Diretrizes Internacionais para a Pesquisa em Epidemiologia CIOMS - 2008

International Ethical Guidelines for Epidemiological Studies Prepared by the Council for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health Organization (WHO) CIOMS Geneva February 2008 (Provisional text - pending printed version) 2 CONTENTS ACKNOWLEDGEMENTS BACKGROUND INTRODUCTION INTERNATIONAL INSTRUMENTS AND GUIDELINES GENERAL ETHICAL PRINCIPLES THE GUIDELINES 1. Ethical justification and scientific validity of epidemiological research involving human subjects 2. Ethical review committees 3. Ethical review of externally sponsored research 4. Individual informed consent 5.

Obtaining informed consent: Essential information for prospective research subjects 6. Obtaining informed consent: Obligations of investigators and sponsors 7. Compensation for participation 8. Benefits, harms and risks of study participation 9. Special limitations on risk when research involves individuals who are not capable of giving informed consent 10. Research in populations and communities with limited resources 11. Choice of control in clinical trials 12. Equitable distribution of burdens and benefits in the selection of groups of subjects in research 3 13. Research involving vulnerable persons 14.

Research involving children 15. Research involving individuals who by reason of mental or behavioural disorders are not capable of giving adequately informed consent 16. Women as research participants 17. Pregnant women as research participants 18. Safeguarding confidentiality 19. Right of injured subjects to treatment and compensation 20. Strengthening capacity for ethical and scientific review and biomedical research 21. Ethical obligation of external sponsors to provide health-care services 22. Disclosure and review of potential conflicts of interest 23. Use of the Internet in epidemiological studies 24.

Use of stored biological samples and related data Appendix 1: Glossary Appendix 2: Items to be included in a protocol (or associated documents) for epidemiological research involving human subjects Appendix 3: The Declaration of Helsinki Appendix 4: Members of the Core Group Appendix 5: Commenters on the draft guidelines (Organizations and individuals) (to be provided at a later date) 1 ACKNOWLEDGEMENTS 1 2 The Council for International Organizations of Medical Sciences (CIOMS) 3 acknowledges the financial contributions of UNAIDS, the WHO Department of 4 Reproductive Health and Research, the Swiss Academy of Medical Sciences, the 5 Royal Netherlands Academy of Arts and Sciences, the Drug Information Association 6 (DIA) and the Exxon Mobile Corporation, Irving, Texas, USA.

CIOMS was at all 7 times free to avail itself of the services and facilities of WHO, for which it is grateful. 8 A number of other institutions and organizations also made valuable contributions by 9 making their experts available at no cost to CIOMS for the meetings held in relation 10 to the revision project. This too has been highly appreciated. 11 The updating process, which began in 2003, was initiated by the CIOMS 12 Secretariat by requesting a number of experts and institutions involved in the 13 formulation of the 1991 CIOMS International Guidelines for Ethical Review of 14 Epidemiological Studies to list new and additional topics to be covered in the updated 15 version.

Subsequently, drafting was carried out initially by an established Core Group 16 of experts listed in Appendix 4, supported by Professor Juhana E. Idänpään-Heikkilä, 17 then CIOMS Secretary-General, and Mr. Sev Fluss, Senior Adviser of CIOMS and 18 chaired by Professor Michel Vallotton, Swiss Academy of Medical Sciences and 19 President of CIOMS. The task of finalizing the various drafts was led by Professor 20 Alexander Capron, Professor of Law and Medicine, University of Southern 21 California, Los Angeles, California, USA as Principal Rapporteur, assisted by 22 Professor Rodolfo Saracci, Director of Research in Epidemiology, National Research 23 Council, Pisa, Italy and Professor Idänpään-Heikkilä and the CIOMS Secretariat.

A 24 consultation was held on the draft Guidelines in June 2007; the participants are listed 25 in Appendix 5. The interest and comments of the many organizations and individuals 26 who responded to the drafts of the Guidelines posted until now on the CIOMS website 27 or otherwise made available are gratefully acknowledged (Appendix 5). A further 28 meeting of the Core Group was held in November 2007 to incorporate the suggestions 29 of the consultants and other comments on the pre-final draft to be posted again in 30 February 2008 on the CIOMS website. 31 32 33 2 BACKGROUND 34 35 The Council for International Organizations of Medical Sciences (CIOMS) is an 36 international non-governmental organization in official relations with the World 37 Health Organization (WHO).

It was founded under the auspices of WHO and the 38 United Nations Educational, Scientific and Cultural and Organization (UNESCO) in 39 1949; among its mandates is to maintain collaborative relations between national and 40 international medical and scientific groups and the United Nations and its specialized 41 agencies, particularly UNESCO and WHO. 42 In the late 1970s, CIOMS began working in collaboration with WHO on ethics 43 in relation to research. The initial objective was to prepare guidelines to indicate how 44 the ethical principles that should govern the conduct of biomedical research involving 45 human subjects, as set forth in the Declaration of Helsinki (first issued by the World 46 Medical Association in 1964 and amended in 1975), could be effectively applied, 47 particularly in developing countries, given their socioeconomic circumstances, laws 48 and regulations, and executive and administrative arrangements.

The first product of 49 this CIOMS/WHO undertaking was the publication in 1982 of Proposed International 50 Ethical Guidelines for Biomedical Research Involving Human Subjects. 51 The period that followed saw rapid advances in medicine and biotechnology, the 52 growth of multinational clinical trials and of research involving children and other 53 vulnerable groups, a shift in attitudes towards regarding human subjects research as 54 largely beneficial rather than threatening, and the outbreak of the HIV/AIDS 55 pandemic. These developments raised new ethical issues not considered in the 56 preparation of Proposed Guidelines.

Moreover, the Declaration of Helsinki was again 57 twice revised (in 1983 and 1989). It therefore was timely to revise and update the 58 1982 guidelines, and CIOMS, with the cooperation of WHO and its Global 59 Programme on AIDS, in 1993 issued International Ethical Guidelines for Biomedical 60 Research Involving Human Subjects. 61 During this period, CIOMS and its collaborators also recognized that ethical 62 guidance was also needed for public health research. Therefore, even before the 63 revision of the biomedical research guidelines was completed, International 64 Guidelines for Ethical Review of Epidemiological Studies were published in 1991.

65 3 In the years that followed, it became apparent that the biomedical guidelines 66 would need to be revised again to address additional issues, especially those arising in 67 controlled clinical trials carried out in low-resource countries by sponsors from richer 68 countries. The updating, which was also necessitated by a major revision in the 69 Declaration of Helsinki in 2000, resulted in the publication by CIOMS and WHO of 70 revised International Ethical Guidelines for Biomedical Research Involving Human 71 Subjects in 2002. Like the 1982 and 1993 versions, the new document was designed 72 to be of use, particularly to low-resource countries, in defining national policies on the 73 ethics of biomedical research (and particularly clinical trials of pharmaceuticals), 74 applying ethical standards in local circumstances, and establishing or redefining 75 adequate mechanisms for ethical review of research involving human subjects.

76 The process of revising the research guidelines that began in the late 1990s 77 made clear that developments in the ethical analysis of all types of research using 78 human subjects had potential implications for the 1991 Guidelines for 79 epidemiological studies. Furthermore, the growing recognition of the importance of 80 epidemiological research to improving the health of the public highlighted the 81 importance of bringing the 1991 guidelines into line with current thinking on ethics 82 and human rights. Therefore, in 2003 CIOMS constituted a core group to consider 83 how the existing ethical guidance for epidemiological studies should be updated.

The 84 group initially attempted to make changes in the 1991 document, but then put aside 85 that draft because of the response from persons involved in ethical review committees 86 that they found it difficult to relate the epidemiological guidelines to CIOMS 2002 87 International Ethical Guidelines for Biomedical Research. The group recognized 88 how widely the latter document has been disseminated, so that it is now the basic 89 conceptual and practical guide when research undergoes ethical review in institutions 90 around the world, particularly in developing countries. Intending to ensure that 91 ethical principles are consistently applied to all types of research, the core group 92 decided to prepare a Supplement to the 2002 document that would address the special 93 features of epidemiological studies.

The group thereby meant to connect the ethics of 94 epidemiological research with the standards and analysis that have been developed for 95 other types of research involving human subjects and to ease the process of review 96 because many of those using the proposed supplement–especially members of, and 97 4 administrators for, ethical review committees–would have experience with using the 98 2002 Guidelines in the context of biomedical research projects. 99 In February 2006, a draft of the supplement was posted on the CIOMS website 100 and opened to comment from interested parties. The response from groups and 101 individuals involved in biomedical research was largely positive, but the same was 102 not true among epidemiologists.

Some accepted the drafters‟ insistence that the 103 manner of presentation did not signify that the field of epidemiology should be 104 regarded as derivative from or secondary to clinical research. (Quite the contrary, as 105 the drafters noted: the field of epidemiology predates many of the methods now used 106 in clinical research.) But many objected that epidemiologists were not necessarily 107 conversant with the 2002 Guidelines and would therefore find it burdensome to have 108 to switch back and forth between the epidemiology supplement and the biomedical 109 research document. They were also concerned that the supplement would not 110 provoke ethical review committees that principally review biomedical research to 111 sufficiently adjust their expectations–and also their membership–to take account of 112 important differences raised by epidemiological studies.

At the same time, the critics 113 also stated that ethical review committees that mostly review public health research 114 and other epidemiological studies would find it simpler to have a stand-alone set of 115 guidelines. 116 A second issue, concerning the scope of the new guidance document for 117 epidemiological research, also emerged from the comments on the 2006 draft. In 118 conducting some studies, epidemiologists alter the physical, chemical, social or 119 psychological conditions to which members of a population are exposed. Such 120 population trials resemble the clinical trials of new drugs and devices that are the 121 principal focus of the 2002 CIOMS International Ethical Guidelines except that 122 sometimes the unit of study is not an individual person but a community or other 123 group.

It would therefore be possible for those interventional or experimental 124 epidemiological studies to be designed and reviewed under the 2002 document, and to 125 restrict the current guidelines to the unique features of observational epidemiological 126 studies. But such an approach would have several disadvantages. 127 First, while the CIOMS 2002 guidelines are familiar to many ethical review 128 committees, some committees that only review epidemiological studies may not be 129 familiar with them. These committees would be better served by a single booklet that 130 5 addresses both observational and interventional epidemiology.

Likewise, 131 epidemiologists, who may move from conducting an observational study to an 132 interventional study, should not have to shift back and forth between one document 133 and the other. Therefore, it is the intention of the core group that this guidance 134 document encompass all types of epidemiological studies. 135 In the case of interventional studies, the present Guidelines are generally the 136 same as those in the 2002 document, 1 but whenever appropriate the commentary has 137 been focused on issues that arise in epidemiological rather than biomedical research. 138 For example, in biomedical research the focus of protection is typically on avoiding 139 physical or psychological harm; in the context of international studies, attention has 140 often been directed towards the responsibilities of commercial sponsors of research 141 from high-resource countries when they test their products in low-resource countries.

142 In epidemiological research, the sponsors are more likely to be public agencies than 143 commercial companies and the risks are more likely to involve socioeconomic harm 144 (such as through dissemination of private information) rather than physical injury. Of 145 course, some epidemiological research does utilize biomedical interventions (such as 146 vaccines) in one or more population groups, where the risk can be physical as well as 147 social. And even observational studies can involve physical risks, if the failure to 148 change conditions means that some of the subjects of a study are exposed to avoidable 149 risks of harm.

Indeed, one of the most infamous cases of unethical research in the 150 Twentieth Century, the so-called Tuskegee Syphilis Study, involved the observation 151 of untreated disease in a group of poor African-Americans in rural Alabama over a 152 forty-year period; the fact that the subjects were unaware of their diagnosis and of the 153 purpose for the public health officials‟ interest in them and were not offered treatment 154 when antibiotics became generally available caused a scandal that propelled the 155 development of formal rules for research with human beings in the United States. 156 The present Guidelines address observational studies by noting, in the 157 commentary, the ways in which it may be appropriate to treat such research 158 differently than interventional studies (for example, regarding informed consent).

159 160 1 Small changes have been made in the wording of Guidelines where this was necessary in light of differences between biomedical and epidemiological research, and three Guidelines (22-24) have been added in the present document; they are not, however, narrowly concerned with epidemiological research and are considered appropriate for inclusion in CIOMS‟s next edition of the International Ethical Guidelines for Biomedical Research Involving Human Subjects. 6 INTRODUCTION 161 162 “Epidemiology is the study of the distribution and determinants of health-related 163 states or events in specified populations, and the application of this study to control of 164 health problems.” (John Last, Dictionary of Epidemiology, 4 th edition) This volume 165 sets forth ethical guidance regarding the first part of this definition, namely, how 166 epidemiologists–as well as those who sponsor, review, or participate in the studies 167 they conduct–should identify and respond to the ethical issues that are raised by the 168 process of producing this information.

169 Epidemiology has made essential contributions to the improvement in human 170 health achieved over the past century. It can be reasonably expected that the field will 171 continue to do so by using ever more powerful and sophisticated scientific tools to 172 increase the understanding of the distribution of health and illness and of their many 173 physical, chemical, biological, behavioural, social and environmental determinants. 174 Indeed, further improving the health of the public depends on making greater use of 175 the tools of epidemiology. At the same time, it is essential that this new knowledge, 176 and the changes for the good that it prompts, be derived from studies conducted 177 according to recognized ethical standards.

By focusing on the distinctive aspects of 178 epidemiological research, this document aims to provide the field with just such a set 179 of ethical standards. 180 Epidemiological research today encompasses a wide spectrum of research 181 ranging from the investigation of disease causation using the tools of molecular 182 biology in populations to the evaluation of health services and from analysis of the 183 social factors conditioning health and disease to large-scale studies of new public 184 health interventions to prevent disease. All aspects of health when studied at the level 185 of the population are the proper domain of epidemiology, which also provides 186 essential inputs for clinicians, policymakers and social analysts, for example on 187 disease frequency or on the effects of different interventions to control a disease.

188 In epidemiology, the term "studies" encompasses both routine applications of 189 epidemiological methods–for example, in public health surveillance or hospital 190 quality evaluation–and investigations designed to produce new scientific knowledge 191 and theories; the latter are addressed in the present Guidelines and commentaries. The 192 text adopts the usage common in biomedical research, in which the term "study" is 193 7 used–along with "investigation" or "trial"–to designate research activities; thus in 194 what follows, references to "epidemiological studies" denote epidemiological research 195 rather than practice.

196 197 Research and practice. In order to avoid imposing on the ordinary practice of 198 medicine all the rules and procedures created over the past six decades to protect 199 research subjects, it is conventional to define "research" as involving activities that 200 are designed to develop or contribute to generalizable knowledge. Generalizable 201 knowledge consists of theories, principles or relationships, or the accumulation of 202 information on which they are based, that can be corroborated by accepted scientific 203 methods of observation and inference. In contrast, when a physician or psychologist 204 varies conventional treatment in an attempt to produce a better result for a patient, one 205 might say that he or she was “experimenting” but since such individualized variations 206 do not produce generalizable knowledge, the activity would be regarded as practice 207 not research.

208 The “generalizable knowledge” definition works well for medical and 209 behavioural studies pertaining to human health, which are commonly denominated 210 “biomedical research” to indicate its relation to health. But the definition works less 211 well in separating practice from research in the field of epidemiology. Many studies 212 using the tools of epidemiology which are performed on a regular basis by public 213 health agencies, such as routine surveillance for disease outbreaks, are correctly 214 viewed as “practice” even though the information produced may contribute to 215 generalizable knowledge.

Thus, in carrying out their activities epidemiologists (and 216 others examining the activities) need to apply careful judgment to determine whether 217 the activity should be classified as research or practice. Of course, as explained more 218 fully in these Guidelines, it does not necessarily follow that everything placed in the 219 former category is problematic or is even subject to all the requirements for advance 220 approval and individualized informed consent usually associated with research. 221 Conversely, some activities that are routinely carried out by epidemiologists do raise 222 ethical issues that may benefit from careful scrutiny or even reconsideration, even if 223 they have long traditions and are sanctioned by regulations or statutes.

224 Ethics and epidemiology. Progress in medical care and disease prevention 225 depends upon an understanding not only of physiological and pathological processes 226 8 but also of the social, cultural, economic, and other environmental determinants of 227 health, including the effects of the health-care system and other social institutions. 228 Producing that understanding requires performing research involving human subjects. 229 Such research should be carried out only by, or strictly supervised by, suitably 230 qualified and experienced investigators under accepted ethical guidelines. 231 Ethical guidelines assist both investigators and ethical review committees in 232 acting responsibly.

Investigators, with whom rests the primary duty to protect the 233 rights and welfare of research subjects and to ensure the scientific quality of research, 234 can benefit through better design and administration of their protocols, including the 235 processes for obtaining consent and communicating their research findings, while 236 ethical review committees can benefit through improved evaluation and oversight of 237 studies. In their respective roles, each has a duty to see that research plans are 238 transparent, that subjects‟ data and biological samples are actually used for valid 239 studies, that study results are made publicly available, and that unnecessary 240 administrative obstacles to research–should they occur–are effectively removed.

241 Because of their merely observational nature, epidemiological studies in the past 242 were widely regarded as not raising any significant ethical issues and were commonly 243 carried out without approval of an ethical review committee. However, recent years 244 have brought increased attention to the ethical conduct of research generally, greater 245 awareness of the potential harms to research subjects including non-physical harm 246 from disclosures of health-related information and hence increased efforts to protect 247 privacy. All of these have implications for observational epidemiological research.

248 Investigators and review committees need to take differences between interventional 249 and observational studies into consideration in designing and approving observational 250 studies. In some cases, the differences can simplify the ethical review process; in 251 others, additional considerations are raised. 252 The mere formulation of ethical guidelines for epidemiological research 253 involving human subjects will hardly resolve all the moral doubts that can arise in 254 association with such research. Nonetheless, the present Guidelines are intended at 255 least to draw the attention of investigators, sponsors and ethical review committees to 256 the need to consider carefully the ethical implications of research protocols and the 257 manner in which research is conducted, and thus to conduce to high scientific and 258 ethical standards in epidemiological research.

259 1 INTERNATIONAL INSTRUMENTS AND GUIDELINES 260 261 The first official international statement on the ethics of medical research was 262 promulgated in 1947 as part of the judgment of the court in Nuremberg that tried the 263 Nazi physicians who had conducted atrocious experiments on unconsenting prisoners 264 and detainees during the Second World War. The judges set forth ten conditions–265 which became known as the Nuremberg Code–for the ethical conduct of research 266 involving human subjects, emphasizing the necessity of voluntary consent. 267 The Universal Declaration of Human Rights, adopted by the United Nations 268 General Assembly in 1948 in the wake of the judgment in The Doctors' Case, states 269 that "No one shall be subjected . . . to cruel, inhuman or degrading treatment or 270 punishment" (Article 5).

The International Covenant on Civil and Political Rights, 271 approved by the General Assembly in 1966 to give the Declaration legal as well as 272 moral force, explicates that this prohibition means that "In particular, no one shall be 273 subjected without his free consent to medical or scientific experimentation." (Article 274 7). (Many countries have incorporated this provision or its equivalent into their 275 constitution or public health laws and regulations.) Subsequent human rights 276 instruments, which provide special protection to women (Convention on the 277 Elimination of All Forms of Discrimination Against Women) and children 278 (Convention on the Rights of the Child), reinforce the connection between human 279 rights and the ethical principles that underlie a number of international guidelines for 280 research with human beings.

281 The most prominent of these began taking shape in the 1950s, when the World 282 Medical Association (WMA) began the process of articulating a set of duties for 283 physicians conducting medical research. Though it drew on the Nuremberg Code, the 284 WMA wanted a set of standards that was generated by the profession itself (free of 285 any association with the wartime physician-criminals) and that encompassed research 286 undertaken in the course of medical care. The resulting declaration, adopted at the 287 WMA meeting in Helsinki in 1964, became a fundamental document in the field of 288 research ethics and has influenced the formulation of international, regional and 289 national legislation and codes of conduct.

The Declaration, which has been amended 290 several times, most recently in 2000 (Appendix 3), is a comprehensive international 291 statement of the ethics of research involving human subjects. It sets out ethical 292 2 guidelines for physicians engaged in both clinical and nonclinical biomedical 293 research. 294 Since the publication of the CIOMS 1993 Guidelines, several international 295 organizations have issued ethical guidance on clinical trials. These have included, 296 from the World Health Organization, in 1995, Guidelines for Good Clinical Practice 297 for Trials on Pharmaceutical Products; and from the International Conference on 298 Harmonisation of Technical Requirements for Registration of Pharmaceuticals for 299 Human Use (ICH), in 1996, Guideline on Good Clinical Practice, designed to ensure 300 that data generated from clinical trials are mutually acceptable to regulatory 301 authorities in the European Union, Japan and the United States of America.

The Joint 302 United Nations Programme on HIV/AIDS published in 2000 the UNAIDS Guidance 303 Document Ethical Considerations in HIV Preventive Vaccine Research; a revised 304 version, Ethical Considerations in Biomedical HIV Prevention Trials, was produced 305 by UNAIDS and WHO in 2007. 306 In 2001 the Council of Ministers of the European Union adopted a Directive on 307 clinical trials, which became binding in law in the countries of the Union in 2004. The 308 Council of Europe, with more than 40 Member States, has approved a Protocol on 309 Biomedical Research (which was opened for ratification by Member States on 25 310 January 2005) to implement the provisions of its 1997 Convention on Human Rights 311 and Biomedicine that relate to biomedical research.

312 313 3 GENERAL ETHICAL PRINCIPLES 314 315 All research involving human subjects should be conducted in accordance with 316 three basic ethical principles, namely respect for persons, beneficence and justice. It is 317 generally agreed that these principles, which in the abstract have equal moral force, 318 guide the conscientious preparation of proposals for scientific studies. In varying 319 circumstances they may be expressed differently and given different moral weight, 320 and their application may lead to different decisions or courses of action. The present 321 guidelines are directed at the application of these principles to research involving 322 human subjects.

323 324 Respect for persons incorporates at least two fundamental ethical 325 considerations, namely: 326 a) respect for autonomy, which requires that those who are capable of 327 deliberation about their personal choices should be treated with respect for their 328 capacity for self-determination; and 329 b) protection of persons with impaired or diminished autonomy, which requires 330 that those who are dependent or vulnerable be afforded security against harm or 331 abuse. 332 333 Beneficence refers to the ethical obligation to maximize benefits and to 334 minimize harms. This principle gives rise to norms requiring that the risks of research 335 be reasonable in the light of the expected benefits, that the research design be sound, 336 and that the investigators be competent both to conduct the research and to safeguard 337 the welfare of the research subjects.

Beneficence further proscribes the deliberate 338 infliction of harm on persons; this aspect of beneficence is sometimes expressed as a 339 separate principle, nonmaleficence (do no harm). 340 341 Justice refers to the ethical obligation to treat each person in accordance with 342 what is morally right and proper, to give each person what is due to him or her. In the 343 ethics of research involving human subjects the principle refers primarily to 344 distributive justice, which requires the equitable distribution of both the burdens and 345 the benefits of participation in research. Differences in distribution of burdens and 346 4 benefits are justifiable only if they are based on morally relevant distinctions between 347 persons; one such distinction is vulnerability.

"Vulnerability" refers to a substantial 348 incapacity to protect one's own interests owing to such impediments as lack of 349 capability to give informed consent, lack of alternative means of obtaining medical 350 care or other expensive necessities, or being a junior or subordinate member of a 351 hierarchical group. Accordingly, special provision must be made for the protection of 352 the rights and welfare of vulnerable persons. 353 Sponsors of research or investigators cannot, in general, be held accountable for 354 unjust conditions where the research is conducted, but they must refrain from 355 practices that are likely to worsen unjust conditions or contribute to new inequities.

356 Neither should they take advantage of the relative inability of low-resource countries 357 or vulnerable populations to protect their own interests, by conducting research 358 inexpensively and avoiding complex regulatory systems of industrialized countries in 359 order to develop products for the lucrative markets of those countries. 360 In general, the research project should leave low-resource countries or 361 communities better off than previously or, at least, no worse off. It should be 362 responsive to their health needs and priorities in that any product developed is made 363 reasonably available to them, and as far as possible leave the population in a better 364 position to obtain effective health care and protect its own health.

365 Justice requires also that the research be responsive to the health conditions or 366 needs of vulnerable subjects. The subjects selected should be the least vulnerable 367 necessary to accomplish the purposes of the research. Risk to vulnerable subjects is 368 most easily justified when it arises from interventions or procedures that hold out for 369 them the prospect of direct health-related benefit. Risk that does not hold out such 370 prospect must be justified by the anticipated benefit to the population of which the 371 individual research subject is representative. 372 An issue, mainly for those countries and perhaps less pertinent now than in the 373 past, has been the extent to which ethical principles are considered universal or as 374 culturally relative – the universalist versus the pluralist view.

