participate in research, the subjects may feel compelled to do so, even when the doctors or teachers are not intentionally putting pressure on the patient or student.
The place in which the survey is conducted is equally important. For example, if a request for participation is made (or instruction about the survey is given) in the classroom, students may feel compelled to participate. Furthermore, if research is supported by a certain public institution or municipality, it is important to consider carefully whether such support might unduly influence the decision of a potential participant to participate in it. On the other hand, it can be argued that publicly supported research could indicate reliable research procedures and benefits for the public interest, which may be reassuring to the participants. It is fair to say that researchers should try to establish continuous communication not only with the municipal officials who support the research but also with individuals or groups in the community. However, many may argue against the ethical validity of the "community agreement," which is obtained from the representatives of a municipality or community residents as an approval of research. Interpretation of the above-mentioned issues may differ from person to person or culture to culture. However, when authorities other than researchers are involved in the preparation and implementation stages of research, the positive and/or negative impact of their participation should be considered.
When the voluntariness of the participants is not ensured, the authenticity of their answers cannot be guaranteed. For example, asking minors questions related to illegal or immoral acts, such as smoking or socially undesirable matters, is not likely to produce reliable responses. Furthermore, when the obtained responses suggest an illegal act, what would be the ethically appropriate thing for a researcher to do? For example, what ethical obligations would researchers have if responses obtained from junior and senior high school students revealed that they smoked or used illegal drugs?
These issues demonstrate how a subject should be asked to participate in a research project, how a subject should be approached, why the selection of an appropriate place to respond to a questionnaire makes a difference, and why voluntary participation should be guaranteed, particularly for those who are potentially vulnerable. Would it be considered an undue inducement to reward participants for their cooperation, especially when their income is low? What is an appropriate sum for such a reward? Would a reward of 1,000 yen (approximately 8 US dollars) per participant for his/her cooperation in the research, such as filling out a questionnaire, be appropriate?
2-2 Whether consent to participate in the study has been obtained or proxy consent is required.
Case 2: As stated above.
Case 5: A clinician administered a questionnaire to schizophrenic outpatients and inpatients at a hospital where he worked. The objective of the study was to know what kind of illness the patients think they have and to obtain an explanation of what they had learned from their doctors and what they wanted to know. Oral consent was obtained from the patients.
In these cases, the question is whether the subjects truly understood the content of the request to participate in the study before they responded to the questions. Especially in Case 5, the subjects had been diagnosed as mentally ill by various diagnostic standards, and it is doubtful that they fully understood the objective of the study. Even though being mentally ill does not necessarily indicate a lack of decision-making capability, it is difficult to tell whether the patients could really understand the abstract objective of the study (the ethical validity of telling the truth) or whether they understood that the questionnaire was research and not a part of their therapy. These can be serious issues with patients who are mentally retarded or demented.
Another question is whether or not proxy consent is required when the respondents of questionnaire-based research are schoolchildren or when their understanding is impaired due to illness. If proxy consent is required, who is going to give consent to the researcher? For example, even if implementation of a questionnaire survey to elementary school students (children under 12) is approved by a classroom teacher or a schoolmaster, would it not be necessary to obtain consent from the respondents themselves or their parents as well? When a questionnaire includes questions related to privacy or illegal acts, such as smoking, drug use, and prostitution, as addressed in 2-5, the consent should be given after careful deliberation. Some critics claim that individual or proxy consent should be obtained before a questionnaire survey is implemented. In such cases, it should be determined whether oral or written consent is necessary. On the other hand, other might critics claim that the act of responding to a questionnaire can be regarded as a sign of consent in itself.
2-3 Issues arising when questionnaires are not completed by the "authentic" subjects
Case 6: A team of researchers asked the families (the caregivers) of patients with senile dementia to fill out a questionnaire in order to identify the relationship between the caregivers' psychological stress and the patients' condition. The questionnaire asked the caregivers to give detailed information about the medical, mental, and social states of the patients with senile dementia. Half of the patients who needed care were capable of communicating with others.
Issues related to the requirement of privacy and consent may arise when the main content of a questionnaire requires a respondent to describe someone other than himself or herself, such as a caregiver describing a patient in the above case. Similar issues could arise when health care providers conduct surveys that profile hospitalized patients or when health care providers use information from a patient's medical record to complete a questionnaire. The current guidelines for epidemiological research allow researchers to access patient records as long as patient anonymity and privacy are protected .
