Roger G Rawbone, MD
(Medical Secretary to the HSE Research Ethics Committee)
The author thanks the Chair and the members of the HSE Research Ethics Committee for their encouragement and constructive comment during the preparation of this paper.
The work of the Health & Safety Executive Research Ethics Committee are presented in the light of 100 studies submitted for consideration over its first six years of operation. Ethical issues which have arisen in the context of this non-therapeutic research which are considered by the committee to be of particular importance are highlighted and exemplified by specific case examples.
The aims of the Health and Safety Commission and Executive (HSC/E), whose existence and functions derive from the Health and Safety at Work etc. Act 1974, are to protect the health, safety and welfare of employees, and to safeguard others, principally the public, who may be exposed to risks from work activities.
Advancing occupational health is a key element within the HSC/E strategic plan (1) and research, by translating broad objectives from HSC's Health Risk Reviews into outputs that meet HSE's policy, technical or operational needs (2), forms an essential component of HSE's occupational health programme.
Research has always raised major ethical questions about the testing of new procedures, methods or hypotheses in human subjects for the furtherance of knowledge rather than for the individual's personal or particular gain. These questions often arise in the context of therapeutic medical research. Historically such therapeutic research, together with a predominantly paternalistic approach to medical care, fell within the umbrella of the doctor-patient relationship; but during the last thirty years, with the shift to patient autonomy, the consideration of ethical issues has become subject to more formal arrangements through ethics committees.
Many, if not the majority, of the research projects supported by the HSE are conducted outside of a clinical setting and involve approaches to individual members of a workforce or healthy volunteers. In 1993 the HSE established an internal Research Ethics Committee (REC) in order to ensure that accepted ethical principles for research (3, 4, 5, 6) were being applied in its own studies. The essential functions of the REC are thus to maintain ethical standards in HSE's research; to protect research participants from harm; to protect participant's rights; and, to provide reassurance and a measure of accountability to society.
This paper describes the HSE REC in the context of the first 100 studies submitted for consideration highlighting and discussing a number of issues particularly relevant to non-therapeutic research.
Between 1993 to 1999 the HSE REC has held 18 meetings and considered a total of 100 submitted projects. The committee is chaired by a member of the HSE board, the director of health (who is not a medical practitioner), and has 12 members together with a medical and administrative secretariat. Of the 12 members 5, from HSE, provide expertise in field Health & Safety inspection, the employment medical advisory service (EMAS), Health & Safety law, medical statistics and laboratory sciences; whilst 7 non HSE members include specialists in occupational health, medical research, occupational psychology, general medical practice, and nursing together with 2 lay representatives. The committee was set up with specific input from the British Medical Association and the Centre for Medical Law and Ethics at King's College, London.
The committee's primary function is to consider all research undertaken internally by HSE which involves human subjects. Externally funded research is expected to be reviewed by the external organisation's local ethics committee (e.g. university or hospital); however, the HSE REC will consider external research in some circumstances: where the funded organisation does not have access to an ethics committee; where the subject of research is an HSE specialisation (e.g. occupational hygiene); or where the research is multi-centre (although the formation MRECs (7) has to some extent removed this reason). Of the 100 projects 6 were submitted for consideration by organisations external to HSE.
Internal HSE research is usually initiated directly against the organisation's objectives or, occasionally, indirectly through a specific employee's work towards further qualification (e.g. membership of the faculty of occupational medicine or MSc in occupational health).
Examples of the wide range of projects considered include
In general projects may be considered as falling into one of two broad groupings - those which are laboratory based experimental studies where volunteers are exposed to situations outside of their normal experience, and those based in a work situation where workers agree to participate in studies which essentially look at aspects of the performance of their normal day to day work activities.
Eighty-six of the 100 projects have been considered by the full committee, 76 at meetings and 10 through postal comment. The remaining 14 projects were cleared outside the committee by the chair. Current committee policy is that proposals should all be discussed in committee; however, clearance by the chair is considered acceptable where the project involves no invasive procedure and no recording of personal data, or where the project is one of a generic series of projects the first of which has already been cleared in committee and the proposed project series declared. Because the committee meets, at most, four times a year, postal evaluation by members has been agreed where a project needs to proceed with minimal delay. In this situation, however, where members express concern, clearance is still deferred until the next formal committee meeting. On average the committee reviews 5 new projects at each of its meetings.
The outcome of committee discussion was to clear 12 of the 86 studies without condition; 55 were cleared conditional upon completion of a stated action, and 17 were deferred pending action and further consideration at a subsequent meeting; 1 study was rejected and 1 study withdrawn from the agenda, by the researcher, just prior to a meeting.