The challenge to 375 international research ethics is to apply universal ethical principles to biomedical 376 research in a multicultural world with a multiplicity of health-care systems and 377 considerable variation in standards of health care. The Guidelines take the position 378 that research involving human subjects must not violate any universally applicable 379 5 ethical standards, but acknowledge that, in superficial aspects, the application of the 380 ethical principles, e.g., in relation to individual autonomy and informed consent, 381 needs to take account of cultural values, while respecting absolutely the ethical 382 standards.

383 Finally, it is important to remember the basic distinction between legal norms 384 and ethical norms. While the former are founded on the latter, there is no necessary 385 one-to-one correspondence between each legal and ethical norm. A law may be 386 regarded as unethical by some people (e.g., a law prescribing the death penalty for 387 certain crimes) and likewise, an ethical norm may be regarded as unlawful in a 388 country (e.g., one involving female genital mutilation). Thus it cannot be expected 389 that ethical guidelines, which translate ethical principles into the form of 390 recommendations (rather than of strict norms), will always coincide with legal 391 prescriptions.

This applies all the more to international guidelines which are issued in 392 the context of legal dispositions varying from one country to another. 393 394 6 THE GUIDELINES 395 396 Guideline 1 397 Ethical justification and scientific validity of 398 epidemiological research involving human beings 399 400 The ethical justification of epidemiological research involving human subjects 401 is the prospect of discovering new ways of improving the health of individuals, 402 groups and populations. Such research can be ethically justifiable only if it is 403 carried out in ways that respect and protect, and are fair to, research subjects 404 and that are morally acceptable within the communities in which the research 405 is carried out.

Moreover, because scientifically invalid research is unethical in 406 that it exposes research subjects to risks without possible benefit, 407 investigators and sponsors must ensure that proposed studies involving 408 human subjects conform to generally accepted scientific principles and are 409 based on adequate knowledge of the pertinent scientific literature. 410 411 Commentary on Guideline 1 412 413 General considerations. Among the essential features of ethically justified research 414 involving human subjects, including research with identifiable human tissue or data, 415 are that the research offers a means of developing information not otherwise 416 obtainable, that the design of the research is scientifically sound, and that the 417 investigators and other research personnel are competent.

The methods to be used 418 should be appropriate to the objectives of the research and the field of study. 419 Investigators and sponsors must also ensure that all who participate in the conduct of 420 the research are qualified by virtue of their education and experience to perform 421 competently in their roles. These considerations should be adequately reflected in the 422 research protocol submitted for review and clearance (Appendix 2 specifies the items 423 to be included in a protocol, when relevant). Scientific review is discussed further in 424 the Commentaries to Guidelines 2 and 3: Ethical review committees and Ethical 425 review of externally sponsored research.

Other ethical aspects of research are 426 discussed in the remaining guidelines and their commentaries. 427 7 428 Observational studies. While observational research normally does not pose a risk of 429 physical harm to individuals, this is not always the case, for several reasons. First, in 430 some non-experimental studies researchers intervene physically with subjects, such as 431 by taking blood or tissue samples. Second, even when an observational study 432 involves only questionnaires or record-examination, subjects may be at risk of 433 physical or psychological harm. For example, interviewing women in a study of 434 domestic violence may expose them to the risk of further violence.

A risk of 435 psychological harm may be present when sensitive questions are asked, for instance 436 asking parents about events surrounding a child's death, or details about sexual habits. 437 Likewise, initiating research on workplace hazards may cause anxiety among both 438 employees and employers. Even research limited to an examination of existing 439 records may entail a risk for the group under investigation (such as stigmatization) or, 440 without causing measurable harm, it may still wrong people by making use of 441 information that they regard as private. Careful planning, open discussions with all 442 concerned parties (such as representatives of workers and managers in occupational 443 health research), vigorous efforts to protect confidential data, and pooling data to 444 larger entities are all part of good study design.

445 446 Guideline 2 447 Ethical review committees 448 449 All proposals to conduct epidemiological research involving human subjects 450 must be submitted for review of their scientific merit and ethical acceptability 451 to one or more scientific review and ethical review committees. The review 452 committees must be independent of the research team, and any direct financial 453 or other material benefit they may derive from the research should not be 454 contingent on the outcome of their review. The investigator must obtain their 455 approval or clearance before undertaking the research. The ethical review 456 committee should conduct further reviews as necessary in the course of the 457 research, including monitoring the progress of the study.

458 459 Commentary on Guideline 2 460 8 461 Inclusion in, or exemption from, review. Research involves human subjects when an 462 investigator will directly obtain information from individuals or groups or will 463 otherwise acquire identifiable private information about them. Proposals for 464 epidemiological studies, like other research with human subjects, must usually 465 undergo prior scientific and ethical review, although some observational studies, such 466 as those utilizing publicly available or anonymous data, may not be subject to prior 467 review and approval by an ethical review committee under the regulations of the local 468 jurisdiction.

When in doubt about whether or not a study involves elements that 469 warrant ethical review, epidemiologists are encouraged to consult the ethical review 470 committee or to submit their studies for review. For example, a study of sensitive 471 topics or behavior (illicit drug use; domestic violence; etc.) may merit review because 472 of its potential effects on a community or group even if the data were to be recorded 473 anonymously. Even when an exemption is claimed, the research protocol should 474 provide justification for the claimed exemption. A public health study not submitted 475 to an ethical review committee should receive administrative confirmation by a 476 competent authority that the study is exempt from review.

Epidemiologists should 477 keep in mind that scientific journals generally require that papers submitted for 478 publication have received prior review by an ethical review committee. 479 480 General observations. Ethical and scientific review committees may function at the 481 institutional, local, regional, or national level, and in some cases at the international 482 level. The regulatory or other governmental authorities concerned should promote 483 uniform standards across committees within a country, and, under all systems, 484 sponsors of research and institutions in which the investigators are employed should 485 allocate sufficient resources to the review process.

Review committees may receive 486 money for the activity of reviewing protocols, but under no circumstances may 487 payment be conditioned on a review committee‟s approval or clearance of a protocol. 488 489 Scientific review. Scientific review and ethical review are intertwined: scientifically 490 unsound research involving humans as subjects is ipso facto unethical in that it may 491 expose them to risk or inconvenience to no purpose; even if there is no risk of injury, 492 the wasting of subjects‟ and researchers‟ time in unproductive activities represents 493 9 loss of a valuable resource.

Epidemiological research involving humans must conform 494 to generally accepted scientific principles, and be based on a thorough knowledge of 495 the scientific literature and other relevant sources of information, as well as adequate 496 preparatory studies. Scientific review must consider, inter alia, the study design, 497 including the provisions for avoiding or minimizing risk and for monitoring safety 498 when applicable, as well as the scientific qualifications of the investigators (including 499 education in the principles of research practice). 500 501 Ethical review. The ethical review committee is responsible for safeguarding the 502 rights, safety, and well-being of the research subjects.

Many ethical review 503 committees consider both the scientific and the ethical aspects of proposed research; 504 when the tasks are separated, the ethical review committee must verify that another 505 competent expert body has determined that the research is scientifically sound. The 506 ethical review committee should also ensure that provisions for monitoring of data 507 and safety are in place, either through the committee itself or another body. 508 509 Once a research proposal has been found scientifically sound, the ethical review 510 committee should consider whether any known or possible risks to the subjects are 511 justified by the expected benefits, direct or indirect, and whether the proposed 512 research methods will minimize harm and maximize benefit.

(See Guideline 8: 513 Benefits, harms and risks of study participation.) If the proposal is sound and the 514 balance of risks to anticipated benefits is reasonable, the committee should then 515 determine whether the procedures proposed for obtaining informed consent, when 516 applicable (see Guideline 4), are satisfactory and the process for selecting subjects is 517 equitable. The committee is also responsible for ensuring that all other ethical 518 concerns arising from a protocol are satisfactorily resolved both in principle and in 519 practice, for keeping records of its decisions, and for taking measures to follow up on 520 the conduct of ongoing research projects.

521 522 National (centralized) or local review. Ethical review committees may be created 523 under the aegis of national or local health administrations, national (or centralized) 524 research councils or other nationally representative bodies. In a highly centralized 525 administration a national, or centralized, review committee may be constituted for 526 10 both the scientific and the ethical review of research protocols. In countries where 527 research is not centrally administered, ethical review is more effectively and 528 conveniently undertaken at a local or regional level. The authority of a local ethical 529 review committee may be confined to a single institution (such as a hospital, research 530 institute, or university) or may extend to all institutions in which research is carried 531 out within a defined geographical area.

532 However committees are created, and however their jurisdiction is defined, they 533 should establish working rules regarding, for instance, frequency of meetings, a 534 quorum of members, decision-making procedures, and review of decisions. The rules 535 should protect the confidentiality of review-committee documents and discussions. 536 The committee should provide its rules to prospective investigators, and should also 537 never compel investigators to submit to unnecessary repetition of review. 538 539 Committee membership. Committees competent to review the scientific and/or ethical 540 aspects of epidemiological research proposals have competence on all relevant topics;

541 such committees must be multidisciplinary, including epidemiologists and other 542 experts in the design and analysis of population health studies. It is important that at 543 least some committee members (or experts co-opted on an ad hoc basis, as needed for 544 particular studies) be knowledgeable and up-to-date about statistical methods as 545 applied to epidemiology including sampling methodology in general, as well as about 546 the populations being studied in particular (e.g., concerning the existence of 547 subpopulations, social structure, hazards at work, etc.). 548 In addition to such experts, the membership should include other professionals 549 such as physicians, nurses, lawyers, ethicists and clergy, as well as lay persons 550 qualified to represent the cultural and moral values of the community and to ensure 551 that the rights of the research subjects will be respected.

Lack of formal education 552 should not disqualify community members from joining in constructive discussion on 553 issues relating to the study and the application of its findings, and when illiterate 554 persons form the focus of a study they should either be considered for membership or 555 invited to have their views expressed. Committees should include both men and 556 women. A number of members should be replaced periodically, with the aim of 557 blending the advantages of experience with those of fresh perspectives. 558 Committees that often review occupational health research should include 559 11 workers‟ representatives, and those that often review research proposals directed at 560 specific diseases or impairments should invite or hear the views of individuals or 561 bodies representing patients with such diseases or impairments.

Similarly, for research 562 involving such subjects as children, students, elderly persons or employees, 563 committees should invite, or solicit the views of, their representatives or advocates. 564 565 Multi-centre research. Some research projects are designed to be conducted in a 566 number of centres in different communities or countries. Generally, to ensure that the 567 results will be valid, the study must be conducted in an identical way at each centre. 568 Such studies include various kinds of epidemiological research and evaluations of 569 health service programmes in addition to clinical trials.

In multi-centre studies, local 570 ethical or scientific review committees are not normally authorized to change 571 inclusion or exclusion criteria or to make other, similar modifications. They should be 572 fully empowered, however, to prevent a study that they believe to be unethical. 573 Moreover, changes that local review committees believe are necessary to protect the 574 research subjects should be documented and reported to the research institution or 575 sponsor responsible for the whole research programme for consideration and due 576 action, to ensure that all other subjects can be protected and that the research will be 577 valid across sites.

578 To ensure the validity of multi-centre research, any change in the protocol 579 should be made at every collaborating centre or institution, or, failing this, explicit 580 inter-centre comparability procedures must be introduced; changes made at some but 581 not all will defeat the purpose of multi-centre research. For some multi-centre studies, 582 scientific and ethical review may be facilitated by agreement among centres to accept 583 the conclusions of a single review committee; its members could include a 584 representative of the ethical review committee at each of the centres at which the 585 research is to be conducted, as well as individuals competent to conduct scientific 586 review.

In other circumstances, a centralized review may be complemented by local 587 review relating to the local participating investigators and institutions. The central 588 committee can review the study from a scientific and ethical standpoint, while the 589 local committees verify the practicability of the study in their communities, including 590 the infrastructures, the state of training, and ethical considerations of local 591 significance. 592 12 In a large multi-centre epidemiological study, individual investigators will not 593 have authority to act independently, with regard to data analysis or to preparation and 594 publication of manuscripts, for instance.

Such a trial usually has a set of committees 595 which operate under the direction of a steering committee and are responsible for such 596 functions and decisions. The function of the ethical review committee in such cases is 597 to review the relevant plans with the aim of avoiding abuses. 598 599 Research in emergency situations. The emerging best practice for research conducted 600 during an emergency–such as population studies of outbreaks of disease or of 601 disasters (and relief efforts)–is to establish the basic research design for various 602 categories of research prior to the emergency.

Among other benefits, this permits 603 prior ethical review of at least the major features of the research design. When prior 604 review has not occurred, a review should be provided as quickly as possible. The 605 special problems in obtaining informed consent in emergencies are addressed in the 606 Commentary on Guideline 6. 607 608 Sanctions. Ethical review committees generally have no authority to impose sanctions 609 on researchers who violate ethical standards in the conduct of research involving 610 humans. They may, however, withdraw ethical approval of a research project if 611 judged necessary.

They should be required to monitor the implementation of an 612 approved protocol and its progression, and to report to institutional or governmental 613 authorities any serious or continuing non-compliance with ethical standards as they 614 are reflected in protocols that they have approved or in the conduct of the studies. 615 Failure to submit a protocol to the committee should be considered a clear and serious 616 violation of ethical standards. 617 Sanctions imposed by governmental, institutional, professional or other 618 authorities possessing disciplinary power should be employed as a last resort.

619 Preferred methods of control include cultivation of an atmosphere of mutual trust, and 620 education and support to promote in researchers and in sponsors the capacity for 621 ethical conduct of research. 622 Should sanctions become necessary, they should be directed at the non-623 compliant researchers or sponsors. They may include fines or suspension of eligibility 624 to receive research funding, to use investigational interventions, or to practise their 625 13 profession. Unless there are persuasive reasons to do otherwise, editors should refuse 626 to publish the results of research conducted unethically and retract any articles that are 627 subsequently found to contain falsified or fabricated data or to have been based on 628 unethical research.

Drug regulatory authorities should consider refusal to accept 629 unethically obtained data submitted in support of an application for authorization to 630 market a product. Such sanctions, however, may deprive of benefit not only the errant 631 researcher or sponsor but also that segment of society intended to benefit from the 632 research; such possible consequences merit careful consideration. 633 634 Potential conflicts of interest. To maintain a review committee‟s independence from 635 the investigators and sponsors and to avoid conflict of interest, any member with a 636 special or particular interest in a proposal (whether direct or indirect) should not take 637 part in assessing the proposal if that interest could subvert the member‟s objective 638 judgment.

Members of ethical review committees should be held to the same standard 639 of disclosure as scientific and medical research staff with regard to financial or other 640 interests that could be construed as conflicts of interest. A practical way of avoiding 641 such conflict of interest is for the committee to insist on a declaration of possible 642 conflict of interest by any of its members. A member who makes such a declaration 643 should then withdraw, when doing so is clearly appropriate, either at the member‟s 644 own discretion or at the request of the other members. Before withdrawing, the 645 member should be permitted to offer comments on the protocol or to respond to 646 questions of other members.

647 Research sponsors (whether commercial enterprises, governments, or 648 foundations) have good reasons to support studies that are ethically and scientifically 649 acceptable, but cases have arisen in which the conditions of funding may have 650 introduced bias. For example, an investigator may have little or no input into trial 651 design, limited access to the raw data, or limited participation in data interpretation, or 652 study results may not be published if they are unfavourable to the sponsor's product or 653 activity. As the persons directly responsible for their work, investigators should not 654 enter into agreements that interfere unduly with their access to the data or their ability 655 to analyse the data independently, to prepare manuscripts, or to publish them.

656 Investigators must disclose potential or apparent conflicts of interest on their 657 part to the ethical review committee or to other institutional committees designed to 658 14 evaluate and manage such conflicts. Guidance on mechanisms for ethical review 659 committees to deal with conflicts of interest appears in Guideline 22. (See also Multi-660 centre research, above.) 661 662 Guideline 3 663 Ethical review of externally sponsored research 664 665 An external sponsoring organization and individual investigators should 666 submit the research protocol for ethical and scientific review in the country of 667 the sponsoring organization, and the ethical standards applied should be no 668 less stringent than they would be for research carried out in that country.

The 669 health authorities of the host country, as well as a national or local ethical 670 review committee, should ensure that the proposed research is responsive to 671 the health needs and priorities of the host country and meets the requisite 672 ethical standards. 673 674 Commentary on Guideline 3 675 676 Definition. The term externally sponsored research refers to research undertaken in 677 one country (the host) but sponsored, financed, and sometimes wholly or partly 678 carried out by an external international or national organization or company with the 679 collaboration or agreement of the appropriate authorities, institutions and personnel of 680 the host country.

681 682 Ethical and scientific review. Committees in both the country of the sponsor and 683 the host country have responsibility for conducting scientific and ethical review, as 684 well as the authority to withhold approval of research proposals that fail to meet their 685 scientific or ethical standards. As far as possible, there must be assurance that the 686 review is independent and that there is no conflict of interest that might affect the 687 judgment of members of the review committees in relation to any aspect of the 688 research. When the external sponsor is an international organization, its review of the 689 research protocol must be in accordance with its own independent ethical-review 690 procedures and standards.

Committees responsible for reviewing and approving 691 proposals for externally sponsored research should have among their members or 692 15 consultants persons who are thoroughly familiar with the customs and traditions of the 693 population or community concerned and sensitive to issues of human dignity. 694 Committees in the external sponsoring country or international organization 695 have a special responsibility to determine whether the scientific methods are sound 696 and suitable to the aims of the research; whether the drugs, vaccines, devices or 697 procedures to be studied meet adequate standards of safety; whether there is sound 698 justification for conducting the research in the host country rather than in the country 699 of the external sponsor or in another country; and whether the proposed research is in 700 compliance with the ethical standards of the external sponsoring country or 701 international organization.

702 Committees in the host country have a special responsibility to determine 703 whether the objectives of the research are responsive to the health needs and priorities 704 of that country. The ability to judge the ethical acceptability of various aspects of a 705 research proposal requires a thorough understanding of a community's customs and 706 traditions. The ethical review committee in the host country, therefore, must have as 707 either members or consultants persons with such understanding; it will then be in a 708 favourable position to determine the acceptability of the proposed means of obtaining 709 informed consent and otherwise respecting the rights of prospective subjects as well 710 as of the means proposed to protect the welfare of the research subjects.

Such persons 711 should be able, for example, to indicate suitable members of the community to serve 712 as intermediaries between investigators and subjects, and to advise on whether 713 material benefits or inducements may be regarded as appropriate in the light of a 714 community's gift-exchange and other customs and traditions. 715 When a sponsor or investigator in one country proposes to carry out research in 716 another, the ethical review committees in the two countries may, by agreement, 717 undertake to review different aspects of the research protocol. In short, in respect of 718 host countries either with developed capacity for independent ethical review or in 719 which external sponsors and investigators are contributing substantially to developing 720 such capacity, ethical review in the external, sponsoring country may be limited to 721 ensuring compliance with broadly stated ethical standards.

The ethical review 722 committee in the host country can be expected to have greater competence for 723 reviewing the detailed plans for compliance, in view of its better understanding of the 724 cultural and moral values of the population in which it is proposed to conduct the 725 16 research; it is also likely to be in a better position to monitor compliance in the course 726 of a study. However, in respect of research in host countries with inadequate capacity 727 for independent ethical review, full review by the ethical review committee in the 728 external sponsoring country or international agency is necessary.

729 730 Industry-sponsored research. In industry-sponsored research on possible occupational 731 hazards, the protection of confidential information on products and production 732 processes should be respected. Such protection should not, however, prevail over the 733 primary interests of identifying potential health effects and of communicating the 734 research results to all involved parties and to the scientific community. 735 736 Guideline 4 737 Individual informed consent 738 739 For all epidemiological research involving humans the investigator must obtain 740 the voluntary informed consent of the prospective subject or, in the case of an 741 individual who is not capable of giving informed consent, the permission of a 742 legally authorized representative in accordance with applicable law.

Waiver of 743 individual informed consent is to be regarded as exceptional, and must in all 744 cases be approved by an ethical review committee unless otherwise permitted 745 under national legislation that conforms to the ethical principles in these 746 Guidelines. 747 748 Commentary on Guideline 4 749 750 General considerations. Voluntary informed consent is a decision to participate in 751 research, taken by a competent individual who has received the necessary 752 information; who has adequately understood the information; and who, after 753 considering the information, has arrived at a decision without having been subjected 754 to coercion, undue influence or inducement, or intimidation.

755 Informed consent is based on the principle that competent individuals are 756 entitled to choose freely whether to participate in research. Informed consent 757 embodies the individual's freedom of choice and respects the individual's autonomy. 758 17 As an additional safeguard, it must always be complemented by independent ethical 759 review of research proposals. This safeguard of independent review is particularly 760 important as many individuals are limited in their capacity to give adequate informed 761 consent; they include young children, adults with severe mental or behavioural 762 disorders, and persons who are unfamiliar with medical concepts and technology (See 763 Guidelines 13, 14 and 15).

764 765 Process. Obtaining informed consent is a process that is begun when initial contact is 766 made with a prospective subject and continues throughout the course of the study. By 767 informing the prospective subjects, by repetition and explanation, by answering their 768 questions as they arise, and by ensuring that each individual understands each 769 procedure, investigators elicit their informed consent and in so doing manifest respect 770 for their dignity and autonomy. Each individual must be given as much time as is 771 needed to reach a decision, including time for consultation with family members or 772 others.

Adequate time and resources should be set aside for informed-consent 773 procedures. 774 775 Language. Informing the individual subject must not be simply a ritual recitation of 776 the contents of a written document. Rather, the investigator must convey the 777 information, whether orally or in writing, in language that suits the individual's level 778 of understanding. The investigator must bear in mind that the prospective subject‟s 779 ability to understand the information necessary to give informed consent depends on 780 that individual's maturity, intelligence, education and belief system.

It depends also on 781 the investigator's ability and willingness to communicate with patience and 782 sensitivity. 783 784 Comprehension. The investigator must then ensure that the prospective subject has 785 adequately understood the information. The investigator should give each one full 786 opportunity to ask questions and should answer them honestly, promptly and 787 completely. In some instances the investigator may administer an oral or a written test 788 or otherwise determine whether the information has been adequately understood. 789 (See also Commentary on Guideline 6) 790 791 18 Documentation of consent.

Consent may be indicated in a number of ways. The 792 subject may imply consent by voluntary actions, express consent orally, or sign a 793 consent form. As a general rule, the subject should sign a consent form, or, in the 794 case of incompetence, a legal guardian or other duly authorized representative should 795 do so. The ethical review committee may approve waiver of the requirement of a 796 signed consent form if the research carries no more than minimal risk–that is, risk that 797 is no more likely and not greater than that attached to routine medical or 798 psychological examination–and if the procedures to be used are only those for which 799 signed consent forms are not customarily required outside the research context.

Such 800 waivers may also be approved when existence of a signed consent form would be an 801 unjustified threat to the subject's confidentiality. Particularly when the information is 802 complicated, it is usually advisable to give subjects information sheets to retain; these 803 may resemble consent forms in all respects except that subjects are not required to 804 sign them. Their wording should be cleared by the ethical review committee. When 805 consent has been obtained orally, for example in a telephone interview, investigators 806 are responsible for providing documentation or proof of consent.

807 808 Renewing consent. When material changes occur in the conditions or the procedures 809 of a study the investigator should once again seek informed consent from the subjects. 810 For example, when a study itself (or another source) generates new information that 811 would have to be disclosed were any subjects being newly recruited to the study, 812 existing subjects should be given such information promptly and asked whether they 813 agree to continue in the study. 814 In long-term studies involving active follow-up, subjects who do not wish to 815 continue will simply stop participating, but in studies involving only passive follow-816 up it is appropriate to inform subjects periodically of the status of the study and to 817 seek their agreement to continue having their on-going records incorporated into the 818 data base.

Prior to the initiation of such long-term studies (i.e., those lasting two or 819 more quinquennia), the plans for such re-consenting should be presented to the ethical 820 review committee responsible for reviewing and approving the study. 821 822 Cultural considerations. In some cultures an investigator may enter a community to 823 conduct research or approach prospective subjects for their individual consent only 824 19 after obtaining permission from a community leader, a council of elders, or another 825 designated authority. Such customs must be respected. In no case, however, may the 826 permission of a community leader or other authority substitute for individual 827 informed consent.

(To avoid a misunderstanding, the person from whom permission is 828 sought should be informed in advance that consent will be still sought from 829 individuals enrolling in research, lest this practice be seen as unanticipated disrespect 830 for his or her authority.) In some populations the use of a number of local languages 831 may complicate the communication of information to potential subjects and the ability 832 of an investigator to ensure that they truly understand it. Many people in all cultures 833 are unfamiliar with, or do not readily understand, scientific concepts such as those of 834 placebo or randomization.

Sponsors and investigators should develop culturally 835 appropriate ways to communicate information that is necessary for adherence to the 836 standard required in the informed consent process. Also, they should describe and 837 justify in the research protocol the procedure they plan to use in communicating 838 information to subjects. For collaborative research in developing countries the 839 research project should, if necessary, include the provision of resources to ensure that 840 informed consent can indeed be obtained legitimately within different linguistic and 841 cultural settings. 842 843 Consultation with community members.