When health care providers, as research participants, are asked to describe their experiences related to patient care or specific cases of ethical dilemmas encountered in the process of patient care, it should be considered whether patient's consent is required. One way to solve this problem is to disguise the sex, age, and illness of a patient to make him unidentifiable. When health care providers are asked questions related to any form of patient information, including that which must be obtained from patient records, the ethical validity of such research depends on the nature of the questions; issues related to privacy and consent requirement are influenced by how detailed the provided information is. For example, it would probably be safe to ask caregivers questions related to mental stress or psychological pressure they experience. However, care should be exercised when asking caregivers questions related to specific problem behaviors of elderly patients in their care because such questions could identify the person in question, depending on the community they belong to and the nature of the question. When collecting information of any kind, it is necessary to confirm whose consent should be obtained.
2-4 Information provided to participants prior to obtaining their informed consent to participate in the research
Case 3: As stated above.
Case 7: Patients were asked to fill out a questionnaire to identify the effects of the therapeutic intervention that they received. The survey was conducted only on those who consented to participate in the research. However, it was unclear what kind of prior explanation was given to them when they were asked to participate.
Researchers should provide subjects with sufficient information prior to obtaining their consent to participate in a study . When a questionnaire is administered in a medical setting (during a medical examination or checkup) or in an educational setting, how explicitly should the prior explanation indicate that the survey is part of the treatment, educational activities, or research? It is clinically important to conduct a survey to measure the effect of therapeutic intervention as shown in Case 7, but, if there is a chance that such a survey could be used in a future study, it must be determined if the respondents should be informed of such a possibility.
On the other hand, when obtaining a subject's informed consent to participate in a study, under what circumstances are researchers allowed to intentionally omit the objective and nature of the research from their prior explanation to the subject? Would it be ethically acceptable not to explain the objectives of the research in order to avoid bias in the responses? In order to answer these questions, we must discuss the kind of ethical standards that should be used to judge the ethical validity of any research that makes use of questionnaires. Would it be possible to apply the long-established current research ethics widely used in medical research involving human subjects or should a separate standard be established? Furthermore, what should be included in the prior explanation to subjects? When responses to a questionnaire are linked with other patient information, such as patient records and test data, how much information about such facts should be included in a prior explanation? If a patient's safety, anonymity, and privacy are protected, would it be unnecessary to obtain a patient's prior consent to access his records and data?
Information that should be disclosed and explained to a subject before his participation in research is usually presented in written form, such as in a letter of request to participate in research or in a "read-before-you-start" section on the front page or within the first few pages of a questionnaire. For example, would it be necessary to add a proviso that respondents do not have to respond to questions that they do not want to answer, or would it be acceptable to omit such a proviso and permit participants to decide on their free will and regard their responses as their consent to respond? Some researchers may intentionally avoid indicating such a proviso for fear of raising the number of missing values. If the validity of the survey results is damaged because of an excessive number of missing values, survey expenditures, as well as the efforts of respondents who provided valid responses, will have been wasted. However, it can be argued that the possibility of excessive missing values alone cannot justify omitting the above proviso from the prior explanation.
The prior explanation which should be included could be the following matters: objectives and procedures of the research; subject's freedom to participate in or to withdraw from the research; names of the researchers; funding sources; sponsors of the research; actual process and procedure of the research, including links to other information; acknowledgements; the proviso that respondents do not have to respond to the questions that they do not want to answer; methods used to retain participant responses; information regarding those who have access to the responses; how the responses are going to be used in the future; consideration to privacy; methods of feedback regarding the results obtained; and whether or not participants must state their names in the questionnaire. Different people have different opinions in this regard.
2-5 Contents of a questionnaire
Case 5: As stated above.
Case 8: A researcher administered a questionnaire to elementary school students in the lower classes regarding their lifestyle and health. The questionnaires were first handed to the schoolmaster, then to the health instructor, and finally to the classroom teacher, who asked the students questions related to unbalanced diet and bowel movement.