The commonest reason for a conditional clearance was the failure to use plain and effective English in the information given to volunteers and in the forms requesting consent. Projects which were deferred were, in general, those which raised specific questions over the scientific content of the proposal and where the committee considered that either the background for the project had not been adequately researched or that the study methodology was not able to meet the defined objectives for the work. The single study rejected by the committee involved the experimental exposure of volunteers, in an environmentally controlled chamber, to a substance where there was a statutory requirement for exposures to be kept 'as low as is reasonably practicable' (i.e. categorised with a Maximum Exposure Limit (8) implying that no safe threshold of exposure can be identified - for example a carcinogen). In this case the committee considered that the study objectives meant that this condition could not be met in the case of a deliberate experimental exposure in a normal volunteer.
The 100 studies submitted during 1993 - 1998 now have the following outcomes. Fifty-six have been completed and reported (either through internal report or peer reviewed publication); 16 were withdrawn after REC clearance (15 before starting); and 28 remain in progress. One study, as has been noted, was rejected.
The studies withdrawn after clearance by the committee warrant comment. The reasons given include:
The one study withdrawn after commencement was discontinued because of insufficient response from the potential study population.
When ethical clearance is secured for a project the research co-ordinator is asked to inform the committee of any relevant ethical issues which arise during the course of the study, and when the committee secretariat receives project reports researchers are explicitly asked whether there have been any issues which might relate to the ethical conduct of the research. Specific ethical issues arose during the course of 3 studies but in each case this had been identified and discussed by the committee with advice being provided to the researchers. One of these related to a finding by the researcher of working methods which contravened acceptable Health & Safety practice, thus raising the issue of breaking confidentiality; a second, also concerning confidentiality, related to a request from an employer for copies of video material identifying specific workers; and the third involved the provision of counselling for volunteers where information obtained through the research had potentially damaging health implications in terms of previous work exposures to toxic substances.
One of the primary considerations of an ethics committee is to ensure that research is conducted to acceptable ethical standards using, for example, the application of the four principles of biomedical ethics - respect for autonomy, non maleficence, beneficence and justice (9). In the case of therapeutic research there is, by definition, a potential benefit to the study subject against which the risks of participation can be balanced. This is not the case in non-therapeutic research which will usually be associated with some risk, even if very small. It is this difference which places an onus on REC's which consider non-therapeutic research to consider the proposals with particular care both from the point of view of the welfare and dignity of the research subject but also the validity of the research itself (10). The benefits of this research can only be seen in relation to the possible future benefits to those who may be involved in similar working practice or environments. The range of expertise of committee members necessarily has to be wide and cover the relevant occupational disciplines in order to ensure the scientific validity of the research proposed. Whilst, as with all ethics committees, the involvement of the lay representatives has to be emphasised (11) it is also important that the committee ensure a wide consultation and support for that research which is to be conducted in the workplace. Appropriate consultation both with employers, and with employees and employee representatives is particularly relevant in this context.
This paper will not discuss the issue of the validity of the research excepting to note that the committee have been concerned that proposals could not be considered in the category of futile research (12). All 17 projects deferred by the REC were for reasons relating to scientific aspects of the proposal. As previously noted this was generally because of an inadequately researched background or because of inappropriate methodology, usually relating to the proposed statistics.
The autonomy of a participant implies that he (or she) is able to act intentionally, freely (that is without controlling influence) and with understanding. It is important that individuals are not being treated merely as a means to achieve the research outcome. The Kantian imperative requires us to treat people as ends in themselves, as persons, and one way of doing this, of trying to ensure that we do not exploit others, is to ask for their consent to what we propose and to make sure that they have a real option to refuse. Consent therefore becomes informed consent where autonomy has driven a shift from the researcher's obligation to disclose information to the quality of a participants consent and understanding.
This has been referred to as autonomous authorisation defined such that "an informed consent occurs if and only if a subject with substantial understanding and in substantial absence of control by others intentionally authorises a professional to do something" (13). In this sense informed consent implies that the individual is competent to act, receives full disclosure, understands the disclosure, acts voluntarily and consents to the intervention. Added to this the autonomy of the individual implies a respect for that individuals privacy and thus a confidentiality of information.