Even when individualized consent is not 844 feasible, investigators may be asked by the ethical review committee to ascertain the 845 views of representative members of the relevant community on the proposed research. 846 Consultation with the community should be sustained throughout the period of the 847 study; eliciting community concerns may require study staff to mobilize the 848 community and provide means for members to express their opinions. The opinions of 849 persons in a position equivalent to those whose biological samples or records will be 850 used in a study offer a relevant point for determining whether such a study would 851 offend community norms of privacy and autonomy.

Such efforts are not the same as 852 obtaining permission from community leaders to undertake a study; rather they are 853 aimed at obtaining the views of people who are in effect proxies for the potential 854 subjects–for example, unions or other workers' organizations for studies involving 855 occupational records, associations that represent population at high risk for disease 856 (such as sex workers‟ groups, in the case of HIV infection), and patient organizations 857 20 for studies involving records or pathology specimens stored at a hospital. In designing 858 their studies, researchers should be guided by this feedback in deciding whether, or to 859 what extent, the persons whose records or specimens will be studied would be likely 860 to object to such use if it were possible to ask them individually; likewise, ethical 861 review committees may request that the researcher supply information from such 862 community consultations as a part of a research proposal to use personally identifiable 863 records or samples without individual informed consent.

The process of community 864 consultation, and the justification for using it, should be specified in the protocol so 865 that the ethical review committee can evaluate what is proposed. 866 867 Community review of, and permission for, studies. Investigators carrying out 868 epidemiological research sometimes include a process of review by representatives of 869 the community in which it is proposed to conduct the study, particularly when the 870 research originates outside that community or even outside the country in which the 871 community is located. Such review can take the form of a "dialogue" with the 872 community about the proposed study and its potential implications, or a more 873 structured consultation that would document the concerns of a socially identifiable 874 group.

In some cases, formal approval may be legally required; for example, under 875 US law, a Native American tribal council must formally approve any research 876 conducted within tribal jurisdiction. In industry-based occupational epidemiology, the 877 agreement and cooperation of employers and employees is a necessary requisite to the 878 conduct of studies. Epidemiologists should usually follow the same approach when 879 developing field investigations, especially when research findings may be presented 880 or interpreted in ways that directly relate to a community or other identifiable group 881 of people or in which the collectivity itself is the unit of analysis.

Those consulted 882 should be in a position to speak on behalf of the community or to reflect its views; 883 researchers should have adequate time and resources to discern how the study 884 population is organized socially and politically and which groups can best speak with 885 authority for the population. Care should, of course, be taken to ensure that those 886 consulted include all relevant groups and do not exclude, for instance, women or 887 members of minority groups. As previously noted, plans for community review 888 should be specified in the protocol, to allow their evaluation by the ethical review 889 committee.

890 21 891 Use of medical records and biological specimens collected for other purposes. 892 People have a right to know that their medical records or biological specimens may be 893 used for research. Records and specimens taken in the course of clinical care, or for 894 an earlier study, may be used for research without the consent of the patients/subjects 895 only if an ethical review committee has determined that the research poses minimal 896 risk, that the rights or interests of the patients will not be violated, that their privacy 897 and confidentiality or anonymity are assured, and that the research is designed to 898 answer an important question and would be impracticable if the requirement for 899 informed consent were to be imposed.

Appropriate standards and procedures are 900 discussed more fully in Guideline 24 and its Commentary. (See also Guideline 18) 901 902 Waiver of consent requirements in epidemiological studies. Investigators should not 903 initiate epidemiological research involving human subjects without first obtaining 904 each subject's informed consent, unless they have received explicit approval to do so 905 from an ethical review committee or the research activity is authorized by legislation 906 or competent authorities in accord with the ethical principles in these Guidelines. 907 Categories of epidemiological research for which consent may be waived include: 908 a. the use of personally non-identifiable materials;

909 b. the use of personally identifiable materials with special justification; 910 c. studies performed within the scope of regulatory authority; 911 d. studies using health-related registries that are authorized under national 912 regulations; and 913 e. cluster-randomized trials. 914 The rules and processes for waiver of consent apply also to situations in which 915 permission is obtained from appropriate surrogates for research involving subjects 916 who lack the capacity to consent for themselves (see Guidelines 14 and 15). 917 918 a. When personally non-identifiable materials are used. As noted under Guideline 2, 919 some epidemiological studies, for example those using publicly available data, may 920 be exempt from ethical review and, a fortiori, from individual informed consent.

In 921 other cases, review may be appropriate but individual consent may not be relevant or 922 required. For example, the individual consent requirement does not arise when the 923 22 materials used in the research are not personally identifiable (meaning that, by 924 definition, the individuals concerned would be unknown to the researcher and hence 925 could not be contacted to obtain consent). 926 927 b. When personally identifiable materials are used. Even when a study involves data 928 or material that carry a person‟s name or that are linked by a code to a person, an 929 ethical review committee may approve observational research using such data or 930 material without requiring individual consent prior to the research.

The committee 931 may do so if it is convinced by the protocol that (a) subjects would be exposed to no 932 more than minimal risk, and (b) either the study involves only publicly available data 933 or the requirement of individual informed consent would make the conduct of the 934 research impracticable. 935 An investigator who proposes not to seek informed consent for a non-936 interventional study that uses personally identifiable information which is not publicly 937 available (including data derived from biological samples and medical records) must 938 justify to an ethical review committee not obtaining consent; the committee should 939 ensure that access to such information is strictly limited in time and extent for the 940 specific research purposes, that allowing the investigator to use it will not 941 compromise the interests or welfare of any persons identified by the data, that any risk 942 of harm will be minimized, that the use accords with locally applicable legal 943 requirements, and that there is no known objection of the individual to such use.

(The 944 obligation of institutions to make available means for people to opt out of having their 945 stored biological material and associated records used for research is discussed in 946 Guideline 24 and the associated Commentary.) 947 The most common justification for using records or samples collected in the 948 past without consent is that it would be impracticable or prohibitively expensive to 949 locate the persons whose samples or records are to be examined; this may happen 950 when, for instance, the study involves reviewing hospital records or performing new 951 tests on blood samples collected at a time when consent to future research uses of 952 such samples was not usually sought (a point further elaborated under Guideline 24).

953 On the other hand, the reluctance of individuals to agree to participate would not 954 constitute impracticability; data from individuals who have specifically rejected such 955 uses in the past may be used only with proper, official authorization in public health 956 23 emergencies. (The special circumstances of consent for research under emergency 957 conditions is elaborated in the Commentary on Guideline 2.) 958 Implicit in the argument for use of personally identifiable material without 959 consent is the claim that the value of the research and the unfeasibility of obtaining 960 consent justify violating a person's interest in becoming a subject of research only 961 with his or her knowledge and agreement.

Thus, the task of the ethical review 962 committee in each case is to evaluate the merits of this claim when set forth by an 963 investigator: how important is the research and could the desired information be 964 produced by another method, what would be the costs and burdens of contacting the 965 persons whose data would be used in the study, how difficult would it be to meet 966 those costs and burdens, and is the imposition of this difficulty justified by the nature 967 of the interests that would be infringed or the potential harm created by allowing the 968 investigator to proceed without consent?

969 The committee should also consider whether any mitigation–such as 970 anonymizing the data–can be undertaken without seriously compromising the 971 scientific merits of the proposed study. When research using personally identifiable 972 data from records or samples collected in the past without an appropriate consent 973 procedure is permitted without consent, the committee should ensure that the 974 investigator (and sponsor) will strictly safeguard the confidentiality of subjects. For 975 this purpose, up-to-date technical means of data encryption may be valuable for 976 safeguarding the confidentiality of records.

977 Anonymization of samples and data will also make it impossible to convey to 978 subjects any findings that might be relevant to the health of the person concerned or 979 family members. Studies that could produce such findings should always include 980 information about the circumstances, if any, under which such findings would be 981 disclosed to the persons concerned or others. It is usually acceptable not to disclose 982 such findings; indeed, it is often imprudent to convey individual findings in research 983 because the significance of the finding will not be well established and the method 984 used may not yet have met the standards used for clinically approved tests.

If it is 985 determined that the research is of a sort that could produce clinically significant 986 findings, an alternative to irreversible anonymization would be to lodge the key to the 987 coding system with an independent third party who would take on the responsibility 988 24 of notifying the persons concerned when a specified potential hazard has been 989 identified. 990 991 c. When studies are performed within the scope of regulatory authority. Consent may 992 also not be required for studies that involve data not publicly available but which are 993 carried out under legislative or regulatory authority for public health, such as disease 994 surveillance.

The extent and limits of such permission are a matter of local law but 995 epidemiologists must still consider whether, in a given case, it is ethical to use their 996 public authority to access personal data for research purposes. When their use of such 997 data does not clearly constitute a public health activity (e.g., when adverse reaction 998 monitoring produces findings which raise a research issue the study of which would 999 go beyond routine surveillance), the epidemiologists should seek individual consent 1000 for the use of the data or demonstrate that the research meets one of the other 1001 conditions for waiving informed consent, as explained in this Commentary.

Even 1002 when individual consent is not required, the usual expectations of risk minimization, 1003 protection of confidentiality, and compliance with all other legal requirements still 1004 apply. 1005 1006 d. Studies using health-related registries. The creation and maintenance of health-1007 related registries (e.g., cancer registries, databanks of genetic and other anomalies in 1008 newborn babies, etc.) provide a major resource for many public health activities, from 1009 disease prevention to resource allocation. Several considerations support the common 1010 practice of requiring that all practitioners submit relevant data to such registries: the 1011 importance of having comprehensive information to provide accurate information 1012 about an entire population; the scientific need to include all cases in order to avoid 1013 undetectable selection bias; and the general ethical principle that burdens and benefits 1014 should be distributed equitably across the population.

Hence, registries that are 1015 established or officially recognized by governmental authorities usually involve 1016 mandatory rather than voluntary collection of data. 1017 Studies using data from such registries (as well as studies that link data from 1018 several registries or that combine registry-data with information from publicly 1019 available sources) thus involve the use of data that have been compiled without the 1020 informed consent of the individuals involved. Such studies should be submitted to an 1021 25 ethical review committee and permission should also be sought from the competent 1022 authority that is legally responsible for the maintenance and use of the registry.

When 1023 an investigator plans to contact persons based on their inclusion in the registry (e.g., 1024 to obtain from them additional information for research purposes beyond the data 1025 supplied by the registry), the investigator should bear in mind that these persons may 1026 be unaware that their data were submitted to the registry or unfamiliar with the 1027 process by which investigators obtain access to the data. Investigators are cautioned to 1028 ensure that their access to the registry information is appropriately explained to the 1029 potential research subjects by the people who run the registry or other public 1030 authorities, preferably before the investigators approach the subjects.

1031 1032 e. Cluster-randomized trials. Epidemiological research can take the form of a trial in 1033 which an intervention is targeted to a whole group of people such as all the students in 1034 a school or all residents of a community, and in which the groups–rather than the 1035 individuals within the group–are randomly assigned to the different arms of the trial. 1036 Examples include a vaccination campaign applied at the school level, fluoridation of 1037 the drinking-water supply to prevent dental caries, a change in healthcare 1038 reimbursement policies, or a change in incineration practices at local waste disposal 1039 sites.

In a cluster-randomized trial, individual persons usually do not have an 1040 opportunity to consent to the study itself but should typically still be made aware that 1041 it will take place. Depending upon the way the study is conducted, individuals may or 1042 may not be able to decline participation in the study. For example, parents could 1043 consent or not consent to their child's vaccination at school or a person could decide 1044 to drink bottled water rather than use water that may be fluoridated; conversely, it 1045 would be difficult for a person to change the air he or she breathed outside in a study 1046 comparing methods of incinerating waste, or for a person to move to a different 1047 jurisdiction where the experimental method for healthcare reimbursement is not being 1048 tested.

As in all studies, investigators have a responsibility to describe in the protocol 1049 the information that will be provided to the decision-makers and to individuals within 1050 the clusters. 1051 In a cluster-randomized trial, the investigator should identify an appropriate 1052 person or body (e.g., a community leader, headmaster, or local health council) that has 1053 authority to give permission for the cluster to participate in the study and to be 1054 26 assigned on a random basis to one arm or another of the study. While this decision-1055 maker may not have been appointed or elected for the specific purpose of giving 1056 permission for the cluster to participate in research, the scope of authority should 1057 encompass interventions of the type in question if provided outside of a research 1058 project; moreover, the decision-maker should ensure that the risks of participation in 1059 the study and the randomization are commensurate with the benefits for the cluster or 1060 for society.

The decision-maker may choose to consult a wider group of community 1061 representatives or advisers before taking the decision to permit the study. 1062 1063 Guideline 5 1064 Obtaining informed consent: 1065 Essential information for prospective research subjects 1066 1067 Before requesting an individual's consent to participate in research, the 1068 investigator must provide the following information, in language or another 1069 form of communication that the individual can understand: 1070 1071 1) that the individual is invited to participate in research, the 1072 reasons for considering the individual suitable for the research, and that 1073 participation is voluntary;

1074 1075 2) that the individual is free to refuse to participate and will be free 1076 to withdraw from the research at any time without penalty or loss of 1077 benefits to which he or she would otherwise be entitled; 1078 1079 3) the purpose of the research, the procedures to be carried out by 1080 the investigator and the subject, and an explanation of how the research 1081 differs from routine medical care; 1082 1083 4) for controlled trials, an explanation of features of the research 1084 design (e.g., randomization, double-blinding), and that the subject will not 1085 be told of the assigned treatment until the study has been completed and 1086 the blind has been broken;

1087 1088 27 5) the expected duration of the individual's participation (including 1089 number and duration of visits to the research centre and the total time 1090 involved) and the possibility of early termination of the trial or of the 1091 individual’s participation in it; 1092 1093 6) whether money or other forms of material goods will be provided 1094 in return for the individual's participation and, if so, the kind and amount; 1095 1096 7) that, after the completion of the study, subjects will be informed 1097 of the findings of the research in general, and individual subjects will be 1098 informed of any finding that relates to their particular health status;

1099 1100 8) that subjects have the right of access to their data on demand, 1101 even if these data lack immediate clinical utility (unless the ethical review 1102 committee has approved temporary or permanent non-disclosure of data, in 1103 which case the subject should be informed of, and given, the reasons for 1104 such non-disclosure); 1105 1106 9) any foreseeable risks, pain or discomfort, or inconvenience to 1107 the individual (or others) associated with participation in the research, 1108 including risks to the health or well-being of a subject’s spouse or partner; 1109 1110 10) the direct benefits, if any, expected to result to subjects from 1111 participating in the research;

1112 1113 11) the expected benefits of the research to the community or to 1114 society at large, or contributions to scientific knowledge; 1115 1116 12) whether, when and how any products or interventions proven by 1117 the research to be safe and effective will be made available to subjects after 1118 they have completed their participation in the research, and whether they 1119 will be expected to pay for them; 1120 1121 13) any currently available alternative interventions or courses of 1122 treatment; 1123 1124 28 14) the provisions that will be made to ensure respect for the privacy 1125 of subjects and for the confidentiality of records in which subjects are 1126 identified;

1127 1128 15) the limits, legal or other, to the investigators' ability to safeguard 1129 confidentiality, and the possible consequences of breaches of 1130 confidentiality; 1131 1132 16) policy with regard to the use of results of genetic tests and 1133 familial genetic information, and the precautions in place to prevent 1134 disclosure of the results of a subject's genetic tests to immediate family 1135 relatives or to others (e.g., insurance companies or employers) without the 1136 consent of the subject; 1137 1138 17) the sponsors of the research, the institutional affiliation of the 1139 investigators, and the nature and sources of funding for the research;

1140 1141 18) the possible research uses, direct or secondary, of the subject’s 1142 medical records and of biological specimens taken in the course of clinical 1143 care (See also Guidelines 4 and 18 Commentaries); 1144 1145 19) whether it is planned that biological specimens collected in the 1146 research will be destroyed at its conclusion, and, if not, details about their 1147 storage (where, how, for how long, and final disposition) and possible 1148 future use, and that subjects have the right to decide about such future use, 1149 to refuse storage, and to have the material destroyed (See Guideline 4 1150 Commentary);

1151 1152 20) whether commercial products may be developed from biological 1153 specimens, and whether the participant will receive monetary or other 1154 benefits from the development of such products; 1155 1156 21) whether the investigator is serving only as an investigator or as 1157 both investigator and the subject’s physician; 1158 1159 22) the extent of the investigator's responsibility to provide medical 1160 29 services to the participant; 1161 1162 23) that treatment will be provided free of charge for specified types 1163 of research-related injury or for complications associated with the 1164 research, the nature and duration of such care, the name of the 1165 organization or individual that will provide the treatment, and whether there 1166 is any uncertainty regarding funding of such treatment;

1167 1168 24) in what way, and by what organization, the subject or the 1169 subject's family or dependants will be compensated for disability or death 1170 resulting from such injury (or, when indicated, that there are no plans to 1171 provide such compensation); 1172 1173 25) whether or not, in the country in which the prospective subject 1174 is invited to participate in research, the right to compensation is legally 1175 guaranteed; 1176 1177 26) that an ethical review committee has approved or cleared the 1178 research protocol. 1179 1180 Commentary on Guideline 5 1181 1182 The points specified in this Guideline–which were developed in the context of 1183 biomedical research–are generally relevant when obtaining informed consent for 1184 interventional research (especially population studies of drugs and devices) but are 1185 not all required in most observational studies.

(In particular, items 4, 12, 13, and 21-1186 24 are unlikely to be relevant.) Depending on the specifics of the study design, the 1187 investigator will need to justify to the ethical review committee why any particular 1188 items have been omitted from the consent process; a committee may, of course, 1189 decide that the researcher should be encouraged, as a prudential matter, to include 1190 some points that are not strictly speaking required. Alternatively, an ethical review 1191 committee may wish to provide investigators with a shorter list of points to be 1192 addressed in the consent process for observational studies.

1193 30 Some of the points specified in this Guideline present special problems in the 1194 context of epidemiological research. The statement in Item #2 that individuals are 1195 “free to withdraw from the research at any time” rests on the principle that it is 1196 ethically unacceptable to force a person to participate in research. In epidemiological 1197 studies, a person‟s “withdrawal” from research can take several forms. The first, 1198 which is a subject‟s request that the gathering of new data about the subject cease 1199 (e.g., in a longitudinal study), must be honored, just as any other withdrawal from on-1200 going participation in a study should be.

The second could involve a request that the 1201 person‟s data (and perhaps biological materials) be removed from a database and/or 1202 repository. Such removal may be very difficult (or impossible if the data have been 1203 anonymized), would risk undermining the validity of studies using the database, and 1204 would typically seem disproportional to the individual‟s interest, since–unlike an on-1205 going intervention study–the person does not bear any burden at present. If an 1206 investigator, with the approval of the ethical review committee, does not intend to 1207 honor requests to remove data and/or biological samples, this policy should be clearly 1208 stated in the consent document.

1209 Item #7 requires two things: that subjects as a group be informed about the 1210 general findings of a study and that individuals be informed about any test results or 1211 other findings relevant to their personal health status. As noted in the Commentary to 1212 Guideline 4, when a study employs anonymization, which makes it impossible to 1213 notify individuals (and, in some cases, even identifiable groups of subjects) of 1214 research findings or personal test results, the ethical review committee should take 1215 this into account in deciding whether to approve the study. Even when they have not 1216 anonymized the data, epidemiologists have often not notified individual subjects of 1217 test results.

In light of contemporary standards for informed consent, however, 1218 epidemiologists should make subjects aware of findings that are clinically relevant to 1219 their individual health. When (e.g., because of the scale of a particular study) an 1220 investigator does not plan to do so, he or she must obtain approval from the ethical 1221 review committee. In all cases, the extent to which findings will be disclosed to 1222 subjects as a group or as individuals should be clearly conveyed in the informed 1223 consent material. 1224 1225 1226 31 Guideline 6 1227 Obtaining informed consent: 1228 Obligations of investigators and sponsors 1229 1230 Investigators have a duty to: 1231 1232 – refrain from unjustified deception, undue influence, or intimidation;

1233 1234 – seek consent only after ascertaining that the prospective subject 1235 has adequate understanding of the relevant facts and of the 1236 consequences of participation and has had sufficient opportunity to 1237 consider whether to participate; 1238 1239 – when individual consent is required, obtain from each prospective 1240 subject a signed form as evidence of informed consent–1241 investigators should justify any exceptions to this general rule and 1242 obtain the approval of the ethical review committee (See Guideline 4 1243 Commentary, Documentation of consent); 1244 1245 – renew the informed consent of each subject if there are significant 1246 changes in the conditions or procedures of the research or if new 1247 information becomes available that could affect the willingness of 1248 subjects to continue to participate; and, 1249 1250 – renew the informed consent of each subject in long-term studies at 1251 pre-determined intervals, even if there are no changes in the design 1252 or objectives of the research.

1253 1254 The principal investigator has a non-delegable duty to ensure that all 1255 personnel working on the study comply with this Guideline. 1256 1257 Sponsors have a duty to ensure that these obligations are fulfilled. 1258 1259 Commentary on Guideline 6 1260 1261 32 The investigator is responsible for ensuring the adequacy of informed consent 1262 from each subject, whether the investigator undertakes this task or delegates it to other 1263 members of the research team. The person obtaining informed consent should be 1264 knowledgeable about the research and capable of answering questions from 1265 prospective subjects.

Investigators in charge of the study must make themselves 1266 available to answer questions at the request of subjects. Any restrictions on the 1267 subject's opportunity to ask questions and receive answers before or during the 1268 research undermines the validity of the informed consent. 1269 1270 Consent by subjects enrolled by mail or electronic means. In some epidemiological 1271 studies, no face-to-face contact occurs between investigators and subjects. For 1272 example, subjects may be asked to provide electronic authorization for the use of their 1273 personal data in a study, or subjects may be asked to complete a questionnaire over 1274 the Internet.

When subjects are enrolled in such studies by mail or electronic means 1275 (e.g., e-mail, Internet, etc.), difficulties may arise in fulfilling investigators‟ duties to 1276 ensure that subjects are able to receive answers to any questions and to ascertain that 1277 subjects adequately understand relevant facts. Potential subjects enrolled in these 1278 ways should therefore be given a means (such as a toll-free phone number or email 1279 address) to enable them to pose questions to, and receive answers from, the research 1280 team concerning the study. Since investigators may not have direct contact, through 1281 such means of communication, with all potential subjects, it is especially important 1282 that the materials used for mail enrolment are worded carefully to maximize the 1283 chances that the subjects enrolled will have an adequate understanding of information 1284 relevant to their participation in the study.

(see also Guideline 23) 1285 1286 Withholding information and deception. Sometimes, to ensure the validity of research, 1287 investigators withhold certain information in the consent process. For example, when 1288 tests will be performed to monitor subjects‟ compliance with a protocol, subjects may 1289 not be told the purpose of the testing, since if they knew their compliance was being 1290 monitored they might modify their behaviour and hence invalidate results. In most 1291 such cases, the prospective subjects are asked to consent to remain uninformed of the 1292 purpose of some procedures until the research is completed; after the conclusion of 1293 the study they are given the omitted information.

More generally, when providing 1294 33 certain information (e.g., regarding the study hypothesis) would jeopardize the 1295 validity of the research, subjects are sometimes not even told that some information 1296 has been withheld until after the research has been completed. Any withholding of 1297 information, and the procedures used to provide information subsequently, must 1298 receive the explicit approval of the ethical review committee based on the necessity of 1299 the withholding, the minimization of attendant risks to subjects, and the adequacy of 1300 the procedures for “debriefing” subjects after their participation in the study.

1301 Active deception of subjects is considerably more controversial than simply 1302 withholding certain information. Lying to subjects is a tactic not commonly employed 1303 in biomedical research. Social and behavioural scientists, however, sometimes 1304 deliberately misinform subjects to study their attitudes and behaviour. For example, 1305 scientists have pretended to be patients to study the behaviour of health-care 1306 professionals and patients in their natural settings. 1307 Some people maintain that active deception is never permissible. Others 1308 would permit it in certain circumstances.

Deception is not permissible, however, in 1309 cases in which the deception itself would disguise the possibility of the subject being 1310 exposed to more than minimal risk. When deception is deemed indispensable to the 1311 methods of a study, the investigators must demonstrate to an ethical review committee 1312 that no other research method would suffice; that significant advances could result 1313 from the research; and that nothing has been withheld that, if divulged, would cause a 1314 reasonable person to refuse to participate. The ethical review committee should 1315 determine the consequences for the subject of being deceived, and whether and how 1316 deceived subjects should be informed of the deception upon completion of the 1317 research.

Such informing, commonly called "debriefing", ordinarily entails explaining 1318 the reasons for the deception. A subject who disapproves of having been deceived 1319 should be offered an opportunity to refuse to allow the investigator to use information 1320 thus obtained. Investigators and ethical review committees should be aware that 1321 deceiving research subjects may wrong them as well as harm them; subjects may 1322 resent not having been informed when they learn that they have participated in a study 1323 under false pretences. In some studies there may be justification for deceiving persons 1324 other than the subjects by either withholding or disguising elements of information.