Case 9: The following questions were asked during a survey administered to several types of patients in different hospitals to measure their subjective quality of life: "Do you think you are worthless?" "Do you have any hopes for your future?" "Are you grateful that you were born?" "Do you like yourself the way you are now?" "Do you have something to live for?"
Health care providers routinely ask their patients about various matters as part of their medical tasks. In questionnaire-based research, many sensitive questions are asked, such as those related to highly private matters or those that are extremely unpleasant, some of which could be harmful to the subjects. However, the definition of sensitivity differs from person to person. Individuals have quite different opinions about the kinds of questions that cause unpleasant or disturbing feelings. For example, questions, such as those asked in Case 9 could have a harmful effect on some respondents. It is important to determine which matters are acceptable and which are not as contents for a questionnaire so that problems can be readily identified and solved.
Is there such a thing as a taboo question? Would such a question be ethically valid if a respondent freely consented to answer it? For example, questions related to social and economic matters, such as a participant's social status or income, are considered sensitive, but such questions could be beneficial in discovering hidden major problems or in establishing social-scale countermeasures. There are quite a lot of epidemiological evidence that the determinants of people's health are strongly related to social and economic factors, indicating the health inequalities [18,19]. For example, in the UK, cancer morbidity and prognosis obtained from a cancer registry are known to be closely linked with the socio-economic status of the people . The government has decided to spend a large amount of money to promote a comprehensive cancer control program [21, 22]. Furthermore, in "Our Healthier Nation," England's national project to improve public health, many measurable objectives have been set considering such socio-economic matters and to reduce the health gaps among classes (http://www.doh.gov.uk/ohn/inequalities.htm). These examples show that, despite its controversial nature, sensitive questions have great advantages as well.
The following matters could be potentially sensitive subjects for questions, according to the authors: body weight, history of contracting sexually transmitted diseases, general clinical history, excessive drinking of alcohol, employment status, sexual experience, history of being sexually assaulted, educational level, income level, habits, history of engaging in illegal acts (such as, a minor smoking or a person's involvement in euthanasia), home environment, economy, income, family suicide history, sexual preferences, history of abortion, parity, suicidal thoughts, feelings of hatred, meaning of life, feeling of loneliness, history of abusing the family or being abused by the family, intimacy within the family, psychological conflicts among the family, religious beliefs, eating behavior, family make-up, psychological stress, physical condition, and performance of immoral acts.
Many others issues remain to be solved, such as who can or should judge the sensitivity of the contents of a questionnaire, what kinds of procedures are required to make such a judgment, what kinds of research would justify asking questions that are potentially harmful to a respondent, and how the ethical validity of such questions can be determined.
2-6 Roles and limitations of a research ethics committee
Case 10: A group of researchers asked the research ethics committee for approval to administer a questionnaire to patients under medical care in order to study the degree of their trust and distrust in medical care and the relationship between health care providers and patients. Two months later, the committee granted approval, although some issues related to the methods of recruiting and rewarding participants were addressed by the reviewers. One of the reviewers pointed out that approval by the ethics committee might not be necessary for this research because the participant's anonymity was guaranteed, rewards were prepared, and participation was based on a subject's free will.
In our experience, approval by a research ethics committee is hardly ever mentioned in published papers of questionnaire-based research, with the exception of a small number of studies in a few countries. Therefore, it is unclear whether or not some studies have received approval. Considering this situation, we should first discuss whether a research ethics committee should review questionnaire-based research, and, if they should, what aspects of such research should be reviewed? Do we need to follow the same procedure for review that is used in other medical research involving human subjects? What criteria are to be used to judge the ethical validity of such research? Is a retrospective review required when the questionnaire is conducted as part of a medical task or treatment or an educational activity? If a committee split as it tried to determine which questions were too sensitive, as mentioned earlier, what procedures could be followed to determine the validity of such questions?
2-7 Matters to be included in a report and publication of the results
Case 7: As stated above.
Case 11: A researcher administered a questionnaire to more than 400 homosexuals to study the relationship between venereal disease and a patient's sexual preference. The method used to identify and select the subjects was not mentioned in the paper.
Case 12: A researcher of bioethics have conducted many questionnaire-based research and published several papers based on the results of these research. Although all the surveys were conducted anonymously, this fact was not specified in the "Methods" section of the papers. Furthermore, the papers did not disclose that participants had been paid a certain sum to fill out the questionnaire.