These issues are given particular consideration by the HSE REC. The question as to whether normal, healthy volunteers, whose motives for participation in non-therapeutic research are (or at least should be) purely altruistic, merit any greater consideration than patients under treatment volunteering for research in a therapeutic clinical setting is open to debate. English law gives no express guidance in the form of statute or precedent about the level of information which researchers are required to disclose. The 'patient' who claims that he was entered into a trial without his consent or with inadequate information having been given to him must bring his case within a recognised civil or criminal action.
For an action in battery then in the context of medical treatment the courts have stated that as long as a patient gives a very general consent to an operation being undertaken then the doctor will not be liable (e.g. Chatterton v Gerson (14)). In other jurisdictions (e.g. Canada) a broader duty of disclosure has been recognised in non-therapeutic trials as exampled by Hall J in the case of Halushka v University of Saskatchewan who stated "The subject of medical experimentation is entitled to a full and frank disclosure of all the facts, probabilities and opinions which a reasonable man might be expected to consider before giving his consent" (15) Interestingly, however, this approach was subsequently said not to apply in the context of a therapeutic procedure (16).
For action in negligence there is again no obligation in English law to make full disclosure and, at least in clinical practice, the standard of disclosure set by the case of Sidaway v Bethlem Royal Hospital Governors (17) is that of a responsible body of medical opinion. If such a standard of disclosure applies in the context of therapeutic research, establishing an action in negligence may prove exceedingly difficult for the research participant. Brazier notes (18) that consent to participation in research is likely to be a more stringent, research-subject centred test than Sidaway. In the case of non-therapeutic research an action may be equally difficult, but it does leave the question as to necessity of full and complete disclosure open.
The HSE experience is that, in comparison with clinical Local Research Ethics Committees (LRECs), the attention given to aspects of informed consent for non-therapeutic research may not be applied as rigorously by LRECs as the HSE committee has chosen to apply them. There is an impression that, particularly where research is being conducted in the workplace under the supervision of a medical practitioner (occupational physician), there remains for LRECs an element of 'reassurance' through an assumed valid paternalistic approach. Although in workplace studies volunteers may be performing their normal work and not exposed to additional risk through this activity many of the studies involve questionnaire administration or biological monitoring, with the collection of blood or urine samples. In this context a questionnaire unnecessarily requesting a construction worker to provide his National Insurance number may be considered an invasion of privacy. It is all too easy, in the workplace situation, to ignore the need for informed consent or to 'hide' the procedures under the guise of health surveillance.
In the context of informed consent and subject autonomy the following issues have received particular attention by the REC.
Overall, changes to the information sheet and consent form were the commonest reasons for the committee to request changes to the proposals submitted. This was also the finding of a recent review of responses by LRECs to submitted proposals (19).
The language of the volunteer calling notice, the information sheet and the consent form have, as submitted, been particularly prone to the use of technical jargon. Whilst in many cases the researcher has justified this on the basis of volunteers being recruited from technical members of the laboratory workforce the committee has consistently taken the view that unless these documents are in plain and effective English one could not defend any subsequent argument of interpretation, nor could the documents be relied on to do their job.
Where biological samples are being taken the approach has been to ensure that the consent form clearly states what will not be measured as well as what will be measured. Thus whilst the latter may be stated in relatively broad terms (e.g. measurement relevant to the study objectives) participants can still explicitly know that there will be no measurement for (e.g.) HIV status or drug abuse.
Feedback of study results is also considered important and particularly where the results cannot be interpreted in isolation (e.g. where the study objective is to define a workplace population exposure range) the committee expect a full explanation of the study outcome to be presented to volunteers as part of the consent procedure. Furthermore, the fact that an individual's results may not be interpretable is not considered a reason for not offering them to the volunteer. Where results may be anticipated as giving rise to particular individual concerns, as may be the case where specific genetic phenotypes are being identified in relation to chemical detoxification, then there are particular concerns about the information and consent form and about the procedures needed to counsel volunteers once their result is made known to them (20). The issue of protecting the confidentiality of the results both from employers and from health professionals, without a participant's specific consent once the results are known, is a point of emphasis.
In relation to the confidentiality of subject information and employers then it is has to be realised that relevant information may be imparted indirectly simply through a subjects volunteering for study. Where the calling notice asks, for example, for volunteers who have a specific condition then by virtue of their volunteering this becomes 'public' knowledge. Such information may impact on the employability of individuals and their colleagues in that employers might use the information to identify individuals who are susceptible and exclude them from particular job opportunities.