1325 Such tactics are often proposed, for example, for studies of the abuse of spouses or 1326 children. An ethical review committee must review and approve all proposals to 1327 34 deceive persons other than the subjects. Subjects are entitled to prompt and honest 1328 answers to their questions; the ethical review committee must determine for each 1329 study whether others who are to be deceived are similarly entitled. 1330 1331 Intimidation and undue influence. Intimidation in any form invalidates informed 1332 consent. Prospective subjects who are patients often depend for medical care upon the 1333 physician/investigator, who consequently has a certain credibility in their eyes, and 1334 whose influence over them may be considerable, particularly if the study protocol has 1335 a therapeutic component.

They may fear, for example, that refusal to participate 1336 would damage the therapeutic relationship or result in the withholding of health 1337 services. The physician/investigator must assure them that their decision on whether 1338 to participate will not affect the therapeutic relationship or other benefits to which 1339 they are entitled. In this situation the ethical review committee should consider 1340 whether a neutral third party should seek informed consent. 1341 The prospective subject must not be exposed to undue influence. The borderline 1342 between justifiable persuasion and undue influence is imprecise, however.

The 1343 researcher should give no unjustifiable assurances about the benefits, risks or 1344 inconveniences of the research, for example, or induce a close relative or a 1345 community leader to influence a prospective subject's decision. (See also Guideline 4: 1346 Individual informed consent.) 1347 1348 Risks. Investigators should be as objective as possible in discussing the details of the 1349 experimental intervention, the pain and discomfort that it may entail, and known risks 1350 and possible hazards. In complex research projects it may be neither feasible nor 1351 desirable to inform prospective participants fully about every possible risk.

They 1352 must, however, be informed of all risks that a „reasonable person‟ would consider 1353 material to making a decision about whether to participate, including risks to a spouse 1354 or partner associated with trials of, for example, psychotropic or genital-tract 1355 medicaments. (See also Guideline 8 Commentary, Risks to groups of persons.) 1356 1357 Exception to the requirement of informed consent for interventional studies to include 1358 persons rendered incapable of informed consent by an acute condition. Certain 1359 persons with acute conditions that render them incapable of giving informed consent 1360 35 may be eligible for inclusion in a study concerning an acute condition in which the 1361 majority of prospective subjects will be capable of informed consent, and in which the 1362 investigational intervention would hold out the prospect of direct benefit and would 1363 be justified accordingly.

When the investigation involves certain procedures or 1364 interventions that would not be of direct benefit yet carry no more than minimal risk 1365 (such as the collection of additional blood for research purposes), the initial protocol 1366 submitted for approval to the ethical review committee should anticipate that some 1367 persons may be incapable of consent, and should propose for such patients a form of 1368 proxy consent, such as permission of the responsible relative. When the ethical review 1369 committee has approved or cleared such a protocol, an investigator may seek the 1370 permission of the responsible relative and enrol such a person.

1371 1372 Guideline 7 1373 Compensation for participation 1374 1375 Subjects may be reimbursed for lost earnings, travel costs and other expenses 1376 incurred in taking part in a study; they may also receive free medical services. 1377 Subjects, particularly those who receive no direct benefit from research, may 1378 also be paid or otherwise compensated for inconvenience and time spent. The 1379 payments should not be so large, however, or the medical services so 1380 extensive as to induce prospective subjects to consent to participate in the 1381 research against their better judgment ("undue inducement").

All payments, 1382 reimbursements and medical services provided to research subjects must 1383 have been approved by an ethical review committee. 1384 1385 Commentary on Guideline 7 1386 1387 Acceptable recompense. Research subjects may be reimbursed for their transport and 1388 other expenses, including lost earnings, associated with their participation in research. 1389 Those who receive no direct benefit from the research may also receive a small sum 1390 of money for inconvenience due to their participation in the research. All subjects 1391 may receive medical services unrelated to the research and have procedures and tests 1392 performed free of charge.

1393 36 1394 Unacceptable recompense. Payments in money or in kind to research subjects should 1395 not be so large as to persuade them to take undue risks or volunteer against their better 1396 judgment. Payments or rewards that undermine a person's capacity to exercise free 1397 choice invalidate consent. It may be difficult to distinguish between suitable 1398 recompense and undue influence to participate in research. An unemployed person or 1399 a student may view promised recompense differently from an employed person. 1400 Someone without access to medical care may or may not be unduly influenced to 1401 participate in research simply to receive such care.

A prospective subject may be 1402 induced to participate in order to obtain a better diagnosis or access to a drug not 1403 otherwise available; local ethical review committees may find such inducements 1404 acceptable. Monetary and in-kind recompense must, therefore, be evaluated in the 1405 light of the traditions of the particular culture and population in which they are 1406 offered, to determine whether they constitute undue influence. The ethical review 1407 committee will ordinarily be the best judge of what constitutes reasonable material 1408 recompense in particular circumstances. When research interventions or procedures 1409 that do not hold out the prospect of direct benefit present more than minimal risk, all 1410 parties involved in the research–sponsors, investigators and ethical review committees 1411 –in both funding and host countries should be careful to avoid undue material 1412 inducement.

1413 1414 Incompetent persons. Incompetent persons may be vulnerable to exploitation for 1415 financial gain by guardians. A guardian asked to give permission on behalf of an 1416 incompetent person should be offered no recompense other than a refund of travel and 1417 related expenses. 1418 1419 Withdrawal from a study. A subject who withdraws from research for reasons related 1420 to the study, such as unacceptable side-effects of the intervention being studied, or 1421 who is withdrawn on health grounds, should be paid or recompensed as if full 1422 participation had taken place. A subject who withdraws for any other reason should be 1423 paid in proportion to the amount of participation.

An investigator who must remove a 1424 subject from the study for wilful noncompliance is entitled to withhold part or all of 1425 the payment. 1426 37 1427 Guideline 8 1428 Benefits, harms and risks of study participation 1429 1430 For all epidemiological research involving human subjects, the investigator 1431 must ensure that potential benefits and harms are reasonably balanced and 1432 risks are minimized. 1433 1434  Interventions or procedures that hold out the prospect of direct 1435 diagnostic, therapeutic or preventive benefit for the individual subject must 1436 be justified by the expectation that they will be at least as advantageous to 1437 the individual subject, in the light of foreseeable harms and benefits, as any 1438 available alternative.

Risks of such 'beneficial' interventions or procedures 1439 must be justified in relation to expected benefits to the individual subject. 1440 1441  Risks of interventions that do not hold out the prospect of direct 1442 diagnostic, therapeutic or preventive benefit for the individual must be 1443 justified in relation to the expected benefits to society. The risks presented 1444 by such interventions must be reasonable in relation to the importance of 1445 the knowledge to be gained. 1446 1447 Commentary on Guideline 8 1448 1449 Ethical grounding. The Declaration of Helsinki in several paragraphs deals with the 1450 well-being of research subjects and the avoidance of harm, specifically: 1451 considerations related to the well-being of the human subject should take precedence 1452 over the interests of science and society (Paragraph 5); clinical testing must be 1453 preceded by adequate laboratory or animal experimentation to demonstrate a 1454 reasonable probability of success without undue risk (Paragraph 11); every project 1455 should be preceded by careful assessment of predictable harms and burdens in 1456 comparison with foreseeable benefits to the subject or to others (Paragraph 16);

1457 physician-researchers must be confident that the risks involved have been adequately 1458 assessed and can be satisfactorily managed (Paragraph 17); and the harms and 1459 burdens to the subject must be minimized, and reasonable in relation to the 1460 38 importance of the objective or the knowledge to be gained (Paragraph 18). 1461 Epidemiological studies may employ a variety of interventions of which some 1462 hold out the prospect of direct therapeutic benefit (beneficial interventions) and others 1463 are administered solely to answer the research question (non-beneficial interventions).

1464 Beneficial interventions are justified as they are in medical practice by the expectation 1465 that they will be at least as advantageous to the individuals concerned, in the light of 1466 both harms and benefits, as any available alternative. Non-beneficial interventions are 1467 assessed differently; they may be justified only by appeal to the knowledge to be 1468 gained, either “generalizable knowledge” (the usual objective of a research project) or 1469 more particularized findings, of use for example by public health officials. 1470 Paragraphs 5 and 18 of the Declaration of Helsinki do not preclude well-1471 informed volunteers, capable of fully appreciating risks and benefits of an 1472 investigation, from participating in research for altruistic reasons or for modest 1473 remuneration.

1474 1475 Experimental studies of preventive interventions. Epidemiologists carry out 1476 experimental studies, in particular randomized population trials, usually to test 1477 preventive intervention programmes, for instance administration of a vaccine or a 1478 drug, or an organized screening programme. Because they involve the totality of a 1479 population or a relevant segment of it (e.g., those believed to be at higher risk of the 1480 target disease), these interventions imply that everybody will be submitted to 1481 whatever inconvenience and potential harm the intervention entails, while only the 1482 minority (often, comparatively small), namely those who would have actually 1483 developed the disease, get the benefit of avoiding it thanks to the intervention (the 1484 same applies, although to a lesser degree, even when the intervention is concentrated 1485 on so-called "high risk" groups).

This is an inherent problem of preventive 1486 programmes; both investigators and ethical review committees need to carefully 1487 weigh the potential harm and inconvenience to programme participants who may not 1488 receive any benefit from the programme, and the participants must receive clear and 1489 full information before giving their consent. Likewise, research may be conducted on 1490 a screening programme for a condition for which no effective treatment exists 1491 provided the results could be of sufficient direct relevance to the health of people 1492 39 other than the participants in the study (for example, in developing a programme for a 1493 transmissible disease, whether contagious or genetic).

1494 1495 Minimizing risk associated with participation in a randomized study. In a randomized 1496 controlled study subjects risk being allocated to receive the intervention that proves 1497 inferior. They are allocated by chance to one of two or more intervention arms and 1498 followed to a predetermined end-point. (Interventions are understood to include new 1499 or established therapies, diagnostic tests and preventive measures.) An intervention is 1500 evaluated by comparing it with another intervention (a control), which is ordinarily 1501 the best current method, selected from the safe and effective treatments available 1502 globally, unless some other control intervention such as placebo can be justified 1503 ethically (see Guideline 11).

1504 To minimize risk when the intervention to be tested is designed to prevent or 1505 postpone a lethal or disabling outcome, the investigator must not, for experimental 1506 purposes, withhold therapy that is known to be superior to the intervention being 1507 tested, unless the withholding can be justified by the standards set forth in Guideline 1508 11. Also, the investigator must provide in the research protocol for the monitoring of 1509 research data by an independent board (Data and Safety Monitoring Board); one 1510 function of such a board is to protect the research subjects from previously unknown 1511 adverse reactions or unnecessarily prolonged exposure to an inferior therapy.

1512 Normally at the outset of a randomized controlled study, criteria are established for its 1513 premature termination (stopping rules or guidelines). 1514 1515 Risks to groups of persons. In order to achieve the social benefits anticipated from 1516 conducting research, results should be made public. Sometimes, however, research in 1517 epidemiology (as well as such other fields as genetics and sociology) may present 1518 risks to the interests of communities, societies, or racially or ethnically defined 1519 groups. Information might be published that could stigmatize a group or expose its 1520 members to discrimination.

Such information, for example, could indicate that the 1521 group has a higher than average prevalence of alcoholism, mental illness or sexually 1522 transmitted disease, or is particularly susceptible to certain genetic disorders. Plans to 1523 conduct such research should be sensitive to such considerations, to the need to 1524 maintain confidentiality during and after the study, and to the need to publish the 1525 40 resulting data in a manner that is respectful of the interests of all concerned, or in 1526 certain exceptional circumstances not to publish them. The ethical review committee 1527 should ensure that the interests of all concerned are given due consideration; often it 1528 will be advisable to have individual consent supplemented by community 1529 consultation.

In assessing the harms and benefits that a protocol presents to a 1530 population, it is appropriate to consider the harm that could result from forgoing the 1531 research or from failing to publish the results. 1532 1533 [The ethical basis for the justification of risk is elaborated further in 1534 Guideline 9] 1535 1536 Guideline 9 1537 Special limitations on risk when research involves individuals 1538 who are not capable of giving informed consent 1539 1540 When there is ethical and scientific justification to conduct research with 1541 individuals incapable of giving informed consent, the potential harm from any 1542 research intervention that does not hold out the prospect of direct benefit for 1543 the individual subject should not be more than minimal.

1544 1545 Commentary on Guideline 9 1546 1547 The minimal-risk standard. Certain individuals or groups may have limited or no 1548 capacity to give informed consent either because, as in the case of prisoners, their 1549 autonomy is limited, or because they have limited cognitive capacity. Research 1550 involving such persons that does not aim to benefit them directly may occur only 1551 when its potential risks are found to be no more than minimal. 1552 In addition, the ethical review committee must find: 1) that the research is 1553 designed to be responsive to the disease or condition affecting the prospective 1554 subjects or to conditions to which they are particularly susceptible;

2) that the 1555 objective of the research is sufficiently important to justify exposure of the subjects to 1556 the increased risk; and 3) that the interventions are reasonably comparable to the 1557 clinical interventions that the subjects have experienced or may be expected to 1558 41 experience in relation to the condition under investigation. The requirement that the 1559 research interventions be reasonably comparable is intended to enable the subjects to 1560 draw on personal experience as they decide whether to accept or reject additional 1561 procedures for research purposes. Their choices will, therefore, be more informed 1562 even though they may not fully meet the standard of informed consent.

1563 1564 Consent required when subject becomes capable of informed consent. If such research 1565 subjects, including children, become capable of giving independent informed consent 1566 during the research, their consent to continued participation should be obtained. 1567 1568 (See also Guidelines 4, 13, 14 and 15.) 1569 1570 Guideline 10 1571 Research in populations and communities 1572 with limited resources 1573 1574 Before undertaking research in a population or community with limited 1575 resources, the sponsor and the investigator must make every effort to ensure 1576 that: 1577 1578 - the research is responsive to the health needs and the priorities of the 1579 population or community in which it is to be carried out; and 1580 1581 - any intervention or product developed, or knowledge generated, will be 1582 made reasonably available for the benefit of that population or community.

1583 1584 Commentary on Guideline 10 1585 1586 This Guideline is concerned with countries or communities in which resources 1587 are limited to the extent that the population is, or may be, vulnerable to exploitation 1588 by sponsors and investigators from the relatively wealthy countries and communities. 1589 This concern, which has arisen principally from experiences with clinical trials of new 1590 drugs, should not stand in the way of carrying out ethically sound epidemiological 1591 42 studies in resource-limited setting. Such studies are, almost by nature, relevant to the 1592 health of the populations or communities in which they are conducted, and the 1593 information gathered in such studies can often be very important for improving 1594 population health in resource-poor countries and communities.

1595 1596 Responsiveness of research to health needs and priorities. To meet the ethical 1597 requirement that research be responsive to the health needs of the population or 1598 community in which it is carried, it is not sufficient simply to determine that a disease 1599 is prevalent in the population and that new or further research is needed: the ethical 1600 requirement of “responsiveness” can be fulfilled only if successful interventions or 1601 other kinds of health benefit are made available to the population. This is applicable 1602 especially to research conducted in countries where governments lack the resources to 1603 make such products or benefits widely available.

Even when a product to be tested in 1604 a particular country is much cheaper than the standard treatment in some other 1605 countries, the government or individuals in that country may still be unable to afford 1606 it. If the knowledge gained from the research in such a country is used primarily for 1607 the benefit of populations that can afford the tested product, the research may rightly 1608 be characterized as exploitative and, therefore, unethical. 1609 When an investigational intervention has important potential for health care in 1610 the host country, the negotiation that the sponsor should undertake to determine the 1611 practical implications of "responsiveness", as well as "reasonable availability", should 1612 include representatives of stakeholders in the host country; these include the national 1613 government, the health ministry, local health authorities, and concerned scientific and 1614 ethics groups, as well as representatives of the communities from which subjects are 1615 drawn and non-governmental organizations such as health advocacy groups.

The 1616 negotiation should cover the health-care infrastructure required for safe and rational 1617 use of the intervention, the likelihood of authorization for distribution, and decisions 1618 regarding payments, royalties, subsidies, technology and intellectual property, as well 1619 as distribution costs, when this economic information is not proprietary. In some 1620 cases, satisfactory discussion of the availability and distribution of successful 1621 products will necessarily engage international organizations, donor governments and 1622 bilateral agencies, international nongovernmental organizations, and the private 1623 sector.

The development of a health-care infrastructure should be facilitated at the 1624 43 onset so that it can be of use during and beyond the conduct of the research. 1625 Additionally, if an investigational intervention has been shown to be beneficial, 1626 the sponsor should continue to provide it to the subjects after the conclusion of the 1627 study and pending its approval by a drug regulatory authority, when relevant. The 1628 sponsor is unlikely to be in a position to make a beneficial investigational intervention 1629 generally available to the community or population until some time after the 1630 conclusion of the study, as it may be in short supply and in any case cannot be made 1631 generally available before a drug regulatory authority has approved it.

1632 When a study‟s expected outcome is scientific knowledge rather than a 1633 commercial product, such planning or negotiation is rarely, if ever, needed. There 1634 must be assurance, however, that the scientific knowledge developed will be used for 1635 the benefit of the population. 1636 1637 Reasonable availability. The issue of "reasonable availability" is complex and will 1638 need to be determined on a case-by-case basis. Relevant considerations include the 1639 length of time for which the intervention or product developed, or other agreed 1640 benefit, will be made available to research subjects, or to the community or 1641 population concerned; the severity of a subject‟s medical condition; the effect of 1642 withdrawing the study drug (e.g., death of a subject); the cost to the subject or health 1643 service; and the question of undue inducement if an intervention is provided free of 1644 charge.

1645 In general, if there is good reason to believe that a product developed or 1646 knowledge generated by research is unlikely to be reasonably available to, or applied 1647 to the benefit of, the population of a proposed host country or community after the 1648 conclusion of the research, it is unethical to conduct the research in that country or 1649 community. This should not be construed as precluding studies designed to evaluate 1650 novel therapeutic concepts. As a rare exception, for example, research may be 1651 designed to obtain preliminary evidence that a drug or a class of drugs has a beneficial 1652 effect in the treatment of a disease that occurs only in regions with extremely limited 1653 resources, and it could not be carried out reasonably well in more developed 1654 communities.

Such research may be justified ethically even if there is no plan in place 1655 to make a product available to the population of the host country or community at the 1656 conclusion of the preliminary phase of its development. If the concept is found to be 1657 44 valid, subsequent phases of the research could result in a product that could be made 1658 reasonably available at its conclusion. 1659 1660 (See also Guidelines 3,12, 20 and 21.) 1661 1662 Guideline 11 1663 Choice of control in clinical trials 1664 1665 As a general rule, research subjects in the control group of a trial of a 1666 diagnostic, therapeutic, or preventive intervention should receive an 1667 established effective intervention.

In some circumstances it may be ethically 1668 acceptable to use an alternative comparator, such as placebo or "no 1669 treatment". 1670 1671 Placebos may be used: 1672 1673 - when there is no established effective intervention; 1674 1675 - when withholding an established effective intervention would expose 1676 subjects to, at most, temporary discomfort or delay in relief of symptoms; 1677 1678 - when use of an established effective intervention as comparator would 1679 not yield scientifically reliable results and use of placebo would not add 1680 any risk of serious or irreversible harm to the subjects.

1681 1682 Commentary on Guideline 11 1683 1684 The controversies that have arisen concerning placebo controls have centred 1685 largely on clinical trials of new drugs undertaken in resource-poor countries by 1686 investigators from resource-rich countries. Nonetheless, ethical issues can also arise 1687 when placebos are proposed as part of the design of interventional studies undertaken 1688 by epidemiologists. 1689 1690 General considerations for controlled clinical trials. The design of trials of 1691 45 investigational diagnostic, therapeutic or preventive interventions raises interrelated 1692 scientific and ethical issues for sponsors, investigators and ethical review committees.

1693 To obtain reliable results, investigators must compare the effects of an investigational 1694 intervention on subjects assigned to the investigational arm (or arms) of a trial with 1695 the effects that a control intervention produces in subjects drawn from the same 1696 population and assigned to its control arm. Although randomization is the preferred 1697 method for assigning subjects to the various arms of a clinical trial, non-experimental 1698 methods, such as cohort and case-control studies to evaluate drugs and devices, may 1699 often be justified scientifically and ethically. Assignment to treatment arms by 1700 randomization, in addition to its usual scientific superiority, offers the advantage of 1701 tending to render equivalent to all subjects the foreseeable benefits and risks of 1702 participation in a trial.

1703 A clinical trial cannot be justified ethically unless it is capable of producing 1704 scientifically reliable results. When the objective is to establish the effectiveness and 1705 safety of an investigational intervention, the use of a placebo control is sometimes 1706 much more likely than the use of an active control to produce a scientifically reliable 1707 result. In many cases the ability of a trial to distinguish effective from ineffective 1708 interventions (its assay sensitivity) cannot be assured unless the control is a placebo. 1709 If, however, an effect of using a placebo would be to deprive subjects in the control 1710 arm of an established effective intervention, and thereby to expose them to serious 1711 harm, particularly if it is irreversible, it would obviously be unethical to use a placebo.

1712 1713 Placebo control in the absence of a established effective alternative. The use of 1714 placebo in the control arm of a clinical trial is ethically acceptable when, as stated in 1715 the Declaration of Helsinki (Paragraph 29), “no proven prophylactic, diagnostic or 1716 therapeutic method exists.” Usually, in this case, a placebo is scientifically preferable 1717 to no intervention. In certain circumstances, however, an alternative design may be 1718 both scientifically and ethically acceptable, and preferable; for example, in certain 1719 vaccine trials an investigator might choose to provide for those in the control arm a 1720 vaccine that is unrelated to the investigational vaccine.

1721 1722 Placebo-controlled studies that entail only minor risks. A placebo-controlled design 1723 may be ethically acceptable, and preferable on scientific grounds, when the condition 1724 46 for which the intervention is being evaluated is only a small deviation in physiological 1725 measurements, such as slightly raised blood pressure or a modest increase in serum 1726 cholesterol, and if delaying or omitting an established effective intervention would 1727 cause only temporary discomfort (e.g., common headache) and no serious adverse 1728 consequences. Likewise, when the investigative intervention is aimed at a relatively 1729 trivial condition, such as the common cold or hair loss, and using a placebo for the 1730 duration of a trial would deprive control subjects of only minor benefits, it is not 1731 unethical to use a placebo-control design.

Even if it were possible to design a so-1732 called “non-inferiority”, or "equivalency", trial using an active control, it would still 1733 not be unethical in these circumstances to use a placebo-control design. The ethical 1734 acceptability of such placebo-controlled studies increases as the period of placebo use 1735 is decreased, and when the study design permits change to the active intervention 1736 (“escape treatment”) if intolerable symptoms occur. In any event, the ethical review 1737 committee must be fully satisfied that the risks of withholding such an intervention 1738 are truly minor and short-lived, that the safety and human rights of the subjects will be 1739 fully protected, that prospective subjects will be fully informed about alternative 1740 treatments, and that the purpose and design of the study are scientifically sound.

1741 1742 Placebo control when active control would not yield reliable results. Another 1743 rationale for using a placebo control rather than an established effective intervention 1744 is that the documented experience with the established effective intervention is not 1745 sufficient to provide a scientifically reliable comparison with the intervention being 1746 investigated; it is then difficult, or even impossible, without using a placebo, to design 1747 a scientifically reliable study. (When a researcher relies on this rationale, the ethical 1748 review committee has the option of seeking expert opinion as to whether use of an 1749 established effective intervention in the control arm would invalidate the results of the 1750 research.) This basis for depriving control subjects of an established effective 1751 intervention in clinical trials is, however, ethically acceptable only when doing so 1752 would not add any risk of serious, particularly irreversible, harm to the subjects.

In 1753 some cases, the condition at which the intervention is aimed (for example, cancer or 1754 HIV/AIDS) will be too serious to deprive control subjects of an established effective 1755 intervention. 1756 1757 47 An exception to this general rule is applicable in some studies designed to 1758 develop a therapeutic, preventive or diagnostic intervention for use in a country or 1759 community in which established effective interventions used in other countries are not 1760 available, and are unlikely to become available in the foreseeable future, for economic 1761 or logistic reasons, when the purpose of such a study is to make a potentially effective 1762 and affordable alternative available to the population.

An example might be an 1763 interventional epidemiological study of a simple method of water purification that 1764 could eliminate most pathogens responsible for serious disease in a country that is 1765 unable to afford more elaborate interventions which are effective in countries with 1766 greater resources. The scientific and ethical review committees must be satisfied that 1767 the established effective intervention cannot be used as comparator because its use 1768 would not yield scientifically reliable results that would be relevant to the health 1769 needs of the study population. This would be the case when existing data about the 1770 effectiveness and safety of the established effective intervention may have been 1771 accumulated under circumstances unlike those of the population in which it is 1772 proposed to conduct the trial (e.g., the disease or condition manifests itself differently 1773 in different populations, or other uncontrolled factors exist in the environment).

