Family Practice Advance Access originally published online on December 1, 2007
Family Practice 2007 24(6):527-528; doi:10.1093/fampra/cmm072
Editorial |
Research ethics approval: comprehensive mechanisms are essential but not available
a Centre for Biomedical Ethics
b Department of General Practice and Primary Care, Primary Care Clinical Sciences Building, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. Correspondence to Heather Draper; Email: h.j.a.draper{at}bham.ac.uk
It is now widely accepted that any research involving humans or identifiable human information should be undertaken according to a pre-specified protocol that is submitted for consideration, comment, guidance, and where appropriate, approval by a properly constituted ethical review committee.1 Similarly, recent legislation such as the Data Protection Act, Human Tissue Act and Mental Capacity Act emphasize the importance of ethical scrutiny of research on human participants, however, in the UK there is no single, national ethics committee that undertakes research appraisal.2
The process of ethical review for research involving National Health Service (NHS) patients or staff was formalized only fairly recently (in the history of research in UK). The NHS Research Ethics Service is administered by the National Research Ethics Service (NRES, previously COREC), but is only responsible for providing ethical review for research occurring within the NHS (i.e. involving NHS patients, their relatives or carers or NHS staff).3 The scrutiny of non-NHS research remains the responsibility of the funding organization or the host institution.2 Currently, there is no equivalent oversight body to NRES for non-NHS research ethics approval. Most universities have now initiated formal processes for ethical review of non-NHS research, prompted in part by the Economic and Social Research Council requirement for ethical approval for all projects it funds. It has been traditional for some disciplines, noticeably academic psychology and sociology, to insist on ethical approval for all research projects carried out by staff and students, but this practice had neither previously been adopted by all university departments nor been standard policy for all UK Universities or other institutions (e.g. charities).
The same ethical imperatives used to establish the process of ethical review in the NHS are applicable to any research involving human participants, or their tissue or information. Although NHS patients may be considered vulnerable, non-NHS patients may be no less vulnerable, and non-patients and workers in other organizations may be as prone to exploitation or harm during research as NHS staff and patients. Respect for the rights of individuals, including the need to protect personal information, should be accorded to all research participants, not only those involved in research within the NHS. It makes sense, then, to generalize the process of ethical review from the NHS to the wider community so why has a central ethical review committee not been introduced? There are probably many reasons, but primary among these is probably the simple fact that it is not always obvious whose responsibility it is to instigate the process of formal review. The NHS has a duty of care to all its patients and staff and the Government Social Research Unit's framework for ethical assurance of all social research conducted by or for the government is an attempt to produce an umbrella ethical framework for all social research conducted by government departments.2 But no one seemed to be prepared to assume a duty of care for the rest of the general public. Moreover, setting in place formal procedures costs money, as does ensuring their continuation and, until it seemed as though revenue might be lost though lack of procedure, there was little financial incentive for academic institutions, for example, to set up processes that their researchers probably felt simply impeded research anyway.
Of course all journals, not just medical ones, could have insisted many years ago on some proof of ethical approval before publishing research on humans or using personal data.4 This might have led to less research that did not have appropriate research ethics committee approval being published, although it may not necessarily have avoided the research being undertaken. The decision about what counts as evidence of ethical approval is not, however, straightforward. Where local committees have been instigated it can be difficult for journals to be certain that the committees giving approval are properly constituted and experienced. Medical journals face this problem with regard to research that does not originate from ‘developed’ [we have taken the term from the Committee on Publication Ethics (COPE)4] countries (see for instance, case 03/12 http://www.publicationethics.org.uk/cases/zerosixthree accessed on October 31, 2007). In such cases, COPE suggests that editors apply the standards for research ethics in developed countries—this effectively means that the editor, and presumably members of the editorial board, acts as a kind of ethics committee in retrospect, though it is not clear how rigorous this process should be. Should the editorial board determine ethical acceptability on the basis of the proposed publication alone or seek further documents such as the original protocol and participant information sheets? If not the latter then it would be relatively easy for unethical practices (e.g. in recruitment or through misleading information) to be hidden, deliberately or otherwise. The practice of editorial review may prevent the publication of research that falls short of the standards expected in developed countries, but it will not necessarily prevent the research from happening in the first place. Cynics might suggest that this does little more than to keep the journals' hands clean. It also assumes that the correct response to valuable but unethically obtained research is that it should never be published. This is debatable.5–9 Just as refusing to publish may deter bad practice, it could equally be argued that once research participants have given their time and their information (data, bodily samples, whatever), it would be unethical for the results of the research not to be published and disseminated. It also makes the assumption that all cases of unethical research are obvious, clear-cut breaches of ethics. Yet the debate in the British Medical Journal10 several years ago suggests that this is not the case. We do not always agree, for instance, about when there might be valid exceptions to the gaining of informed consent. And given that reasonable people disagree about this, it seems unfair on those who were the subjects of research, and potential research subjects of the future (because if the original research is not published then it is likely that further research subjects will need to be recruited to answer the original research question), not to mention the researchers themselves, not to publish research results that fall into the grey areas.