When reporting the research or publishing it in the form of a paper, what information should be given in the "Methods," "Subjects," and "Questionnaire" sections? This is one of the most critical questions that need to be answered for the appropriate publication of research using questionnaire surveys. How detailed should the following information concerning the ethical consideration be given to the research, whether a reward was given to the respondents, whether or not it was an anonymous survey, and whether the contents of the prior explanation were printed on the front page of the questionnaire (or in the letter of request to participate in the survey)? Should the report or paper indicate that the particular study has been approved by the research ethics committee?
For example, in Case 11, no information of how the subjects were approached is given. It could have been reported in the original paper but omitted by the editor of the journal due to space limitations. However, it might not have been in the original paper to begin with. Issues like these are also related to a journal's publication policies and ethical standards. It is very difficult to judge the ethical validity of research involving questionnaires based on the information presented in published articles alone.
In this paper, we addressed and categorized potential problems in the implementation of questionnaire-based research. We posed questions on many of the conventional practices prevalent among researchers, including ourselves. Addressing issues in such a manner can be criticized as captious. On the other hand, it can be argued that our ethical awareness has improved through studies and reviews of the history of unethical medical research, which enables us to highlight problems that had been overlooked for a long time.
We addressed the problems at the following three stages of research: examination of validity; implementation; and reporting and publishing. Most of the problems addressed in this paper are subtle issues in terms of ethical validity. Some may claim that they are minor issues because questionnaire-based research could seldom be directly harmful to the participants. However, in our opinion, many of the issues addressed in this paper are related to the essence of research ethics, such as the protection of the subjects' voluntariness and privacy, whether or not the subjects' consent to participate in the study has been obtained, and whether the due process of ethical review has been followed. In order to discuss such issues, we must first provide answers to the following questions: How do we define harm? How far should we respect an individual's right to control the circulation of information relating to him? Which parts of the research procedure should be included in the report? How can we seek a balance between the progress of medical science and the protection of the rights of individuals?
Regardless of many unresolved problems regarding the ethics of questionnaire-based research, we would argue that, as a minimal requirement, the protocols of questionnaire-based research should be reviewed by the research ethics committee or IRB, which is a standard procedure now in some countries, or its equivalent such as an independent third-party organization that has no conflict of interests with the research in question. This is because medical professionals have ulterior professional and personal interest in successfully conducting medical research such as acquiring a reputation and receiving promotion. Such ulterior interests may disturb the researcher's ability to judge the significance of the research and the risk-benefit ratio involved (unpublished observations). However, we must be aware that such review systems have limitations. Agreement by an ethics committee does not necessarily guarantee the ethical validity of the research. Furthermore, an ethics committee would splits over subtle issues. In addition, the significance of a research ethics committee is lost when researchers submit a false report. For example, there was a case in which a researcher submitted a false report to the ethics committee and then conducted genetic analysis on tissue collected from subjects without obtaining their prior consent . Therefore, a formal review system must be accompanied by educating researchers to raise their ethical awareness . It is not known what kinds of research would be considered appropriate by researchers with high ethical sensibility. However, it would be fair to say that researchers who consider it important to think ethically and who always determine their actions based on ethical considerations are more likely to conduct more harmless research than those who give highest priority to the medical and scientific aspects of the research and the profits that come from the results.
Although the issues addressed in this paper are subtle, they are important ethical issues that require careful consideration and comprehensive discussion. Therefore, in this paper, we limited our discussion to presenting the issues. We hope that this paper becomes the basis for discussing the ideal direction for questionnaire-based research. Our conclusion for the present in this paper is that we must have a true understanding of research ethics developed in the past as well as nurture our own ethical sensibility to promote the appropriate use of questionnaires in research. Further consideration will be required on what actually needs to be done for it. Each researcher in charge of this type of research must carefully consider its ethical validity while keeping the balance between progress in medical science and health care and the protection of the subject's human rights. Moreover, not to mention, the purpose of the research should be accountable and socially acceptable and should serve the needs of the society. Research that is heavily profit-oriented or is conducted out of simple curiosity will probably have difficulties meeting the ethical requirements of society. However, it can also be said that excessive "ethical" restriction could deter researchers from research that could potentially be beneficial to society in the long run.