In 1774 these circumstances an ethical review committee may approve a clinical trial in which 1775 the comparator is a placebo or no treatment or a local remedy. The ethical 1776 acceptability of such a proposed investigational intervention depends upon its being 1777 responsive to the health needs of the population from which the research subjects 1778 would be recruited and upon there being assurance that, if it proves to be safe and 1779 effective, it will be made reasonably available to that population. 1780 1781 An "equivalency trial" as an alternative to a placebo-controlled trial. An 1782 alternative to a placebo-control design would be an “equivalency trial”, which would 1783 compare an investigational intervention with an established effective intervention.

1784 Equivalency trials are not designed to determine whether an investigational 1785 intervention is superior to an established effective one but rather whether it is, in 1786 effectiveness and safety, equivalent, or almost equivalent, to the latter; it would be 1787 hazardous to conclude, however, that an intervention that meets this equivalency 1788 standard is better than nothing or whatever intervention is available in the country 1789 simply because the intervention used as the control was itself previously shown to be 1790 48 better than a placebo, since there may be substantial differences between the results of 1791 superficially identical clinical trials carried out in different countries or at different 1792 times.

1793 1794 Means of minimizing harm to placebo-control subjects. Even when placebo 1795 controls are justified on one of the bases set forth in the Guideline, there are means of 1796 minimizing the possibly harmful effect of being in the control arm. 1797 First, a placebo-control group need not be untreated. An add-on design may be 1798 employed when the investigational therapy and a standard treatment have different 1799 mechanisms of action. The treatment to be tested and placebo are each added to a 1800 standard treatment. Such studies have a particular place when a standard treatment is 1801 known to decrease mortality or irreversible morbidity but a trial with standard 1802 treatment as the active control cannot be carried out or would be difficult to interpret 1803 [International Conference on Harmonisation (ICH) Guideline: Choice of Control 1804 Group and Related Issues in Clinical Trials, 2000].

In testing for improved treatment 1805 of life-threatening diseases such as cancer, HIV/AIDS, or heart failure, add-on 1806 designs are a particularly useful means of finding improvements in interventions that 1807 are not fully effective or may cause intolerable side-effects. They have a place also in 1808 respect of treatment for epilepsy, rheumatism and osteoporosis, for example, because 1809 withholding of established effective therapy could result in progressive disability, 1810 unacceptable discomfort or both. 1811 Second, as indicated in Guideline 8 Commentary, when the intervention to be 1812 tested in a randomized controlled trial is designed to prevent or postpone a lethal or 1813 disabling outcome, the investigator minimizes harmful effects of placebo-control 1814 studies by providing in the research protocol for the monitoring of research data by an 1815 independent Data and Safety Monitoring Board (DSMB).

One function of such a 1816 board is to protect the research subjects from previously unknown adverse reactions; 1817 another is to avoid unnecessarily prolonged exposure to an inferior intervention. The 1818 board fulfils the latter function by means of interim analyses of the data pertaining to 1819 efficacy to ensure that the trial does not continue beyond the point at which an 1820 investigational therapy is demonstrated to be effective. Normally, at the outset of a 1821 randomized controlled trial, criteria are established for its premature termination 1822 (stopping rules or guidelines).

1823 49 In some cases the DSMB is called upon to perform "conditional power 1824 calculations", designed to determine the probability that a particular clinical trial 1825 could ever show that the investigational therapy is effective. If that probability is very 1826 small, the DSMB is expected to recommend termination of the clinical trial, because 1827 it would be unethical to continue it beyond that point. 1828 In most cases of research involving human subjects, it is unnecessary to appoint 1829 a DSMB. To ensure that research is carefully monitored for the early detection of 1830 adverse events, the sponsor or the principal investigator appoints an individual to be 1831 responsible for advising on the need to consider changing the system of monitoring 1832 for adverse events or the process of informed consent, or even to consider terminating 1833 the study.

1834 1835 Guideline 12 1836 Equitable distribution of burdens and benefits in the 1837 selection of groups of subjects in research 1838 1839 Groups or communities to be invited to be subjects of research should be 1840 selected in such a way that the burdens and benefits of the research will be 1841 equitably distributed. The exclusion of groups or communities that might 1842 benefit from study participation must be justified. 1843 1844 Commentary on Guideline 12 1845 1846 General considerations: Equity requires that no group or class of persons should 1847 bear more than its fair share of the burdens of participation in research.

Similarly, no 1848 group should be deprived of its fair share of the benefits of research, short-term or 1849 long-term; such benefits include the direct benefits of participation as well as the 1850 benefits of the new knowledge that the research is designed to yield. When burdens or 1851 benefits of research are to be apportioned unequally among individuals or groups of 1852 persons, the criteria for unequal distribution should be morally justifiable and not 1853 arbitrary. In other words, unequal allocation must not be inequitable. Subjects should 1854 be drawn from the qualifying population in the general geographic area of the trial 1855 without regard to race, ethnicity, economic status or gender unless there is a sound 1856 50 scientific reason to do otherwise.

1857 In the past, groups of persons were excluded from participation in research for 1858 what were then considered good reasons. As a consequence of such exclusions, 1859 information about the diagnosis, prevention and treatment of diseases in such groups 1860 of persons is limited. This has resulted in a serious class injustice. If information 1861 about the management of diseases is considered a benefit that is distributed within a 1862 society, it is unjust to deprive groups of persons of that benefit. Such documents as 1863 the Declaration of Helsinki and the UNAIDS Guidance Document Ethical 1864 Considerations in HIV Preventive Vaccine Research, and the policies of many 1865 national governments and professional societies, recognize the need to redress these 1866 injustices by encouraging the participation of previously excluded groups in basic and 1867 applied biomedical research. [NOTE: Have epidemiologists’ groups done likewise?] 1868 Members of vulnerable groups also have the same entitlement to access to the 1869 benefits of investigational interventions that show promise of therapeutic benefit as 1870 persons not considered vulnerable, particularly when no superior or equivalent 1871 approaches to therapy are available.

1872 There has been a perception, sometimes correct and sometimes incorrect, that 1873 certain groups of persons have been overused as research subjects. In some cases such 1874 overuse has been based on the administrative availability of the populations. Research 1875 hospitals are often located in places where members of the lowest socioeconomic 1876 classes reside, and this has resulted in an apparent overuse of such persons. Other 1877 groups that may have been overused because they were conveniently available to 1878 researchers include students in investigators‟ classes, residents of long-term care 1879 facilities and subordinate members of hierarchical institutions.

Impoverished groups 1880 have been overused because of their willingness to serve as subjects in exchange for 1881 relatively small stipends. Prisoners have been considered ideal subjects for Phase I 1882 drug studies because of their highly regimented lives and, in many cases, their 1883 conditions of economic deprivation. 1884 Overuse of certain groups, such as the poor or the administratively available, is 1885 unjust for several reasons. It is unjust to selectively recruit impoverished people to 1886 serve as research subjects simply because they can be more easily induced to 1887 participate in exchange for small payments.

A further injustice occurs when such 1888 people are called upon to bear the burdens of research while others who are better off 1889 51 enjoy the benefits. However, although the burdens of research should not fall 1890 disproportionately on socio-economically disadvantaged groups, neither should such 1891 groups be categorically excluded from research protocols. It would not be unjust to 1892 selectively recruit poor people to serve as subjects in research designed to address 1893 problems that are prevalent in their group – malnutrition or poor living conditions, for 1894 example. Similar considerations apply to institutionalized groups or those whose 1895 availability to the investigators is for other reasons administratively convenient.

1896 Not only may certain groups within a society be inappropriately overused as 1897 research subjects, but also entire communities or societies may be overused. This has 1898 been particularly likely to occur in countries or communities with insufficiently well-1899 developed systems for the protection of the rights and welfare of human research 1900 subjects. Such overuse is especially questionable when the populations or 1901 communities concerned bear the burdens of participation in research but are 1902 extremely unlikely ever to enjoy the benefits of new knowledge and products 1903 developed as a result of the research.

(See Guideline 10: Research in populations and 1904 communities with limited resources.) 1905 1906 Guideline 13 1907 Research involving vulnerable persons 1908 1909 Special justification is required for inviting vulnerable individuals to serve as 1910 research subjects and, if they are selected, the means of protecting their rights 1911 and welfare must be strictly applied. 1912 1913 Commentary on Guideline 13 1914 1915 Vulnerable persons are those who are relatively (or absolutely) incapable of 1916 protecting their own interests. More formally, they may have insufficient power, 1917 intelligence, education, resources, strength, or other needed attributes to protect their 1918 own interests.

1919 1920 General considerations. The central problem presented by plans to involve vulnerable 1921 persons as research subjects is that such plans may entail an inequitable distribution of 1922 52 the burdens and benefits of research participation. Ethical justification of their 1923 involvement usually requires that investigators satisfy ethical review committees that: 1924 1925 – the research could not be carried out equally well with less vulnerable 1926 subjects; 1927 – the research is intended to obtain knowledge that will lead to improved 1928 diagnosis, prevention or treatment of diseases or other health problems 1929 characteristic of, or unique to, the vulnerable class– either the actual subjects 1930 or other similarly situated members of the vulnerable class;

1931 – research subjects and other members of the vulnerable class from which 1932 subjects are recruited will ordinarily be assured reasonable access to any 1933 diagnostic, preventive or therapeutic products that will become available as a 1934 consequence of the research; 1935 – the risks attached to interventions or procedures that do not hold out the 1936 prospect of direct health-related benefit will not exceed those associated with 1937 routine medical or psychological examination of such persons unless an 1938 ethical review committee authorizes a slight increase over this level of risk 1939 (Guideline 9); and, 1940 – when the prospective subjects are either incompetent or otherwise 1941 substantially unable to give informed consent, their agreement will be 1942 supplemented by the permission of their legal guardians or other appropriate 1943 representatives (Guidelines 14 and 15).

1944 1945 Vulnerable groups. Major classes of individuals conventionally considered vulnerable 1946 are those with limited capacity or freedom either to consent or to decline to consent. 1947 They include children, and persons who because of mental or behavioural disorders 1948 are incapable of giving informed consent. Less obvious as a vulnerable group are 1949 prospective subjects who are junior or subsidiary members of a hierarchical group; the 1950 quality of their consent requires careful consideration, since their agreement to 1951 volunteer may be unduly influenced, whether justified or not, by the expectation of 1952 preferential treatment if they agree or by fear of disapproval or retaliation if they 1953 refuse.

Examples of such groups are medical and nursing students, subordinate 1954 hospital and laboratory personnel, employees of pharmaceutical companies, and 1955 53 members of the armed forces or police. Because they work in close proximity to 1956 investigators, they tend to be called upon more often than others to serve as research 1957 subjects, and this could result in inequitable distribution of the burdens and benefits of 1958 research. 1959 Elderly persons are commonly regarded as vulnerable. With advancing age, 1960 people are increasingly likely to acquire attributes that define them as vulnerable.

1961 They may, for example, be institutionalized or develop varying degrees of dementia. 1962 If and when they acquire such vulnerability-defining attributes, and not before, it is 1963 appropriate to consider them vulnerable and to treat them accordingly. 1964 Other groups or classes may also be considered vulnerable. They include 1965 residents of nursing homes, people receiving welfare benefits or social assistance and 1966 other poor people and the unemployed, patients in emergency rooms, some ethnic and 1967 racial minority groups, homeless persons, nomads, refugees or displaced persons, 1968 prisoners, patients with incurable disease, individuals who are politically powerless, 1969 and members of communities unfamiliar with modern medical concepts.

To the extent 1970 that these and other classes of people have attributes resembling those of classes 1971 identified as vulnerable, the need for special protection of their rights and welfare 1972 should be reviewed and applied, where relevant. 1973 Although, on the whole, investigators must study less vulnerable groups before 1974 involving more vulnerable groups, some exceptions are justified. In general, children 1975 are not suitable for Phase I drug trials or for Phase I or II vaccine trials, but such trials 1976 may be permissible after studies in adults have shown some therapeutic or preventive 1977 effect.

For example, a Phase II vaccine trial seeking evidence of immunogenicity in 1978 infants may be justified when a vaccine has shown evidence of preventing or slowing 1979 progression of an infectious disease in adults, or Phase I research with children may 1980 be appropriate because the disease to be treated does not occur in adults or is 1981 manifested differently in children. 1982 1983 54 Guideline 14 1984 Research involving children 1985 1986 Before undertaking research involving children, the investigator must ensure 1987 that: 1988 – the research might not equally well be carried out with adults;

1989 – the purpose of the research is to obtain knowledge relevant to the health 1990 needs of children; 1991 – a parent or legal representative of each child has given permission; 1992 – the agreement (assent) of each child has been obtained to the extent of 1993 the child's capabilities; and 1994 – a child's refusal to participate or continue in the research will be 1995 respected. 1996 1997 Commentary on Guideline 14 1998 1999 Justifications for involving children in interventional studies. The participation of 2000 children is indispensable for research into diseases of childhood and conditions to 2001 which children are particularly susceptible (cf. vaccine trials), as well as for clinical 2002 trials of drugs that are designed for children as well as adults.

In the past, many new 2003 products were not tested for children though they were directed towards diseases also 2004 occurring in childhood; thus children either did not benefit from these new drugs or 2005 were exposed to them though little was known about their specific effects or safety in 2006 children. Now it is widely agreed that, as a general rule, the sponsor of any new 2007 therapeutic, diagnostic or preventive product that is likely to be indicated for use in 2008 children is obliged to evaluate its safety and efficacy for children before it is released 2009 for general distribution.

2010 2011 Justifications for involving children in other epidemiological studies. Observational 2012 epidemiological research, such as studies on how genetic and environmental factors 2013 present in childhood affect adult health, may be carried out even when the purpose is 2014 not "to obtain knowledge relevant to the health needs of children" provided that the 2015 other requirements are met. Since the potential benefits (in terms of etiological 2016 55 knowledge derived from the study) are relevant to adults while the potential harm 2017 would affect the children, such studies are usually permissible only in the context of 2018 the extremely reduced risks found in observational research.

A further justification 2019 would arise when the children in the study would also be potential beneficiaries of the 2020 study results when they become adults. 2021 Research on occupational hazards for children at work, which would produce 2022 knowledge relevant to children‟s health but may not meet the other requirements (e.g., 2023 if it might instead be carried out with adults), should nonetheless be regarded as 2024 permissible and even necessary, if nothing else to document the persistence and extent 2025 of child labor practices. 2026 2027 Assent of the child in studies for which competent subjects’ consent is required.

The 2028 willing cooperation of the child should be sought, after the child has been informed to 2029 the extent that the child's maturity and intelligence permit. The age at which a child 2030 becomes legally competent to give consent differs substantially from one jurisdiction 2031 to another; in some countries the "age of consent" established in their different 2032 provinces, states or other political subdivisions varies considerably. Often children 2033 who have not yet reached the legally established age of consent can understand the 2034 implications of informed consent and go through the necessary procedures; they can 2035 therefore knowingly agree to serve as research subjects.

Such knowing agreement, 2036 sometimes referred to as assent, is insufficient to permit participation in research 2037 unless it is supplemented by the permission of a parent, a legal guardian or other duly 2038 authorized representative. 2039 Some children who are too immature to be able to give knowing agreement, or 2040 assent, may be able to register a 'deliberate objection', an expression of disapproval or 2041 refusal of a proposed procedure. The deliberate objection of an older child, for 2042 example, is to be distinguished from the behaviour of an infant, who is likely to cry or 2043 withdraw in response to almost any stimulus.

Older children, who are more capable of 2044 giving assent, should be selected before younger children or infants, unless there are 2045 valid scientific reasons for involving younger children first. 2046 A deliberate objection by a child to taking part in research should always be 2047 respected even if the parents have given permission, unless the child needs an 2048 intervention that is not available outside the context of research, the investigational 2049 56 intervention shows promise of therapeutic benefit, and there is no acceptable 2050 alternative therapy. In such a case, particularly if the child is very young or immature, 2051 a parent or guardian may override the child's objections.

If the child is older and more 2052 nearly capable of independent informed consent, the investigator should seek the 2053 specific approval or clearance of the scientific and ethical review committees for 2054 initiating or continuing with the investigational treatment. If child subjects become 2055 capable of independent informed consent during the research, their informed consent 2056 to continued participation should be sought and their decision respected. 2057 A child with a likely fatal illness may object or refuse assent to continuation of 2058 a burdensome or distressing intervention. In such circumstances parents may press an 2059 investigator to persist with an investigational intervention against the child's wishes.

2060 The investigator may agree to do so if the intervention shows promise of preserving or 2061 prolonging life and there is no acceptable alternative treatment. In such cases, the 2062 investigator should seek the specific approval or clearance of the ethical review 2063 committee before agreeing to override the wishes of the child. 2064 2065 Permission of a parent or guardian. The investigator must obtain the permission of a 2066 parent or guardian in accordance with local laws or established procedures in all 2067 studies for which individual consent would be required from subjects capable of 2068 giving consent (see Guideline 4).

It may be assumed that children over the age of 12 2069 or 13 years are usually capable of understanding what is necessary to give adequately 2070 informed consent, but their consent (assent) should normally be complemented by the 2071 permission of a parent or guardian, even when local law does not require such 2072 permission. Even when the law requires parental permission, however, the assent of 2073 the child must be obtained. 2074 In some jurisdictions, some individuals who are below the general age of 2075 consent are regarded as “emancipated” or “mature” minors and are authorized to 2076 consent without the agreement or even the awareness of their parents or guardians.

2077 They may be married or pregnant or be already parents or living independently. Some 2078 studies involve investigation of adolescents‟ beliefs and behaviour regarding sexuality 2079 or use of recreational drugs; other research addresses domestic violence or child 2080 abuse. For studies on these topics, ethical review committees may waive parental 2081 permission if, for example, parental knowledge of the subject matter may place the 2082 57 adolescents at some risk of questioning or even intimidation by their parents. 2083 Because of the issues inherent in obtaining assent from children in institutions, 2084 such children should only exceptionally be subjects of research.

In the case of 2085 institutionalized children without parents, or whose parents are not legally authorized 2086 to grant permission, the ethical review committee may require sponsors or 2087 investigators to provide it with the opinion of an independent, concerned, expert 2088 advocate for institutionalized children as to the propriety of undertaking the research 2089 with such children. 2090 2091 Observation of research by a parent or guardian. A parent or guardian who gives 2092 permission for a child to participate in research should be given the opportunity, to a 2093 reasonable extent, to observe the research as it proceeds, so as to be able to withdraw 2094 the child if the parent or guardian decides it is in the child's best interests to do so.

2095 2096 Psychological and medical support. Research involving children should be conducted 2097 in settings in which the child and the parent can obtain adequate medical and 2098 psychological support. As an additional protection for children, an investigator may, 2099 when possible, obtain the advice of a child's family physician, paediatrician or other 2100 health-care provider on matters concerning the child's participation in the research. 2101 2102 (See also Guidelines 8, 9 and 13.) 2103 2104 Guideline 15 2105 Research involving individuals who by reason of mental or 2106 behavioural disorders are not capable of giving 2107 adequately informed consent 2108 2109 Before undertaking research involving individuals who by reason of mental or 2110 behavioural disorders are not capable of giving adequately informed consent, 2111 the investigator must ensure that: 2112 2113 – such persons will not be subjects of research that might equally well 2114 be carried out on persons whose capacity to give adequately 2115 58 informed consent is not impaired;

2116 – the purpose of the research is to obtain knowledge relevant to the 2117 particular health needs of persons with mental or behavioural 2118 disorders; 2119 – the consent of each subject has been obtained to the extent of that 2120 person's capabilities, and a prospective subject's refusal to 2121 participate in research is always respected, unless, in exceptional 2122 circumstances, there is no reasonable medical alternative and local 2123 law permits overriding the objection; and, 2124 – in cases where prospective subjects lack capacity to consent, 2125 permission is obtained from a responsible family member or a 2126 legally authorized representative in accordance with applicable law.

2127 2128 2129 Commentary on Guideline 15 2130 2131 General considerations. Most individuals with mental or behavioural disorders are 2132 capable of giving informed consent; this Guideline is concerned only with those who 2133 are not capable or who because their condition deteriorates become temporarily 2134 incapable. The investigator must obtain the approval of an ethical review committee 2135 to include such persons in research. They should never be subjects of research that 2136 might equally well be carried out on persons in full possession of their mental 2137 faculties, but they are clearly the only subjects suitable for a large part of research into 2138 the origins and treatment of certain severe mental or behavioural disorders.

2139 2140 Consent of the individual in studies for which competent subjects’ consent is required. 2141 The willing cooperation of persons whose mental and behavioural condition interferes 2142 with their ability to consent should be sought to the extent that their mental state 2143 permits, and any objection on their part to taking part in any study that has no 2144 components designed to benefit them directly should always be respected. The 2145 objection of such an individual to an investigational intervention intended to be of 2146 therapeutic benefit should be respected unless there is no reasonable medical 2147 alternative and local law permits overriding the objection.

2148 2149 59 Permission of a surrogate for a subject incapable of giving informed consent. The 2150 investigator must obtain the permission of a surrogate in accordance with local laws 2151 or established procedures in all studies for which individual consent would be 2152 required from subjects capable of giving consent (see Guideline 4). The permission of 2153 an immediate family member or other person with a close personal relationship with 2154 the individual should be sought, but it should be recognized that these proxies may 2155 have their own interests that may call their permission into question.

Some relatives 2156 may not be primarily concerned with protecting the rights and welfare of the patients. 2157 Moreover, a close family member or friend may wish to take advantage of a research 2158 study in the hope that it will succeed in “curing” the condition. Some jurisdictions do 2159 not permit third-party permission for subjects lacking capacity to consent. Legal 2160 authorization may be necessary to involve in research an individual who has been 2161 committed to an institution by a court order. 2162 2163 Serious illness in persons who because of mental or behavioural disorders are unable 2164 to give adequately informed consent.

Persons who because of mental or behavioural 2165 disorders are unable to give adequately informed consent and who have, or are at risk 2166 of, serious illnesses such as HIV infection, cancer or hepatitis should not be deprived 2167 of the possible benefits of investigational drugs, vaccines or devices that show 2168 promise of therapeutic or preventive benefit, particularly when no superior or 2169 equivalent therapy or prevention is available. Their entitlement to access to such 2170 therapy or prevention is justified ethically on the same grounds as is such entitlement 2171 for other vulnerable groups.

2172 Persons who are unable to give adequately informed consent by reason of 2173 mental or behavioural disorders are, in general, not suitable for participation in formal 2174 clinical trials except those trials that are designed to be responsive to their particular 2175 health needs and can be carried out only with them. 2176 2177 (See also Guidelines 8, 9 and 13.) 2178 2179 2180 2181 2182 60 Guideline 16 2183 Women as research participants 2184 2185 Investigators, sponsors or ethical review committees should not exclude 2186 women of reproductive age from epidemiological research. The potential for 2187 becoming pregnant during a study should not, in itself, be used as a reason for 2188 precluding or limiting participation.

However, a thorough discussion of risks to 2189 the pregnant woman and to her fetus is a prerequisite for the woman’s ability 2190 to make a rational decision to enrol in an interventional study. In this 2191 discussion, if participation in the research might be hazardous to a fetus or a 2192 woman if she becomes pregnant, the sponsors/investigators should guarantee 2193 the prospective subject a pregnancy test and access to effective contraceptive 2194 methods before the research commences. Where such access is not possible, 2195 for legal or religious reasons, investigators should not recruit for such 2196 possibly hazardous research women who might become pregnant.

2197 2198 Commentary on Guideline 16 2199 2200 Women in most societies have been discriminated against with regard to their 2201 involvement in research. Women who are biologically capable of becoming pregnant 2202 have been customarily excluded from formal clinical trials of drugs, vaccines and 2203 medical devices owing to concern about undetermined risks to the fetus. 2204 Consequently, relatively little is known about the safety and efficacy of most drugs, 2205 vaccines or devices for such women, and this lack of knowledge can be dangerous. 2206 A general policy of excluding from research women biologically capable of 2207 becoming pregnant is unjust in that it deprives women as a group of the benefits of the 2208 new knowledge derived from such studies.

Further, it is an affront to their right of 2209 self-determination. It is particularly important that occupations that predominantly 2210 involve women workers are not excluded from epidemiological research on potential 2211 occupational hazards. Nevertheless, when given the opportunity to participate in 2212 research that could pose risks to the fetus, women of childbearing age should be 2213 helped to understand that such risk would arise if they become pregnant during the 2214 research. 2215 Although this general presumption favours the inclusion of women in research, 2216 61 it must be acknowledged that in some parts of the world women are vulnerable to 2217 neglect or harm in research because of their social conditioning to submit to authority, 2218 to ask no questions, and to tolerate pain and suffering.