This issue of Family Practice publishes the results of a study that did not have ethical approval.11 It did not have ethical approval because there was no mechanism in place, we are told, to obtain this approval. NHS research ethics committees are not funded to look at research that falls outside of their remit—so why should they. So what were the researchers to do? This problem may be lessened now that academic institutions have taken seriously their responsibility to monitor research proposals that are to be carried out by their staff and students (by and large) but some potential researchers—those not connected to formal institutions or using NHS patients/staff—may still be left without access to an approval process. Some may decide to set up their own committees—but will these be acceptable to journals given that they may fall outside formal structures and who will vouch for the standards and qualities of such committees? Others may decide to do as these researchers did and apply the standards of developed countries as they perceive them.
We think that Family Practice made the right call in deciding to publish this study (neither of us was party to this decision in any way). But it is unlikely to be the last time that the discretion of journal editors is called for as projects initiated prior to July 2007 begin to report and other researchers feel forced to rely on their own discretion for want of a committee to gain approval from. Some may choose, as one frustrated researcher recently suggested, to include a single NHS patient so as to be eligible for NRES. We are left with a ‘disparate and confusing system for both researchers and potential research participants’.2 It is past time that we had a national organization that ensures that all research on human subjects is approved in a satisfactory and standard fashion.
Declaration
Funding: None.
Ethical approval: Not applicable.
Conflicts of interest: None declared.
Notes
Draper H and Wilson S. Research ethics approval: comprehensive mechanisms are essential but not available. Family Practice 2007; 24: 527–528.
References
1 World Medical Association Declaration Of Helsinki. Ethical Principles for Medical Research Involving Human Subjects 2004. http://www.wma.net/e/policy/b3.htm (accessed on October 31, 2007).
2 Parliamentary Office of Science and Technology. Ethical Scrutiny of Research (2005) Postnote Number 243 Parliamentary Copyright 2005. The Parliamentary Office of Science and Technology, 7 Millbank, London, UK.
3 National Research Ethics Service. Requirements for Research in the NHS. http://www.nres.npsa.nhs.uk/applicants/review/apply/research.htm#approval (accessed on October 31, 2007).
4 Guidelines on Good Publication Practice Committee on Publication Ethics (COPE). http://www.publicationethics.org.uk/guidelines (accessed on October 31, 2007).
5 Brackbill Y, Hellegers A. Ethics and the editors. Hastings Cent Rep (1980) 10:20–24.[Web of Science][Medline]
6 Godlovitch S. Forbidding nasty knowledge: on the use of ill-gotton information. J Appl Philos (1997) 14:1–17.[Medline]
7 McDonald A. Ethics and the editors: when should unethical research be published? Can Med Assoc J (1985) 133:803–805.[Medline]
8 Moe K. Should the Nazi data ever be cited. Hastings Cent Rep (1984) 14:5–7.[Web of Science][Medline]
9 Schafer A. The blood-stained data: can we use Nazi experimental data in good conscience? Can Doct (1987) 53:GP4–GP7.[Medline]
10 Doyal L, Tobias J. Informed Consent in Medical research (2001) London: BMJ Books.
11 Doshani A, Pitchforth E, Mayne C, Tincello D. Culturally sensitive continence care: a qualitative study among South Asian Indian women in Leicester. Fam Pract (2007) doi:10.1093/fampra/cmm058.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||