The use of personal information for the sake of public interest and the protection of subject's human rights must be balanced and this can only be done by the collaborative effort of society as a whole. Therefore, more opportunities should be provided for ordinary citizens, researchers, ethicists, and experts from other fields to get together and exchange opinions about this topic.
The authors would like to thank Professor Akira Akabayashi for his comments on this article. This research was supported in part by Grant-in-Aid for Scientific Research by the Ministry of Health, Welfare, and Labor (Quality of life Research Group, Principle Investigator: Shunichi Fukuhara, MD, MSc, DMSc.)
1 Barnbaum DR, Byron M. Research Ethics, New Jersey, Prentice Hall, 2000: 1-14.
2 Beecher H. Ethics and Clinical Research. N Eng J Med 1966; 274: 1354-60.
3 McNeiil PM. Experimentation on Human Beings, In: Kuhse H and Singer H, ed. A Companion to Bioethics, Oxford, Blackwell publishers, 1998: 369-378.
4 Tsuchiya T. Why Japanese Doctors Performed Human Experiments in China 1933-1945? Eubios J Asian and Int Bioethics 2000; 10: 179-180.
5 Appelbaum PS. Drug-Free Research in Schizophrenia: An Overview of the Controversy. IRB 1996; 18: 1-5.
6 World Medical Association. Declaration of Helsinki. Ethical Principles for Medical Research involving Human Subjects, World Medical Association, October 2000.
7 Royal College of Physicians Committee on Ethical Issues in Medicine. Research based on archived information and samples. Recommendations from the Royal College of Physicians Committee on ethical issues in medicine. J R Coll Physicians Lond 1999, 33: 264-6
8 Beauchamp TL, Cook RR, Fayerweather WE, Raabe GK, Thar WE, Cowles SR, GH Spivey GH. Ethical guidelines for epidemiologists. J Clin Epidemiology 1991, 44: 151S-169S
9 Council for International Organizations of Medical Science (CIOMS). International
guidelines for ethical reviews of epidemiological studies, Geneva 1991.
10 Kouseisho Kagakukenkyuuhi Hojokin. Guidelines for informed consent in epidemiological research, version1.0, April 10, 2000, Tokyo (In Japanese).
11 Asai A, Ohnishi M, Nishigaki E, Sekinoto M, Fukuraha S, Fukui T. Attitudes of the Japanese public and doctors towards use of archived information and samples without informed consent: Preliminary findings based on focus group interviews. BMC Med ethics 2002: 3; 1 (http://www.biomedcentral.com/1472-6939/).
12 Asai A, Ohnishi M. Ethical consideration in epidemiological studies. Journal of Japan Association of Bioethics 2001; 1: 122-8 (In Japanese)
13 Hamajima N, Tajima K. Patient's views on reference to clinical data. J Epidemiology 1997; 6: 17-19.
14 Capron AM. Protection of research subjects: So special rules apply in epidemiology? J Clin epidemiology 1991, 44: S81-S89.
15 Asahi Shinbun (newspaper), November 22, 2000 (In Japanese)
16 Anonymous. Psychological research involving children, Jottings, Monash Bioethics Review 2000; 19: 75-6.
17 Evans M, Robling M, Maggs Rapport F, Houston H, Kinnersley P, Wilikinson C. It doesn't cost anything just to ask, does it? the ethics of questionnaire-based research. Journal of Medical Ethics 2002; 28: 41-44.
18 Berkman L, Kawachi I (ed). Social Epidemiology. Oxford, Oxford University Press, 2000.
19 Marmot M, Wilkinson R (ed). Social Determinants of Health. Oxford, Oxford University Press, 1999.
20 Coleman MP, Babb P, Sloggett A, Quinn M, De Stavola B. Socioeconomic inequalities in cancer survival in England and Wales. Cancer 2001; 91: S208-S216.
21 Steimle S. UK's Tony Blair announces crusade to fight cancer. J Natl Cancer Inst 1999; 91:1184-5.
22 Anonymous, Britain Upgrades Its Cancer Program. J Natl Cancer Inst 1999; 91:1999.
23 Analyzing patients' genes without informed consent and presenting a false report to a local research ethics committee. Asahi Shinbun (Newspaper) 2001, March 28 (In Japanese)