When women in such 2219 situations are potential subjects in research, investigators need to exercise special care 2220 in the informed consent process to ensure that they have adequate time and a proper 2221 environment in which to take decisions on the basis of clearly given information. 2222 2223 Individual consent of women: In research involving women of reproductive age, 2224 whether pregnant or non-pregnant, only the informed consent of the woman herself is 2225 required for her participation. In no case should the permission of a spouse or partner 2226 replace the requirement of individual informed consent.

If women wish to consult 2227 with their husbands or partners or seek voluntarily to obtain their permission before 2228 deciding to enrol in research, that is not only ethically permissible but in some 2229 contexts highly desirable. A strict requirement of authorization of spouse or partner, 2230 however, violates the substantive principle of respect for persons. 2231 A thorough discussion of risks to the pregnant woman and to her fetus is a 2232 prerequisite for the woman‟s ability to make a rational decision to enrol in a study. 2233 For women who are not pregnant at the outset of a study but who might become 2234 pregnant while they are still subjects, the consent discussion should include 2235 information about the alternative of voluntarily withdrawing from the study and, 2236 where legally permissible, terminating the pregnancy.

Also, if the pregnancy is not 2237 terminated, they should be guaranteed a medical follow-up. 2238 2239 (See also Guideline 17.) 2240 2241 Guideline 17 2242 Pregnant women as research participants 2243 2244 Pregnant women should be presumed to be eligible for participation in 2245 epidemiological research. Investigators and ethical review committees should 2246 ensure that prospective subjects who are pregnant are adequately informed 2247 about the risks and benefits to themselves, their pregnancies, the fetus and 2248 their subsequent offspring, and to their fertility. 2249 62 2250 Interventional studies should be performed in this population only if it is 2251 relevant to the particular health needs of a pregnant woman or her fetus, or to 2252 the health needs of pregnant women in general, and, when appropriate, if it is 2253 supported by reliable evidence from animal experiments, particularly as to 2254 risks of teratogenicity and mutagenicity.

2255 2256 Commentary on Guideline 17 2257 2258 The justification of research involving pregnant women is complicated by the fact that 2259 it may present risks and potential benefits to two beings–the woman and the fetus–as 2260 well as to the person the fetus is destined to become. Even when evidence concerning 2261 risks is unknown or ambiguous, the decision about acceptability of risk to the fetus 2262 should be made by the woman as part of the informed consent process. Though this 2263 decision should be made by the mother, it is desirable in research directed at the 2264 health of the fetus to obtain the father‟s opinion as well, when possible.

2265 Especially in communities or societies in which cultural beliefs accord more 2266 importance to the fetus than to the woman‟s life or health, women may feel 2267 constrained to participate, or not to participate, in research. Special safeguards should 2268 be established to prevent undue inducement to pregnant women to participate in 2269 research in which interventions hold out the prospect of direct benefit to the fetus. 2270 Where fetal abnormality is not recognized as an indication for abortion, pregnant 2271 women should not be recruited for research in which there is a realistic basis for 2272 concern that fetal abnormality may occur as a consequence of participation as a 2273 subject in research.

2274 Investigators should include in protocols on research on pregnant women a plan 2275 for monitoring the outcome of the pregnancy with regard to both the health of the 2276 woman and the short-term and long-term health of the child. 2277 2278 (See also Commentary on Guidelines 14 and 16.) 2279 2280 63 Guideline 18 2281 Safeguarding confidentiality 2282 2283 A healthcare provider should not submit any identifiable data about a patient to 2284 an investigator or to a database unless the patient permits such submission of 2285 data or it is authorized or mandated by law. The custodian of a database, and 2286 an investigator who receives data for research, must establish secure 2287 safeguards for the confidentiality of the data.

Subjects should be told the 2288 limits, legal or other, to the investigators' ability to safeguard confidentiality 2289 and the possible consequences of breaches of confidentiality”. 2290 2291 Commentary on Guideline 18 2292 2293 In addition to the requirements set forth in this Guideline, a growing body of laws 2294 have been adopted in many countries establishing detailed legal requirements 2295 regarding the protection of the confidentiality and security of health-related data. 2296 2297 Confidentiality between investigator and subject. Research relating to individuals and 2298 groups may involve the collection and storage of information that, if disclosed to third 2299 parties, could cause harm or distress.

Investigators should arrange to protect the 2300 confidentiality of such information by, for example, omitting information that might 2301 lead to the identification of individual subjects, limiting access to the information, 2302 anonymizing data, or other means. During the process of obtaining informed consent 2303 the investigator should inform the prospective subjects about the precautions that will 2304 be taken to protect confidentiality. 2305 The obligation to preserve confidentiality of research data encompasses all 2306 identifying information because the disclosure of such information can cause 2307 physical, psychological, social or economic harm to individuals, couples, families or 2308 other social groups or infringe their intimacy.

One way of achieving confidentiality is 2309 to use only unidentifiable data; for instance, when testing unlinked anonymous blood 2310 samples for HIV infection or when unlinked anonymized or already partially 2311 aggregated data are provided by existing registries (e.g., of deaths) to the 2312 epidemiologist for descriptive studies. 2313 64 When linked data and samples are used, epidemiologists customarily discard 2314 personal identifying information when consolidating data for purposes of statistical 2315 analysis; this also occurs when investigators have linked different sets of data 2316 regarding individuals with the consent of individual subjects.

When personal 2317 identifiers remain on records used for a study, investigators should explain to ethical 2318 review committees why this is necessary and how confidentiality will be protected. 2319 Participation in HIV/AIDS drug and vaccine trials may impose upon the 2320 research subjects significant associated risks of social discrimination or harm; such 2321 risks merit consideration equal to that given to adverse medical consequences of the 2322 drugs and vaccines. Efforts must be made to reduce their likelihood and severity. For 2323 example, subjects in vaccine trials must be enabled to demonstrate that their HIV 2324 seropositivity is due to their having been vaccinated rather than to natural infection.

2325 This may be accomplished by providing them with documents attesting to their 2326 participation in vaccine trials, or by maintaining a confidential register of trial 2327 subjects, from which information can be made available to outside agencies at a 2328 subject's request. 2329 2330 Limits of confidentiality. Prospective subjects should be informed of limits to the 2331 ability of investigators to ensure strict confidentiality and of the foreseeable adverse 2332 consequences of breaches of confidentiality. Some jurisdictions require the reporting 2333 to appropriate agencies of, for instance, certain communicable diseases or evidence of 2334 child abuse or neglect.

Health authorities may have the legal right to inspect study 2335 records, and a sponsor's compliance audit staff may require and obtain access to 2336 confidential data. Although employers should be informed of occupational health 2337 study findings only at the group level, the risk exists, particularly in small 2338 organizations, that the employer will be able to identify the subjects. Pooling data 2339 from a number of comparable organizations may reduce–but not completely 2340 foreclose–this risk. Conversely, research that links data from different sources (e.g., 2341 health records, employment records, etc.) may increase the risk that individuals can be 2342 identified.

These and similar limits to the ability to maintain confidentiality should be 2343 anticipated and disclosed to prospective subjects (see Guideline 5, #15). 2344 2345 65 Data security. Study materials and databases may contain data which besides being 2346 confidential also need to be ensured long life spans, which may in extreme cases 2347 cover several generations. Standards and methods need to be developed for the secure 2348 preservation of data that are, or could be, held for longitudinal studies. Investigators 2349 are responsible for ensuring data security and legitimate access to data by protecting 2350 them against physical injury, criminal action and during any change which may be 2351 associated with changes of technical systems.

Several general principles are useful in 2352 judging the adequacy of data protection: 2353 2354 • Plans for data protection and custody of data, copies and backup facilities, 2355 whether in the hands of an institution or an individual investigator, should be 2356 outlined in the research plan and reviewed by the ethical review committee. 2357 • Limitations on access and legal requirements for disclosure, if any, should be 2358 clearly outlined in the research plan. 2359 • The level of identifiability should be appropriate to the scientific goals of the 2360 research, as well as appropriate to adequately protecting research subjects.

2361 • The informed consent process should include a description of how data and/or 2362 samples will be handled and who will have access; when there will be different 2363 levels of data protection, the information should be explicit about this, explaining 2364 in general terms the modes of protection at each level. 2365 2366 Confidentiality between physician and patient. Physicians and other health care 2367 professionals record the details of their observations and interventions in medical 2368 records. Patients have the right to expect that these health-care professionals will hold 2369 all information about them in strict confidence and disclose it only to those who need, 2370 or have a legal right to, the information, such as other attending physicians, nurses, or 2371 other health-care workers who perform tasks related to the diagnosis and treatment of 2372 patients.

2373 The use of such records in epidemiological studies without the informed consent of 2374 the patients concerned may be approved by an ethical review committee when this is 2375 66 consistent with the requirements of applicable law and with the conditions discussed 2376 in the Commentary on Guideline 4, and provided that there are secure safeguards of 2377 confidentiality; information may also be provided without patient consent to a register 2378 or database when authorized or mandated by law. Access by researchers to patients' 2379 medical records must be approved in advance by an ethical review committee and 2380 supervised by a person who is fully aware of the confidentiality requirements.

When 2381 the practice of collecting patient records for use in research without informed consent 2382 has been approved in a particular setting (such as a hospital or a clinic), patients 2383 should be notified of this practice; notification is usually by means of a statement in 2384 patient-information brochures. It should be made clear to persons that they have an 2385 option to restrict the secondary, research uses of information that is submitted for 2386 billing, prescribing, or other purposes and that if they choose to do so, the care 2387 provided will not be affected. 2388 For populations covered by automated health databases, an ethical review 2389 committee with expertise regarding the scope of access researchers will have to the 2390 medical information and the types of research they want to conduct should ensure that 2391 confidentiality rules and procedures are in place and certify the public health value of 2392 the research (e.g., the Data Access Review Committee for the Canadian Saskatchewan 2393 Health database, the Scientific and Ethical Advisory Group for the UK General 2394 Practice Research Database).

The ethical review committee should also determine 2395 how patients should be advised of such practices, usually by means of a statement in 2396 patient-information brochures, and the ethical or legal need to provide patients with 2397 the choice to “opt out” of secondary use of certain parts of their medical record. 2398 When already existing collections of medical records that were assembled and 2399 stored without an explicit notification and consent procedure (including an 2400 67 opportunity to “opt out”) offer important and otherwise unobtainable data, an ethical 2401 review committee needs to decide whether the use of such records is justified.

2402 Arguments pertaining to this decision are discussed within the more general 2403 framework of the waiving of consent in the Commentary on Guideline 4, under the 2404 section "Waiver of consent requirements." 2405 2406 Disclosure of test results to individuals. When genetic or other diagnostic tests will be 2407 reported to the subject or to the subject‟s physician, the subject should be informed that such 2408 disclosure will occur and that the samples to be tested will be clearly labeled. Investigators 2409 should not disclose results of such diagnostic tests to relatives of subjects without the 2410 subjects` consent.

In places where immediate family relatives would usually expect to be 2411 informed of such results, the research protocol, as approved or cleared by the ethical review 2412 committee, should indicate the precautions in place to prevent such disclosure of results 2413 without the subjects‟ consent; such plans should be clearly explained during the process of 2414 obtaining informed consent. 2415 2416 Issues of confidentiality in genetic research. An investigator who proposes to perform 2417 genetic tests of known clinical or predictive value on biological samples that can be 2418 linked to an identifiable individual must obtain the informed consent of the individual 2419 or, when indicated, the permission of a legally authorized representative.

(Issues 2420 raised by research on stored samples are addressed in Guideline 24 and the associated 2421 Commentary.) 2422 2423 Special confidentiality issues for groups in genetic research. When unidentifiable 2424 biological samples (that is, those that have been fully anonymized and unlinked) are 2425 used in genetic research in a specific population or community, the results obtained 2426 cannot be fed back to individual participants, but in such cases research findings and 2427 advice to the relevant group may be communicated by suitable means. These 2428 68 processes should be fully explained to the prospective subjects as part of informed 2429 consent (see Guideline 5).

2430 Epidemiologists and ethical review committees should however be aware that, 2431 under specific circumstances, the genetic information gathered in a study (on 2432 pharmacogenetics or pharmacogenomics, for example) may have a significant impact 2433 on the subject and his/her family extending over generations, and in some instances 2434 on the whole population group to which the subject concerned belongs. 2435 With genetic population studies, the possibility of new forms of discrimination 2436 based on genotype may emerge. If genetic variations in certain diseases or conditions 2437 are significantly more common in a particular community or ethnic group, this 2438 information may result in stigmatization and stereotyping and in discrimination in 2439 health care services or in the fields of life insurance, employment, reproductive rights, 2440 etc.

The importance of confidentiality is heightened when genetic information might 2441 be used to discriminate or infringe the human rights, fundamental freedoms or dignity 2442 of individuals, families, groups or communities (see Guideline 8). 2443 2444 Guideline 19 2445 Right of injured subjects to treatment and compensation 2446 2447 Investigators should ensure that research subjects who suffer injury as a 2448 result of their participation are entitled to free medical treatment for such injury 2449 and to such financial or other assistance as would compensate them equitably 2450 for any resultant impairment, disability or handicap.

In the case of death as a 2451 result of their participation, their dependants are entitled to compensation. 2452 Subjects must not be asked to waive the right to compensation. 2453 2454 Commentary on Guideline 19 2455 2456 Guideline 19 is concerned with two distinct but closely related entitlements. The 2457 first is the uncontroversial entitlement to free medical treatment and compensation for 2458 accidental injury inflicted by procedures or interventions performed exclusively to 2459 accomplish the purposes of research (non-therapeutic procedures). The second is the 2460 entitlement of dependants to material compensation for death or disability occurring 2461 69 as a direct result of study participation.

Implementing a compensation system for 2462 research-related injuries or death is likely to be complex, however. 2463 2464 Equitable compensation and free medical treatment. Compensation is owed to 2465 research subjects who are disabled as a consequence of injury from procedures 2466 performed solely to accomplish the purposes of research. Compensation and free 2467 medical treatment are generally not owed to research subjects who suffer expected or 2468 foreseen adverse reactions to investigational therapeutic, diagnostic or preventive 2469 interventions when such reactions are not different in kind from those known to be 2470 associated with established interventions in standard medical practice.

2471 The ethical review committee should determine in advance: i) the injuries for 2472 which subjects will receive free treatment and, in case of impairment, disability or 2473 handicap resulting from such injuries, be compensated; and ii) the injuries for which 2474 they will not be compensated. Prospective subjects should be informed of the 2475 committee's decisions, as part of the process of informed consent. As an ethical 2476 review committee cannot make such advance determination in respect of unexpected 2477 or unforeseen adverse reactions, such reactions must be presumed compensable and 2478 should be reported to the committee for prompt review as they occur.

2479 Subjects must not be asked to waive their rights to compensation or required to 2480 show negligence or lack of a reasonable degree of skill on the part of the investigator 2481 in order to claim free medical treatment or compensation. The informed consent 2482 process or form should contain no words that would absolve an investigator from 2483 responsibility in the case of accidental injury, or that would imply that subjects would 2484 waive their right to seek compensation for impairment, disability or handicap. 2485 Prospective subjects should be informed that they will not need to take legal action to 2486 secure the free medical treatment or compensation for injury to which they may be 2487 entitled.

They should also be told what medical service or organization or individual 2488 will provide the medical treatment and what organization will be responsible for 2489 providing compensation. 2490 2491 Obligation of the sponsor with regard to compensation. Before the research begins, 2492 the sponsor, whether a pharmaceutical company or other organization or institution, 2493 or a government (where government insurance is not precluded by law), should agree 2494 70 to provide compensation for any physical injury for which subjects are entitled to 2495 compensation, or come to an agreement with the investigator concerning the 2496 circumstances in which the investigator must rely on his or her own insurance 2497 coverage (for example, for negligence or failure of the investigator to follow the 2498 protocol, or where government insurance coverage is limited to negligence).

In certain 2499 circumstances it may be advisable to follow both courses. Sponsors should seek 2500 adequate insurance against risks to cover compensation, independent of proof of fault. 2501 2502 Guideline 20 2503 Strengthening capacity for ethical and scientific review 2504 and epidemiological research 2505 2506 Many countries lack the capacity to assess or ensure the scientific quality or 2507 ethical acceptability of epidemiological research proposed or carried out in 2508 their jurisdictions. In externally sponsored collaborative studies, sponsors and 2509 investigators have an ethical obligation to ensure that the research projects for 2510 which they are responsible in such countries contribute effectively to national 2511 or local capacity to design and conduct epidemiological research, and to 2512 provide scientific and ethical review and monitoring of such research.

2513 2514 Capacity-building may include, but is not limited to, the following activities: 2515 2516  establishing and strengthening independent and competent ethical review 2517 processes/ committees 2518 2519  strengthening research capacity 2520 2521  developing technologies appropriate to public health, health care and 2522 epidemiological research 2523 2524  training of research and health-care staff 2525 2526  educating the community from which research subjects will be drawn. 2527 2528 71 Commentary on Guideline 20 2529 2530 External sponsors and investigators have an ethical obligation to contribute to a host 2531 country's sustainable capacity for independent scientific and ethical review and 2532 epidemiological research.

Strengthening capacity for conducting epidemiological 2533 research, as well as for undertaking scientific and ethical review of epidemiological 2534 projects, should be regarded as a specific need in many countries, notably because 2535 adequate capacity for biomedical research does not automatically entail adequate 2536 capacity for epidemiological research. This further extends to capacity in very 2537 specialized domains of epidemiological research such as genetic, occupational or 2538 social epidemiology. 2539 Before undertaking research in a host country with little or no such capacity, 2540 external sponsors and investigators should include in the research protocol a plan that 2541 specifies the contribution they will make.

The amount of capacity building reasonably 2542 expected should be proportional to the magnitude of the research project. A brief 2543 epidemiological study involving only review of medical records, for example, would 2544 entail relatively little, if any, such development, whereas a considerable contribution 2545 is to be expected of an external sponsor of, for instance, a large-scale vaccine field-2546 trial expected to last two or three years. 2547 The specific capacity-building objectives should be determined and achieved 2548 through dialogue and negotiation between external sponsors and host-country 2549 authorities.

External sponsors would be expected to employ and, if necessary, train 2550 local individuals to function as investigators, research assistants or data managers, for 2551 example, and to provide, as necessary, reasonable amounts of financial, educational 2552 and other assistance for capacity-building. To avoid conflict of interest and safeguard 2553 the independence of review committees, financial assistance should not be provided 2554 directly to them; rather, funds should be made available to appropriate authorities in 2555 the host-country government or to the host research institution. 2556 2557 (See also Guidelines 10 and 22.) 2558 2559 Guideline 21 2560 Ethical obligation of external sponsors 2561 72 to provide health-care services 2562 2563 External sponsors are ethically obliged to ensure the availability of: 2564 2565  health-care services that are essential to the safe conduct of the research;

2566 2567  treatment for subjects who suffer injury as a consequence of research 2568 interventions; and, 2569 2570  services that are a necessary part of the commitment of a sponsor to make 2571 a beneficial intervention or product developed as a result of the research 2572 reasonably available to the population or community concerned. 2573 2574 Commentary on Guideline 21 2575 2576 Obligations of external sponsors to provide health-care services will vary with the 2577 circumstances of particular studies and the needs of host countries. Some types of 2578 interventional epidemiological studies are intended to find out whether a screening 2579 programme for a disease may lead to an improvement in prognosis, by means of early 2580 diagnosis and treatment.

The intervention cannot be limited to administering a 2581 screening test and examining whether the disease has been detected at an earlier stage 2582 than through standard clinical practice, but should also include provision of the 2583 pertinent treatment. 2584 The sponsors' obligations in particular studies should be clarified before the 2585 research is begun. The research protocol should specify what health-care services will 2586 be made available, during and after the research, to the subjects themselves, to the 2587 community from which the subjects are drawn, or to the host country, and for how 2588 long.

In addition, investigators should specify what action if any they will take when 2589 medical conditions are detected within a study population that are not related to the 2590 study but that need treatment, for instance, obesity or hypertension when recruiting 2591 subjects in an observational cohort study of diet and cancer. The details of these 2592 arrangements should be agreed by the sponsor, officials of the host country, other 2593 interested parties, and, when appropriate, the community from which subjects are to 2594 be drawn. The agreed arrangements should be specified in the consent process and 2595 73 document.

2596 Although sponsors are, in general, not obliged to provide health-care services 2597 beyond that which is necessary for the conduct of the research, it is morally 2598 praiseworthy to do so. Such services typically include treatment for diseases 2599 contracted in the course of the study. It might, for example, be agreed to treat cases of 2600 an infectious disease contracted during a trial of a vaccine designed to provide 2601 immunity to that disease, or to provide treatment of incidental conditions unrelated to 2602 the study. 2603 The scope and limits of the obligation to ensure that subjects who suffer injury 2604 as a consequence of research interventions obtain medical treatment free of charge, 2605 and that compensation be provided for death or disability occurring as a consequence 2606 of such injury, are the subject of Guideline 19.

2607 When prospective or actual subjects are found to have diseases unrelated to the 2608 research, or cannot be enrolled in a study because they do not meet the health criteria, 2609 investigators should, as appropriate, advise them to obtain, or refer them for, medical 2610 care. In general, also, in the course of a study, sponsors should disclose to the proper 2611 health authorities information of public health concern arising from the research. 2612 The obligation of the sponsor to make reasonably available for the benefit of the 2613 population or community concerned any intervention or product developed, or 2614 knowledge generated, as a result of the research is considered in Guideline 10.

2615 2616 Guideline 22 2617 Disclosure and review of potential conflicts of interest 2618 2619 The investigator is responsible for ensuring that the materials submitted to an 2620 ethical review committee include a declaration of any potential conflicts of 2621 interest affecting the study. Ethical review committees should develop forms 2622 that facilitate the reporting of such potential conflicts and materials explaining 2623 their use for investigators. Ethical review committees should evaluate each 2624 study in the light of any declared conflicts and ensure that appropriate means 2625 of mitigation are provided.

If a potentially serious conflict of interest cannot be 2626 adequately mitigated, the committee should not approve the project. 2627 2628 74 Commentary on Guideline 22 2629 2630 Types of conflicting interests. Conflicts can arise from a sponsor's interest in the 2631 study's outcome; such interests include those of a health ministry or other public 2632 agency and are not limited to commercial sponsors. Such conflicts may include a 2633 financial stake held by the investigator or senior members of the research team (as 2634 well as their close family members) in the sponsor of the research (such as an equity 2635 interest), payments to the investigator that depend on the rapidity with which subjects 2636 are recruited or certain results reported, restrictions on the investigator's freedom to 2637 analyze the data or publish research results, or dependence of a research centre on 2638 substantial, ongoing support from a particular sponsor, private or public.

2639 2640 Potential conflicts of interest related to project support. Epidemiological studies may 2641 receive funding from commercial firms. Such sponsors have good reason to support 2642 research methods that are ethically and scientifically acceptable, but cases have arisen 2643 in which the conditions of funding may have introduced bias. For example, 2644 investigators have sometimes had little or no input into study design, limited access to 2645 the raw data, or limited participation in data interpretation, and the results of some 2646 studies have not been published when they were unfavourable to the sponsor's 2647 product.

(This risk of bias may also arise with other sources of support, such as 2648 government or foundations.) As the persons directly responsible for their work, 2649 investigators should not enter into agreements that interfere unduly with their access 2650 to the data or their ability to analyze the data independently, prepare manuscripts, or 2651 publish them. Investigators must also disclose potential or apparent conflicts of 2652 interest on their part to the ethical review committee or to other institutional 2653 committees designed to evaluate and manage such conflicts. Ethical review 2654 committees should therefore ensure that these conditions are met (see also the 2655 Commentary on Guideline 2, Multi-centre research).

2656 2657 Institutional conflicts. Officials overseeing research also need to be aware of – and, as 2658 necessary, take steps to mitigate – institutional conflicts of interest which may arise 2659 when a research centre derives substantial support (perhaps covering years of 2660 funding) from a single sponsor or handful of sponsors; in such circumstances, it may 2661 75 be difficult for persons acting on behalf of the organization, including members of the 2662 ethical review committee, to reach judgments adverse to the sponsor's interests or 2663 wishes. The fact that the ethical review committee (or the institution where it 2664 operates) is paid a fee for reviewing a study does not present an inherent conflict of 2665 interest, provided that the fee is reasonably related to the costs of conducting the 2666 review, is not dependent on the outcome of the review, is uniform for all projects of 2667 comparable complexity, and is set and negotiated by persons other than those actually 2668 engaged in the ethical review process.

2669 2670 Standardized disclosure. Investigators will most likely come to recognize potential 2671 conflicts of interest if they are prompted to scrutinize research sponsorship as an 2672 expected part of preparing a description of their projects for the ethical review 2673 committee. Thus, the development of a standardized disclosure form and related 2674 educational and explanatory materials (by a committee or group of committees, such 2675 as a research ethics association) is recommended as a good way to ensure that 2676 investigators understand the potential for conflicts of interest and routinely report 2677 relevant facts about their own studies to review committees and in all publications.

It 2678 is important that such a document provide a definition of potential conflict of interest. 2679 The explanatory materials should also help investigators to understand that a potential 2680 conflict of interest is not necessarily disqualifying but may be managed either through 2681 disclosure (both before the study, in consent materials, and when any results are 2682 reported) or other means. 2683 2684 Mitigation of conflicts. The means that ethical review committees may wish to 2685 consider for mitigating conflicts of interest include an agreed process for peer review 2686 of the study design, analysis, results, and interpretation; guarantees of the 2687 investigator's right to determine the scientific design and to use the data and publish 2688 results, free of undue restrictions from the sponsor; the existence of multiple sources 2689 of support for the study; etc.

When appropriate, the committee may also require that 2690 potential conflicts of interest be part of the information provided in seeking subjects' 2691 consent to participate, beyond describing "the nature and sources of funding for the 2692 research", which is an element of informed consent under Guideline 5. 2693 2694 76 Guideline 23 2695 Use of the Internet in epidemiological research 2696 2697 If the Internet is used as a tool to identify respondents or to collect data in 2698 epidemiological research, the investigator must ensure that an appropriate 2699 informed consent procedure is applied and that data confidentiality is 2700 maintained.

2701 2702 Commentary on Guideline 23 2703 2704 There are several ways in which researchers can use the Internet while performing 2705 epidemiological research. First, while collecting data, researchers may use the Internet 2706 to actually perform the research itself (online research); visitors to sites may be 2707 enrolled as respondents and questionnaires may be made accessible through the 2708 Internet. In open Internet locations, investigators may observe, as a source of data, 2709 what others are saying and doing without necessarily interacting directly with other 2710 visitors to the site in question.

(Such virtual “spaces” are public but may be regarded 2711 as private by users who are not adequately attentive to the ability of observers to 2712 “participate” invisibly.) Second, the Internet plays an increasingly important role for 2713 researchers in building databases; researchers may send electronic files containing the 2714 results of their research to other researchers for collaborative purposes or to aid in the 2715 construction of a centralized repository on information on a particular topic. This is 2716 the case, for instance, in multi-centre trials. Finally, after completion of the study, 2717 researchers may want to make some results available through the Internet.

The 2718 principles of scientific validity of the study, informed consent, confidentiality, and 2719 balancing of potential benefit and harm are generally applicable to all of these uses of 2720 the Internet, but research using the Internet can have several unique features. 2721 2722 Using the Internet to collect data and build databases. Subjects‟ privacy, 2723 confidentiality and security are at stake when research is conducted through the 2724 Internet. Researchers should be explicit about their presence while doing online 2725 research and seek the informed consent of participants. As part of the informed 2726 consent process, participants should be informed of the means and degree of 2727 77 protection applied to the data as well as where the data and their backup will be 2728 stored, for how long and who will have access to them.

As no face-to-face contact 2729 takes place between participants and investigators, subjects' agreement to participate 2730 should be based on a clear disclosure of the purposes for which data are being 2731 collected and who (investigator and institution) is collecting or accessing them; the 2732 investigator is responsible for maintaining records that document informed consent. 2733 (see also Guideline 6) 2734 Subjects‟ privacy, confidentiality and security are at stake when data are 2735 conveyed to others electronically. Researchers should make sure that confidentiality 2736 of information is guaranteed during data collection, transfer to other centres and the 2737 building of a common database.

Registration forms and questionnaires with personal 2738 identifiers should receive a high degree of security. Passwords and the best available 2739 technology, such as encryption, should be used in order to make sure that only 2740 authorized persons are able to read the data. 2741 2742 Results made available on the Internet. After completion of a study, the accuracy and 2743 completeness of the information made available on the Internet become relevant. 2744 Researchers should be explicit in indicating whether the information provided is 2745 preliminary or definitive, and how complete it is.

2746 2747 Electronic collection of health-related data through new technologies. Subjects‟ 2748 privacy, confidentiality and security are also at stake when data are collected through 2749 electronic devices carried by or implanted in individuals. Epidemiological studies 2750 using such methods must attend to the resulting issues. 2751 2752 Guideline 24 2753 Use of stored biological samples and related data 2754 2755 When collecting and storing human biological samples (and related data, such 2756 as health or employment records) for future epidemiological research, the 2757 investigator must obtain the voluntary informed consent of the individual 2758 donor or, in the case of an individual who is not capable of giving informed 2759 consent, the permission of a legally authorized representative in accordance 2760 78 with applicable law.

The consent should specify: the conditions and duration 2761 of storage; who will have access to the samples; the foreseeable uses of the 2762 samples, whether limited to an already fully defined study or extending to a 2763 number of wholly or partially undefined studies; and the intended goal of such 2764 use, whether only for research, basic or applied, or also for commercial 2765 purposes. The ethical review committee should satisfy itself that the proposed 2766 collection and storage protocol and the consent procedure meet these 2767 specifications. 2768 2769 The protocol of every study using stored human biological samples (and 2770 related data) must be submitted to an ethical review committee, which should 2771 satisfy itself that the proposed use of the samples comes within the scope 2772 specifically agreed to by the subjects.

2773 2774 For stored samples collected for past research, clinical or other purposes 2775 without informed consent to their use for research, the ethical review 2776 committee may consider waiving the consent if it proves materially unfeasible 2777 to obtain it, provided that it concludes that doing so would not harm the rights 2778 or welfare of the persons from whom the samples were collected. 2779 2780 Commentary on Guideline 24 2781 2782 Epidemiologists have long analyzed biological samples and are now increasingly 2783 using the tools of molecular genetics to understand the interaction of factors that 2784 contribute to disease.

When combined with information from medical and other 2785 sources (such as dietary or occupational records), data from biological samples 2786 provide a powerful tool in deciphering the role of environmental and genetic factors 2787 in human health and disease. Consent to the use of samples collected and immediately 2788 analyzed for the purpose of a specific epidemiological study come under Guideline 4 2789 and has been discussed in the Commentary on that Guideline. 2790 Particular issues arise, however, for the use of stored samples, repositories of 2791 which are fast multiplying as a key resource for research, including in particular in the 2792 field of epidemiology.

These issues are different in degree, if not in nature, from those 2793 concerning the use solely of recorded data, such as medical records. While from the 2794 79 latter it is only possible to generate new information by linking different recorded 2795 data, for instance drug use with a subsequent health outcome, analytical 2796 determinations of all kinds carried out on biological samples can generate new data, 2797 and consequently new information, in a virtually limitless amount. 2798 This inherent information-generating potential requires that strict measures be 2799 taken, to the satisfaction of the ethical review committee examining the protocol for 2800 establishment and management of a repository, for assuring not only the physical 2801 protection and maintenance of the samples but also appropriate confidentiality of the 2802 link between biological specimens and personal identifiers of the donating subjects.

2803 This responsibility falls upon the custodian of the repository. It is the responsibility of 2804 the person who obtains and submits the sample to a repository (e.g., a physician in the 2805 course of a diagnostic or screening procedure, or an epidemiologist in the course of a 2806 field study) to ensure that donors whose samples and related data will be stored have 2807 been informed about the potential future uses of such material, and that the samples 2808 will be stored and made available in accordance with conditions explicitly agreed by 2809 them (see Guideline 5, points 18-20). The informed consent should be reviewed and 2810 approved by the ethical review committee responsible for the repository, in addition 2811 to any review required by an ethical review committee at the institution where the 2812 samples are collected.

2813 Three sources of stored samples are commonly in use: 2814 a. repositories of samples collected and stored with informed consent for long-2815 term, epidemiological studies (for example, so-called "population biobanks"); 2816 b. repositories of samples collected and stored in the context of a specific 2817 research [without explicit and fully-informed consent (in line with practices 2818 prevailing at the time)]; 2819 c. repositories of samples (typically surgically excised tissues, bioptic 2820 fragments, and leftover blood collected for diagnostic purposes) collected and 2821 stored in the context of routine clinical care or pathological or forensic 2822 examination.

2823 2824 a. Repositories of samples collected at present and stored for long-term, 2825 epidemiological studies. The value of repositories for longitudinal studies of specific 2826 diseases is now widely recognized; likewise, several large population biobanks are 2827 80 being established to allow studies across many diseases, through correlations of 2828 genetic, environmental, occupational, and other health data. Such repositories share an 2829 important characteristic: the persons whose samples are stored explicitly agree to this 2830 future use through an informed consent procedure approved by an ethical review 2831 committee.

However, since such future research inherently involves the testing of as-2832 yet unformulated hypotheses and the carrying out of analytical determinations 2833 unforeseeable at the time samples are collected, the information disclosed must of 2834 necessity lack much of the specificity usually expected in an acceptable informed 2835 consent process. 2836 The ideal and most direct way out of this dilemma is to seek from the 2837 participants a new consent each time a new hypothesis is going to be tested, a 2838 procedure which, though cumbersome, may be feasible in studies where participants 2839 are contacted and followed up at regular intervals (say, every one or two years).

Even 2840 this procedure, however, leaves out people who die in the interval, a feature that may 2841 seriously bias the study results; it will be up to the ethical review committee, notably 2842 in the light of the response obtained from the subjects who are actually requested to 2843 give a new consent, to advise for or against the use of the samples from deceased 2844 persons. A second best approach is to make the consent given at enrolment specific 2845 enough regarding the type of factors and health endpoints to be investigated in the 2846 future (even if any actual hypotheses cannot be indicated, being as yet unknown) to 2847 constitute the basis for a genuinely "informed" agreement on the part of donors.

This 2848 solution may be the only practicable one in studies where subjects are "passively" 2849 followed up, for instance through disease registries, but are not contacted by the 2850 investigators. A third conceivable solution would be consent to an open-ended 2851 donation of the sample to be used for biomedical and epidemiological research, 2852 conditional upon the approval of an ethical review committee. This solution is highly 2853 debatable and [likely to be] unacceptable under the ethical standards applied in 2854 several countries. It may also be deceivingly simple because it implies that, in order to 2855 give a current informed consent, the participant should in any case be made aware 2856 (unless he/she explicitly refuses) of the spectrum of studies that the blanket formula 2857 "biomedical or epidemiological research" encompasses and which kind of studies, if 2858 any, it excludes.

2859 In no case can a clearance given by an ethical review committee to establish a 2860 81 repository also be regarded as a clearance to carry out an actual study using the 2861 samples in the repository; a new clearance is required after scientific and ethical 2862 review of every specific study protocol. 2863 Especially in the context of repositories established for longitudinal study of a 2864 particular disease, the informed consent should clearly stipulate what return of 2865 information–if any–derived from analysis of the samples is foreseen, should the 2866 subject so wish. In general, information of uncertain scientific validity or meaning 2867 would not qualify for transmission to the participant.

It may also be reasonable to 2868 consider not all health-related information generated by the investigations conducted 2869 on the biological samples but only information potentially beneficial to the subject 2870 and/or his/her relatives, for example diagnostic information on gene variants or 2871 phenotypic traits established as relevant to health, particularly when amenable to 2872 some form of beneficial intervention or information on markers of an infectious 2873 disease or of a harmful environmental exposure, especially if avoidable. 2874 2875 b. Repositories of samples collected and stored in the past with no informed consent 2876 in the context of research.

When already existing repositories of biological samples 2877 collected and stored without an explicit consent procedure offer important and 2878 otherwise unobtainable data, an ethical review committee needs to decide whether the 2879 use of such samples is justified in the absence of explicit consent. Arguments 2880 pertaining to this decision are discussed within the more general framework of 2881 waiving of consent in the Commentary on Guideline 4, under the section "Waiver of 2882 consent requirements". 2883 2884 c. Repositories of routinely collected samples. Secondary use of samples collected in 2885 the context of clinical or preventive (screening) practice does not raise ethical 2886 objections if the informed consent by the patient makes clear that samples can also be 2887 used in the future for research purposes, provided these are explicitly specified.

Given 2888 the likelihood that such materials will be of interest to future researchers, it would be 2889 good clinical practice to insist that patients always be offered several options: to have 2890 their samples used only for their own treatment or benefit and then discarded; to allow 2891 stored samples to be used for research directly related to the condition for which they 2892 have been treated; or to allow stored samples to be used for unrelated research, with 2893 82 or without personal identifiers (as noted above, this blanket option would be 2894 unacceptable under the ethical standards applied in several countries).

These options 2895 may be presented during a conversation with the patient or in an information 2896 document upon admittance to the hospital. It should be made clear to persons, that it 2897 is reasonable to choose to "opt out" and that such a choice will not adversely affect 2898 the care provided to them. (Of course, if the person allows future studies using 2899 identifiable samples which then generates new information of definite clinical value 2900 to that person, ordinary good practice dictates that the person be contacted again even 2901 if years have elapsed in between). In any case, the person should be told that any 2902 research uses of the stored samples will be subject to approval by the relevant ethical 2903 review committee.

Consent for each study to be conducted with samples collected 2904 routinely without explicit consent for future research use must be sought. Only if this 2905 is unobtainable, such as if the patient proves after a reasonable attempt to contact him 2906 to be untraceable or is dead, may an ethical review committee consider the option of 2907 allowing the use of the samples for projects which cannot be carried out in alternative 2908 ways; these conditions are likely to hold, for example, for "historical" collections of 2909 samples stored when contemporary informed consent policies were not applied.

2910 2911 Genetic research. When individual consent or permission has not been obtained to 2912 perform a genetic test that is of known predictive value or that gives reliable 2913 information about a known heritable condition, the investigator must see that 2914 biological samples are fully anonymized and unlinked before performing the test; this 2915 ensures that no information about specific individuals can be derived from such 2916 research or passed back to them. 2917 When biological samples are not fully anonymized and when it is anticipated 2918 that there may be valid clinical or research reasons for linking the results of genetic 2919 tests to research subjects, the investigator in seeking informed consent should assure 2920 prospective subjects that their identity will be protected by secure coding of their 2921 samples (encryption) and by restricted access to the database, and should explain this 2922 process to them.

2923 (See also Guidelines 5, 6 and 7.) 2924 2925 83 Appendix 1: 2926 2927 Glossary 2928 2929 This glossary defines terms used in the text of the Guidelines and Commentaries. 2930 Several definitions are based on, or adapted from, those found in John Last‟s 2931 Dictionary of Epidemiology, 4 th ed. (Oxford University Press) to which the reader is 2932 more generally referred for terms encountered in epidemiological study protocols and 2933 reports. Within a definition italicized words refer to other terms found in the glossary. 2934 2935 Analytic study. An epidemiological study to test the hypothesis that a factor is the 2936 cause of an health effect, for instance that the factor causes a disease or that it 2937 prevents a disease.

The commonest types of analytic studies are case-control, cohort 2938 and cross-sectional studies. Analytic studies are contrasted with descriptive studies, 2939 which do not test hypotheses. In addition to these types of studies, all of which are 2940 observational, analytic studies also encompass interventional studies. 2941 2942 Anonymous. A record, biological sample or item of information that in no 2943 circumstance can be linked to an identified person. 2944 2945 Benefit. A favourable consequence arising from a study, for example the 2946 demonstration that a vaccine is effective in a randomized controlled trial or the 2947 identification of a workplace hazard in an observational study.

Benefits are often 2948 contrasted to "risks" (as in a “risk/benefit ratio”) but the term “risk” is ambiguous 2949 because it connotes both an adverse consequence and the probability of its occurrence 2950 (i.e., risk in the formal epidemiological meaning). To avoid this ambiguity, the term 2951 "risk" is better replaced by "harm" when the consequence is certain or has already 2952 occurred, or "potential harm" when it remains a possibility. In the context of planned 2953 research, the balance to be struck is thus between potential benefits (to society and 2954 possibly to the subjects) and potential harms (principally to subjects), paying attention 2955 both to the type and magnitude of these benefits and harms and the probability that 2956 they will occur.

Potential benefits and harms “to subjects" may not be restricted to 2957 them, but may extend to their family members or, more generally, to a group to which 2958 84 they belong. For instance, findings of a higher than average prevalence of certain 2959 genetic traits or diseases among study subjects may offer a means of early assessment 2960 and prevention (a benefit for the group of which they are a part) but may also 2961 stigmatize the family or the group in the eyes of others (a harm for the group). 2962 2963 Case-control study. An observational study comparing cases with a disease (for 2964 example, lung cancer) with non-diseased control subjects from the same population as 2965 the cases being studied.

The relationship of a factor (for example, tobacco smoking) 2966 to the disease (here, lung cancer) is examined by comparing how frequently the factor 2967 or its different levels (the number of cigarette smoked) is present among cases and 2968 among controls. Information about the factor(s) of interest be gathered by 2969 interviewing people or by consulting existing records, for example, prescription 2970 records for a study of adverse effects of a drug. 2971 2972 Cluster sampling. A method of selecting subjects from a population in which each 2973 unit selected is a group of subjects (e.g., all children in a school or all people in a 2974 town district) rather than an individual.

Clusters are usually selected through random 2975 sampling. 2976 2977 Cohort study. An observational study in which the occurrence of a disease or other 2978 health condition is recorded in any designated group of subjects who are followed up 2979 over a period of time, usually years or, in some studies, decades. At the start of the 2980 observation, the subjects are classified according to the factor(s) whose relation with 2981 the disease is being investigated. For example, blood pressure may be used to classify 2982 subjects in a study of coronary heart disease; the study would consist of comparing 2983 the frequency with which coronary heart disease occurred subsequently in subgroups 2984 of subjects with different blood pressure levels.

In some cohort studies, the subjects 2985 are contacted and asked questions and/or undergo measurements and blood tests by 2986 the investigator at the time of enrolment in the cohort and at fixed intervals thereafter, 2987 while in other studies the cohort can be formed using existing records (e.g., hospital 2988 or employment records) with no technical need to contact the subjects. 2989 2990 85 Competent person. A person capable of understanding the meaning of the 2991 information she is presented with and of taking decisions based on it. Certain persons, 2992 such as children up to a specified age are typically deemed by the law to be legally 2993 incompetent, while others, including people whose mental capacity or thought 2994 processes are impaired by mental or physical illness, can be found by a court or other 2995 body to be incompetent to make some or all decisions.

2996 2997 Control (noun and adjective). Designates the group of subjects against which the 2998 group(s) of subjects of interest in a study are compared. For example in a case-control 2999 study the subjects with the disease of interest, say lung cancer, may be compared with 3000 subjects without the disease, the control or reference group, to find out whether the 3001 former were more frequently exposed than the latter to carcinogenic fumes. In a 3002 randomized controlled trial (RCT) of a new drug, the subjects given the intervention 3003 being studied are compared with the “control” subjects who receive a routinely used 3004 drug or, under certain circumstances, a placebo.

3005 3006 Control (verb). In public health, “to control” means to prevent a disease (or its causal 3007 factors) or to treat it .A disease which can be prevented or treated, or both, is 3008 “controllable”. In the analysis of an epidemiological study, control means to remove 3009 the influence of those factors such as age and gender that may be differently 3010 distributed in two groups of subjects which are being compared so as to avoid having 3011 those factors distort the comparison of the two groups, for instance of their respective 3012 death rates. 3013 3014 Cross-sectional study. An observational study in which the presence of a disease (or 3015 other health condition) and the presence of factor(s) of interest are simultaneously 3016 ascertained at a point in time in order to examine their relationship.

The ascertainment 3017 is often carried out in random representative samples of a population. For example, a 3018 factor such as blood pressure and a health condition as defined by an 3019 electrocardiogram may be measured in subjects selected at random within each age- 3020 and sex-specific stratum of a population. 3021 3022 86 Descriptive study. An observational study portraying the occurrence of a disease or 3023 of other health-related events in relation to geographical areas, calendar periods and 3024 demographic characteristics of populations, such as age, sex, educational level, 3025 occupation, socioeconomic conditions, etc.

These studies can be carried out as “ad 3026 hoc” research investigations or as institutional and regular activities of disease 3027 surveillance within public health practice. In both contexts they contribute to 3028 generating hypotheses on the factors potentially determining the observed disease 3029 patterns. These hypotheses can then be tested in analytic studies whose results may in 3030 turn be used to verify how much the factors account for the disease patterns. 3031 Descriptive studies usually make use of individual records as available in existing 3032 databases or registries (of deaths, of notifiable communicable diseases, of cancer, etc.) 3033 and do not require identification of the persons to whom the records belong.

3034 3035 Factor. Generically any event, characteristic or other definable entity potentially or 3036 actually capable of affecting health or contributing to a health-related condition. 3037 Factors include age, sex, body characteristics (such as height, weight, blood pressure, 3038 genetic traits, etc.), economic status, occupation, residence, and a wide range of 3039 personal behaviour and environmental causes external to the body including diet, 3040 drugs, etc. 3041 3042 Genetic epidemiology. The branch of epidemiology dealing with biologically 3043 inherited causes of health and diseases.

It is a bridging discipline between 3044 epidemiology and genetics, and it encompasses the study of the interactions between 3045 genes and environmental factors in disease causation. 3046 3047 Harm. An adverse consequence arising from a study, as opposed to a benefit. A 3048 potential harm is often referred to as a risk,” but that term is ambiguous because it 3049 encompasses both the magnitude and the probability of a harm occurring. 3050 3051 Information. Items of knowledge contained in materials, namely records (e.g., from 3052 hospitals, interviews, recorded measurements on people, etc.) or biological samples 3053 which can be tested in the laboratory for a variety of components.

Records and 3054 biological samples may or may not be identified as belonging to a particular person 3055 87 and may or may not be linked to each other for the purpose of a study. The 3056 combinations of these different possibilities in various contexts (epidemiology, 3057 clinical trials, and genetic research) have been classified and labeled in different ways. 3058 In the present document, two major categories of information and materials have been 3059 utilized: (personally) identifiable information, which refer to, or can provide a link to, 3060 a particular person and (personally) non-identifiable information, which cannot be 3061 linked to a person.

The two types of information respectively derive from (personally) 3062 identifiable material and (personally) non-identifiable material . 3063 3064 Identifiable material. This includes three types of materials: 3065  Nominal record or sample: records and samples that carry the person's name 3066 or unique identifier, such as a social security number. 3067  Linked, coded record or biological sample: a record or sample that does not 3068 carry a name but is coded and thus, by possessing or by "breaking" the coding 3069 system, could be linked to the person to whom the record refers or from whom 3070 the sample was obtained.

Depending on the circumstances, the code may be 3071 known only to the person concerned or the key to the code may be held by the 3072 person who collected the material (such as the physician of the person 3073 concerned), by the repository where the record or sample is held, and/or by an 3074 investigator who is using the material in a study. 3075  Linked, double-coded record or sample: similar to a linked, coded record or 3076 biological sample except that two different codes are used for each record or 3077 sample; one key, which connects the codes on different samples and records 3078 (and allows data derived from analysing samples to be compared to data from 3079 records), is created by the repository and used by investigators, while a 3080 separate coding system that links each record or sample to the person 3081 concerned is held by a third party (such as the physician who submitted the 3082 record or sample) and is not available to the investigator.

Although double-3083 coding makes linking samples or records to a particular person much more 3084 difficult, the existence of the codes means that such linkage might occur, 3085 either accidentally or through diligent effort. 3086 3087 88 Intervention. An intentional change induced by the investigator in the status of the 3088 study subjects in order to investigate its effects on health. Examples are the 3089 administration of a drug, vaccine, or health education programme. In contrast, 3090 procedures used to acquire data, such as administering a questionnaire, conducting an 3091 interview, taking a blood sample or performing an X-ray, are not regarded as 3092 “interventions” in the technical sense because they are not performed in order to 3093 produce a measurable effect on the subject.

3094 3095 Interventional or intervention study. An epidemiological study based on an 3096 intervention; synonymous with “experimental study”. Such studies test the effects of 3097 interventions (often termed "treatments" in the technical literature, not to signify that 3098 they are therapeutic but that they change the circumstances) which are assigned to 3099 subjects in a population following a study protocol. For example, an intervention 3100 would be a screening test for early recognition and management of a disease to be 3101 compared with no screening or with screening with lesser frequency; or a treatment 3102 could be a vaccine to prevent a disease of viral origin to be compared with no vaccine 3103 or a different vaccine.

Whenever possible, subjects are assigned interventions at 3104 random (a randomized controlled trial). Random allocation means that, other than the 3105 intervention itself, all possibly relevant factors (both those already known to affect the 3106 outcomes being studied and those not yet identified) are on average equally 3107 distributed between groups receiving the different modalities; consequently, assuming 3108 the sample size is large enough to yield statistically significant results, random 3109 allocation ensures that any observed difference in outcomes can be confidently 3110 regarded as a real effect of the intervention.

3111 3112 Investigation. A study carried out for research purposes. It may also denote a study 3113 carried out for clinical diagnostic purposes and, sometimes, a specific diagnostic 3114 procedure (e.g., a breast echography, colonoscopy, or CT investigation). 3115 3116 Linked, coded record or biological sample. A type of identifiable material. 3117 3118 Linked, double-coded record or biological sample. A type of identifiable material. 3119 3120 89 Minimal risk. In this expression “risk” is taken in its common meaning of a possible 3121 but not certain adverse effect (on health). Minimizing risk implies reducing to the 3122 feasible minimum the number and magnitude of such possible effects as well as the 3123 probability that they will occur.

A study is often said to involve “minimal risk” when 3124 the potential harms involved are comparable to those as experienced in “ordinary life” 3125 by a person of a given age and gender or by an apparently healthy person undergoing 3126 routine medical surveillance. 3127 3128 Molecular epidemiology. The use in epidemiological studies of techniques of 3129 molecular biology, better understood as a level and method of measurement rather 3130 than a branch of epidemiology with substantive research content. 3131 3132 Nominal record or sample. A type of identifiable material. 3133 3134 Non-identifiable material.

Includes unlinked records or biological samples that were 3135 either collected on an anonymous basis or have been made anonymous (anonymized) 3136 in such a way that they do not carry any direct or indirect personal identifier . For 3137 these materials, no link is possible between the records or samples and the identity of 3138 the person who was the source of the record or sample. 3139 3140 Observational study. Synonymous of non-experimental study. An epidemiological 3141 study that does not involve an intervention. Observational studies have a wider range 3142 of applicability than intervention studies as they can be employed to investigate both 3143 putative hazardous or beneficial factors (e.g., in the environment, in diet), whereas, 3144 for obvious ethical reasons, intervention studies are typically limited to potentially 3145 beneficial factors.

The results of observational studies, however, cannot usually be 3146 regarded with the same degree of confidence than the results from intervention 3147 studies. In observational studies the groups differently exposed to a factor (for 3148 example subjects with a high and low consumption of fats) may also differ in other 3149 factors, some of which are unknown and uncontrollable and may be the real cause of 3150 an observed effect (for example myocardial infarction). Therefore no single study can 3151 as a rule be regarded as providing firm evidence on the causal role, either hazardous 3152 or protective, of a factor.

Multiple studies, carried out in different settings and 3153 90 producing consistent results, are necessary and should therefore not be considered as 3154 redundant or unethical. 3155 3156 Placebo. An inert medication or procedure given to “please” subjects so that they 3157 think they are receiving an active treatment for their condition. The effects, beneficial 3158 and sometimes even adverse, observed following the administration of a placebo are 3159 usually attributed to psychological processes (e.g., “the power of suggestion”). 3160 3161 Publicly available record or information. Any record or information, whether 3162 carrying personal identifiers or not, that the law treats as publicly accessible, such as a 3163 telephone directory, registry of deaths, or, in a number of countries, the register of 3164 nominal tax records.

Since anyone can use these records, no special authorization or 3165 permission of any type - legal and/or ethical - is required for epidemiologists to 3166 consult them. 3167 3168 Random allocation, random assignment or randomization. Allocation of subjects 3169 to groups, for example to two pharmacological treatments, by a procedure that gives 3170 each subject the same probability of being assigned to either of the groups. Nowadays 3171 this is usually implemented by the use of a computer-generated sequence of random 3172 numbers; for example, each successive subject would assigned to one intervention if 3173 the corresponding random number is an even number and to the other if it is an odd 3174 number.

Random allocation guarantees that all factors capable of influencing the 3175 study outcome (e.g., disease duration), other than the intervention being studied, are 3176 on average equally distributed between the two groups. Random allocation is the 3177 defining feature of a randomized controlled trial. 3178 3179 Randomized controlled trial (RCT). An intervention study involving random 3180 allocation of the subjects to different treatment modalities (factors); “randomized 3181 population trial” or “randomized prophylactic trial” are equivalent terms used for 3182 trials carried out to test a preventive measure in a healthy population.

3183 3184 91 Random sampling. A method of selecting units from a population in which each unit 3185 of the population has a known probability of selection. The unit can be the individual 3186 or, in cluster sampling, a group of individuals. 3187 3188 Register and Registry. A register is an ordered collection of records, for instance of 3189 births or of deaths. A registry is an organized system to develop, maintain and use one 3190 or more registers, for example a national registry may keep the registers of births and 3191 deaths. By extension the institution responsible for the system is also often called a 3192 registry (e.g., a cancer registry).

3193 3194 Risk. The probability that an event, favourable or adverse, will occur within a defined 3195 time interval. Although often contrasted to benefit (as in a “risk/benefit ratio”), the 3196 term “potential harm” is better for that context, leaving “risk” in its formal 3197 epidemiological sense to express the probability of a (typically adverse) event or 3198 outcome. 3199 3200 Social epidemiology. The branch of epidemiology dealing with socially relevant 3201 variables in relation to health. These variables characterize either the place of persons 3202 in society (e.g., gender, education, income, profession) or the structure and function of 3203 social institutions (e.g., family, school, government).

3204 3205 Trial. A generic term that in a clinical context denotes a research activity involving 3206 the administration of an intervention to humans to evaluate its safety and efficacy. 3207 3208 Unlinked record or biological sample. A non-identifiable material. 3209 3210 92 Appendix 2. 3211 3212 Items to be included in a protocol (or associated documents) for 3213 epidemiological research involving human subjects 3214 3215 This comprehensive checklist essentially reproduces Appendix 1 of the International 3216 Ethical Guidelines for Biomedical Research Involving Human Subjects. Since 3217 interventional epidemiological studies, such as a population-randomized controlled 3218 trial of a new vaccine, are similar to biomedical trials, this checklist is applicable;

3219 however, in observational studies, a number of items will not be relevant. In all cases 3220 it is up to the principal investigator to judge which items are pertinent - and to what 3221 extent - to a given study; likewise, the ethical review committee must be satisfied that 3222 the items included meet the requirements of the present Guidelines. 3223 3224 1. Title of the study; 3225 3226 2. A summary of the proposed research in lay/non-technical language; 3227 3228 3. A clear statement of the justification for the study, its significance in development 3229 and in meeting the needs of the country/population in which the research is carried 3230 out;

3231 3232 4. The investigators` views of the ethical issues and considerations raised by the 3233 study and, if appropriate, how it is proposed to deal with them; 3234 3235 5. Summary of published studies and of ongoing research pertinent to the topic, 3236 including relevant animal, preclinical and clinical studies; 3237 3238 6. A statement that the principles set out in these Guidelines will be implemented; 3239 3240 7. An account of previous submissions, if any, of the protocol for ethical review and 3241 their outcome; 3242 3243 8. A brief description of the site(s) where the research is to be conducted, including 3244 93 information about the adequacy of facilities for the safe and appropriate conduct 3245 of the research, and relevant demographic and epidemiological information about 3246 the country or region concerned;

3247 3248 9. Name and address of the sponsor; 3249 3250 10. Names, addresses, institutional affiliations, qualifications and experience of the 3251 principal investigator and other investigators; 3252 3253 11. The objectives of the study, its hypotheses or research questions, its assumptions, 3254 and its variables; 3255 3256 12. A detailed description of the design of the study, including whether it is an 3257 observational or interventional study, and if the latter, a description, among other 3258 things, of how subjects will be assigned to treatment groups (including the 3259 method of randomization, if used), and whether the study will be blinded (single 3260 blind, double blind) or open;

3261 3262 13. The number of research subjects needed to achieve the study objective, and how 3263 this was statistically determined; 3264 3265 14. The criteria for inclusion or exclusion of potential subjects, and justification for 3266 the exclusion of any groups on the basis of age, sex, social or economic factors, or 3267 for other reasons; 3268 3269 15. The justification for involving as research subjects any persons with limited 3270 capacity to consent or members of vulnerable social groups, and a description of 3271 special measures to minimize risks and discomfort to such subjects; 3272 3273 16.

The process of recruitment, e.g., advertisements, and the steps to be taken to 3274 protect privacy and confidentiality during recruitment; 3275 3276 17. Description and explanation of any interventions (the method of treatment 3277 94 administration, including route of administration, dose, dose interval and 3278 treatment period for investigational and comparator products used); 3279 3280 18. When relevant, the plans and justification for withdrawing or withholding 3281 standard measures in the course of the research, including any resulting risks to 3282 subjects; 3283 3284 19. Any other treatment that may be given or permitted, or contraindicated, during 3285 the study;

3286 3287 20. Clinical and laboratory tests and other tests that are to be carried out on subjects 3288 or on biological samples obtained from the subjects; 3289 3290 21. The standardized case-report forms to be used, description and evaluation of the 3291 methods and frequency of measurement in gathering data from subjects, 3292 follow-up procedures, and, if applicable, the measures proposed to determine the 3293 extent to which subjects actually use or are exposed to the intervention; 3294 3295 22. Rules or criteria according to which subjects may be removed from the study or 3296 clinical trial, or, in a multi-centre study, a centre may be discontinued, or the 3297 study may be terminated;

3298 3299 23. Methods of recording and reporting adverse events or reactions, and provisions 3300 for dealing with complications; 3301 3302 24. The known or foreseen risks of adverse reactions, including the risks attached to 3303 each proposed intervention and to any drug, vaccine or procedure to be tested; 3304 3305 25. For research carrying more than minimal risk of physical injury, details of plans, 3306 including insurance coverage, to provide treatment for such injury, including the 3307 funding of treatment, and to provide compensation for research-related disability 3308 or death; 3309 3310 95 26.

Provision for continuing access of subjects to the intervention after the study, 3311 indicating its modalities, the individual or organization responsible for providing 3312 it or paying for it, and for how long it will continue; 3313 3314 27. For research on pregnant women, a plan, if appropriate, for monitoring the 3315 outcome of the pregnancy with regard to both the health of the woman and the 3316 short-term and long-term health of the child; 3317 3318 28. The potential benefits of the research to subjects and to others; 3319 3320 29. The expected benefits of the research to the population, including new knowledge 3321 that the study might generate;

3322 3323 30. The means proposed to obtain individual informed consent and the procedure 3324 planned to communicate information to prospective subjects, including the name 3325 and position of the person responsible for obtaining consent; 3326 3327 31. When a prospective subject is not capable of informed consent, satisfactory 3328 assurance that permission will be obtained from a duly authorized person, or, in 3329 the case of a child who is sufficiently mature to understand the implications of 3330 informed consent but has not reached the legal age of consent, that knowing such 3331 child‟s agreement, or assent, will be obtained, as well as the permission of a 3332 parent or a legal guardian or other duly authorized representative;

3333 3334 32. An account of any economic inducements or other remuneration to prospective 3335 subjects for participation, and of any financial obligations assumed by the 3336 subjects, such as payment for medical services; 3337 3338 33. Plans and procedures, and the persons responsible, for communicating to subjects 3339 information arising from the study (on harm or benefit, for example), or from 3340 other research on the same topic, that could affect subjects‟ willingness to 3341 continue in the study; 3342 3343 96 34. Plans to inform subjects about the results of the study; 3344 3345 35. The provisions for protecting the confidentiality of personal data, and respecting 3346 the privacy of subjects, including the precautions that are in place to prevent 3347 disclosure of the results of a subject's genetic tests to immediate family relatives 3348 without the consent of the subject;

3349 3350 36. Information about how the code, if any, for the subjects' identity is established, 3351 where it will be kept and when, how and by whom it can be broken in the event 3352 of an emergency; 3353 3354 37. Any foreseen further uses of personal data or biological materials; 3355 3356 38. A description of the plans for statistical analysis of the study, including plans for 3357 interim analyses, if any, and criteria for prematurely terminating the study as a 3358 whole if necessary; 3359 3360 39. Plans for monitoring the continuing safety of drugs or other interventions 3361 administered for purposes of the study or trial and, if appropriate, the appointment 3362 for this purpose of an independent data-monitoring (data and safety monitoring) 3363 committee;

3364 3365 40. A list of the references cited in the protocol; 3366 3367 41. The source and amount of funding of the research: the organization that is 3368 sponsoring the research and a detailed account of the sponsor's financial 3369 commitments to the research institution, the investigators, the research subjects, 3370 and, when relevant, the community; 3371 3372 42. The arrangements for dealing with financial or other conflicts of interest that 3373 might affect the judgement of investigators or other research personnel: informing 3374 the institutional conflict-of-interest committee of such conflicts of interest; the 3375 communication by that committee of the pertinent details of the information to the 3376 97 ethical review committee; and the transmission by that committee to the research 3377 subjects of the parts of the information that it decides should be passed on to 3378 them;

3379 3380 43. The time schedule for completion of the study; 3381 3382 44. For research that is to be carried out in a developing country or community, any 3383 contribution that the sponsor will make to capacity-building for scientific and 3384 ethical review and for biomedical research in the host country, and an assurance 3385 that the capacity-building objectives are in keeping with the values and 3386 expectations of the subjects and their communities; 3387 3388 45. Particularly in the case of an industrial or commercial sponsor, a contract 3389 stipulating who possesses the right to publish the results of the study, and a 3390 mandatory obligation to prepare with, and submit to, the principal investigators 3391 the draft of the text reporting the results;

3392 3393 46. In the case of a negative outcome, an assurance that the results will be made 3394 available, as appropriate, through publication or, if relevant to the type of study, 3395 by reporting to the drug registration authority; 3396 3397 47. Circumstances in which it might be considered inappropriate to publish findings, 3398 such as when the findings of an epidemiological, sociological or genetics study 3399 may present risks to the interests of a community or population or of a racially or 3400 ethnically defined group of people, and the procedures by which such a 3401 determination would be made; and 3402 3403 48.

A statement that any proven evidence of falsification of data will be dealt with in 3404 accordance with the policy of the sponsor or of the legal authorities to take 3405 appropriate action against such unacceptable procedures. 3406 3407 98 Appendix 3. 3408 3409 WORLD MEDICAL ASSOCIATION 3410 DECLARATION OF HELSINKI 3411 3412 Ethical Principles for Medical Research 3413 Involving Human Subjects 3414 3415 3416 Adopted by the 18th WMA General Assembly, Helsinki, Finland, June 1964, and amended by the 29th WMA General Assembly, Tokyo, Japan, October 1975 35th WMA General Assembly, Venice, Italy, October 1983 41st WMA General Assembly, Hong Kong, September 1989 48th WMA General Assembly, Somerset West, Republic of South Africa, October 1996 and the 52nd WMA General Assembly, Edinburgh, Scotland, October 2000 Note of Clarification on Paragraph 29 added by the WMA General Assembly, Washington 2002 Note of Clarification on Paragraph 30 added by the WMA General Assembly, Tokyo 2004 A.

INTRODUCTION 1. The World Medical Association has developed the Declaration of Helsinki as a statement of ethical principles to provide guidance to physicians and other participants in medical research involving human subjects. Medical research involving human subjects includes research on identifiable human material or identifiable data. 2. It is the duty of the physician to promote and safeguard the health of the people. The physician's knowledge and conscience are dedicated to the fulfillment of this duty. 3. The Declaration of Geneva of the World Medical Association binds the physician with the words, "The health of my patient will be my first consideration," and the International Code of Medical Ethics declares that, "A physician shall act only in the patient's interest when providing medical care which might have the effect of weakening the physical and mental condition of the patient." 4.

Medical progress is based on research which ultimately must rest in part on experimentation involving human subjects. 5. In medical research on human subjects, considerations related to the well-being of the human subject should take precedence over the interests of science and society. 6. The primary purpose of medical research involving human subjects is to improve prophylactic, diagnostic and therapeutic procedures and the understanding of the aetiology and pathogenesis of disease. Even the best proven prophylactic, diagnostic, and therapeutic methods must continuously be challenged through research for their effectiveness, efficiency, accessibility and quality.

7. In current medical practice and in medical research, most prophylactic, diagnostic and therapeutic procedures involve risks and burdens. 8. Medical research is subject to ethical standards that promote respect for all human beings and protect their health and rights. Some research populations are vulnerable and need special protection. The particular needs of the economically and medically disadvantaged must be recognized. Special attention is also required for those who cannot give or refuse 99 consent for themselves, for those who may be subject to giving consent under duress, for those who will not benefit personally from the research and for those for whom the research is combined with care.

9. Research Investigators should be aware of the ethical, legal and regulatory requirements for research on human subjects in their own countries as well as applicable international requirements. No national ethical, legal or regulatory requirement should be allowed to reduce or eliminate any of the protections for human subjects set forth in this Declaration. B. BASIC PRINCIPLES FOR ALL MEDICAL RESEARCH 10. It is the duty of the physician in medical research to protect the life, health, privacy, and dignity of the human subject. 11. Medical research involving human subjects must conform to generally accepted scientific principles, be based on a thorough knowledge of the scientific literature, other relevant sources of information, and on adequate laboratory and, where appropriate, animal experimentation.

12. Appropriate caution must be exercised in the conduct of research which may affect the environment, and the welfare of animals used for research must be respected. 13. The design and performance of each experimental procedure involving human subjects should be clearly formulated in an experimental protocol. This protocol should be submitted for consideration, comment, guidance, and where appropriate, approval to a specially appointed ethical review committee, which must be independent of the investigator, the sponsor or any other kind of undue influence. This independent committee should be in conformity with the laws and regulations of the country in which the research experiment is performed.

The committee has the right to monitor ongoing trials. The researcher has the obligation to provide monitoring information to the committee, especially any serious adverse events. The researcher should also submit to the committee, for review, information regarding funding, sponsors, institutional affiliations, other potential conflicts of interest and incentives for subjects. 14. The research protocol should always contain a statement of the ethical considerations involved and should indicate that there is compliance with the principles enunciated in this Declaration. 15. Medical research involving human subjects should be conducted only by scientifically qualified persons and under the supervision of a clinically competent medical person.

The responsibility for the human subject must always rest with a medically qualified person and never rest on the subject of the research, even though the subject has given consent. 16. Every medical research project involving human subjects should be preceded by careful assessment of predictable risks and burdens in comparison with foreseeable benefits to the subject or to others. This does not preclude the participation of healthy volunteers in medical research. The design of all studies should be publicly available. 17. Physicians should abstain from engaging in research projects involving human subjects unless they are confident that the risks involved have been adequately assessed and can be satisfactorily managed.

Physicians should cease any investigation if the risks are found to outweigh the potential benefits or if there is conclusive proof of positive and beneficial results. 18. Medical research involving human subjects should only be conducted if the importance of the objective outweighs the inherent risks and burdens to the subject. This is especially 100 important when the human subjects are healthy volunteers. 19. Medical research is only justified if there is a reasonable likelihood that the populations in which the research is carried out stand to benefit from the results of the research. 20. The subjects must be volunteers and informed participants in the research project.

21. The right of research subjects to safeguard their integrity must always be respected. Every precaution should be taken to respect the privacy of the subject, the confidentiality of the patient's information and to minimize the impact of the study on the subject's physical and mental integrity and on the personality of the subject. 22. In any research on human beings, each potential subject must be adequately informed of the aims, methods, sources of funding, any possible conflicts of interest, institutional affiliations of the researcher, the anticipated benefits and potential risks of the study and the discomfort it may entail.

The subject should be informed of the right to abstain from participation in the study or to withdraw consent to participate at any time without reprisal. After ensuring that the subject has understood the information, the physician should then obtain the subject's freely-given informed consent, preferably in writing. If the consent cannot be obtained in writing, the non-written consent must be formally documented and witnessed. 23. When obtaining informed consent for the research project the physician should be particularly cautious if the subject is in a dependent relationship with the physician or may consent under duress.

In that case the informed consent should be obtained by a well- informed physician who is not engaged in the investigation and who is completely independent of this relationship. 24. For a research subject who is legally incompetent, physically or mentally incapable of giving consent or is a legally incompetent minor, the investigator must obtain informed consent from the legally authorized representative in accordance with applicable law. These groups should not be included in research unless the research is necessary to promote the health of the population represented and this research cannot instead be performed on legally competent persons.

25. When a subject deemed legally incompetent, such as a minor child, is able to give assent to decisions about participation in research, the investigator must obtain that assent in addition to the consent of the legally authorized representative. 26. Research on individuals from whom it is not possible to obtain consent, including proxy or advance consent, should be done only if the physical/mental condition that prevents obtaining informed consent is a necessary characteristic of the research population. The specific reasons for involving research subjects with a condition that renders them unable to give informed consent should be stated in the experimental protocol for consideration and approval of the review committee.

The protocol should state that consent to remain in the research should be obtained as soon as possible from the individual or a legally authorized surrogate. 27. Both authors and publishers have ethical obligations. In publication of the results of research, the investigators are obliged to preserve the accuracy of the results. Negative as well as positive results should be published or otherwise publicly available. Sources of funding, institutional affiliations and any possible conflicts of interest should be declared in the publication. Reports of experimentation not in accordance with the principles laid down in this Declaration should not be accepted for publication.

C. ADDITIONAL PRINCIPLES FOR MEDICAL RESEARCH COMBINED WITH MEDICAL CARE 101 28. The physician may combine medical research with medical care, only to the extent that the research is justified by its potential prophylactic, diagnostic or therapeutic value. When medical research is combined with medical care, additional standards apply to protect the patients who are research subjects. 29. The benefits, risks, burdens and effectiveness of a new method should be tested against those of the best current prophylactic, diagnostic, and therapeutic methods. This does not exclude the use of placebo, or no treatment, in studies where no proven prophylactic, diagnostic or therapeutic method exists.

(See footnote) 30. At the conclusion of the study, every patient entered into the study should be assured of access to the best proven prophylactic, diagnostic and therapeutic methods identified by the study. (See footnote) 31. The physician should fully inform the patient which aspects of the care are related to the research. The refusal of a patient to participate in a study must never interfere with the patient-physician relationship. 32. In the treatment of a patient, where proven prophylactic, diagnostic and therapeutic methods do not exist or have been ineffective, the physician, with informed consent from the patient, must be free to use unproven or new prophylactic, diagnostic and therapeutic measures, if in the physician's judgement it offers hope of saving life, re-establishing health or alleviating suffering.

Where possible, these measures should be made the object of research, designed to evaluate their safety and efficacy. In all cases, new information should be recorded and, where appropriate, published. The other relevant guidelines of this Declaration should be followed. Note: Note of clarification on paragraph 29 of the WMA Declaration of Helsinki The WMA hereby reaffirms its position that extreme care must be taken in making use of a placebo-controlled trial and that in general this methodology should only be used in the absence of existing proven therapy. However, a placebo-controlled trial may be ethically acceptable, even if proven therapy is available, under the following circumstances: - Where for compelling and scientifically sound methodological reasons its use is necessary to determine the efficacy or safety of a prophylactic, diagnostic or therapeutic method; or - Where a prophylactic, diagnostic or therapeutic method is being investigated for a minor condition and the patients who receive placebo will not be subject to any additional risk of serious or irreversible harm.

All other provisions of the Declaration of Helsinki must be adhered to, especially the need for appropriate ethical and scientific review. Note: Note of clarification on paragraph 30 of the WMA Declaration of Helsinki The WMA hereby reaffirms its position that it is necessary during the study planning process to identify post- trial access by study participants to prophylactic, diagnostic and therapeutic procedures identified as beneficial in the study or access to other appropriate care. Post-trial access arrangements or other care must be described in the study protocol so the ethical review committee may consider such arrangements during its review.

3417 3418 102 Appendix 4. 3419 3420 Members of the CIOMS/WHO Core Group established 3421 to carry out drafting of the provisional text of Guidelines 3422 3423 VALLOTTON, Michel B. (Chairman) 3424 Professor, Swiss Academy of Medical Sciences and President, CIOMS 3425 3426 CAPRON, Alexander (Principal Rapporteur) 3427 Professor of Law and Medicine, University of Southern 3428 California, Los Angeles, CA, USA, formerly Director, Department of Ethics, 3429 Trade, Human Rights and Health Law, WHO 3430 3431 SARACCI, Rodolfo (Co-rapporteur) 3432 Director of Research in Epidemiology, National Research Council, Pisa, Italy 3433 and formerly Chairman, Ethical Review Committee, International Agency for 3434 Research on Cancer, WHO, Lyon, France 3435 3436 FARLEY, Tim 3437 Coordinator, Controlling Sexually Transmitted and Reproductive Tract Infections, 3438 Department of Reproductive Health and Research, WHO 3439 3440 FLUSS, Sev S.

(Secretary) 3441 Senior Adviser, CIOMS 3442 3443 IDÄNPÄÄN-HEIKKILÄ, Juhana E. 3444 Professor, Senior Adviser and formerly Secretary-General, CIOMS 3445 3446 KREUTZ, Gottfried 3447 Professor, Secretary-General, CIOMS 3448 3449 MATSUDA, Ichiro 3450 Professor, Science Council of Japan 3451 3452 PETERSON, Herbert 3453 Professor, Chair, Department of Maternal and Child Health, University of North 3454 Carolina, Chapel Hill, NC, USA, and formerly Coordinator, Promoting Family 3455 Planning, Department of Reproductive Health Research, WHO 3456 3457 RANTANEN, Jorma 3458 Professor, formerly Director-General, Institute of Occupational Health, Helsinki, 3459 Finland, and President, International Commission on Occupational Health 3460 3461 van DELDEN, Johannes J.M.

3462 Professor of Medical Ethics, University of Utrecht, Netherlands, Royal Netherlands 3463 Academy of Arts and Sciences 3464 3465 VAYENA, Effy 3466 Technical Officer, Department of Reproductive Health and Research, WHO 3467