One of the features of research conducted by the HSE is that unlike other research conducted on human subjects volunteers are not recruited through the health care system. Communication with the subjects health care providers is therefore an issue but the committee make it clear to subjects that their participation in the research and any results (or incidental findings) which may be relevant to their health can only be disclosed with that individuals consent. The committee will not endorse a bland a priori consent to release any relevant data but prefer communication with the G.P. through the subject.
The importance of ensuring a voluntary participation in studies includes effective avoidance of coercion, persuasion and manipulation.
In many of the studies being assessed by the HSE committee research participants are either employees within industry or HSE employees who work in the laboratory facilities. It is essential to ensure that there is no coercion through employees being given any impression that their employment may be in any way affected should they decline to volunteer for study. In the case of HSE employees the Royal College of Physicians (RCP) guidelines on 'Research on Healthy Volunteers' states (4) "Unless the study is educational it is normally undesirable to recruit students who are in close contact with the investigator, e.g. on his medical teaching 'firm' or in his class. This is because students are, or may feel, vulnerable to pressure from someone in a position to influence their careers, by assessment in an examination or otherwise." A comparable situation may be said to exist in the case of HSE employees which makes an independent ethical approval the more necessary. The opportunities for persuasion through a forceful presentation of the importance of a study should also be minimised by ensuring appropriate written communication for all studies.
The issue of manipulation, particularly through financial reward for participation (21), has received specific discussion by the HSE committee. The committees conclusion is that a two component reimbursement is acceptable in the form of an honorarium in recognition of taking part (to a maximum of £30.00) and a payment to cover documented expenses and loss of earnings.
The RCP report (4) states that "a risk greater than minimal is not acceptable where the research subjects are healthy volunteers" and this is the starting point of the HSE committees consideration of these issues.
Apart from expecting a full risk benefit analysis for each project submitted the major concerns of the committee relate to the potential for HSE laboratory employees repeatedly to volunteer for studies in which they have a perceived particular interest. Many of the studies submitted for consideration involve volunteers being subjected to controlled chemical exposures in a laboratory based facility in order to assess biological monitoring techniques or investigate the relative importance of different routes of absorption (e.g. respiratory or dermal) This potential repeated use of laboratory staff relates back to the previous discussion of coercion, but here the committee has been instrumental in ensuring that there are systems in place within HSE to both monitor and control the number of times a person may volunteer within a given time period, the length of time between experimental chemical exposures and the limits on the quantity of blood that may be drawn. Records are also kept of whether individuals are volunteering across different departments within the organisation or only for studies within their own department.
The other issue of interest in the context of subject welfare is that of no fault compensation. It has been noted that knowledge of a no fault compensation scheme may strengthen participant protection over and above informed consent and minimal risk; against this, however, is the threat to the protection of minimal risk through increasing the likelihood that researchers will be willing to engage in research that involves greater than minimal risk. In the case of internal research HSE, as a government agency, does not hold insurance against such events and so the option of no fault compensation does not arise. However, it is considered important that volunteers to studies within HSE are made aware that there is no compensation scheme against no fault injury and that a clear statement of this is included in the consent form.
A specific example will further illustrate a number of issues which may arise during a workplace study in relation to subject welfare. A study involving the investigation of leakage around face masks (respiratory protective equipment - RPE) in asbestos workers was conducted in a controlled setting where asbestos strippers, wearing RPE supplied by their employers, simulated their normal work activity. During the course of the study it became clear that the RPE being used had not been correctly maintained and that a number of the workers had not received adequate training in its use, both of which meant that there had been a potential high exposure to asbestos. The extent of this finding was unexpected and raised questions firstly over compliance of Health & Safety legislation by the employers of the research subjects and secondly over the need to provide the option of medical counselling for subjects when informed of their potential exposure to this toxic substance, the effects of which would not be manifest for a number of years.
During the past six years the HSE REC has built up a considerable experience of evaluating the ethical issues arising in non-therapeutic research within the field of Health & Safety. The importance of issues relating to the welfare and dignity of the research subject deserve particular attention. Where research is conducted within the workplace, it is important to ensure that subjects, and employers, are fully aware of the nature of the project and give a fully informed consent. Particular attention must be given to subject confidentiality where even participation may alert the employer to personal information. Health & Safety research raises specific ethical issues in relation to subject welfare which may involve at least compromising confidentiality where, during the conduct of the research, it transpires that employers are contravening Health & Safety practice. We consider that it is important that the expertise of specialised research ethics committees, such as that of the HSE, are presented and discussed in order to raise awareness of issues which may have a broader impact in the work of committees generally and to facilitate the development of a consistent approach to non-therapeutic research.
1. Health and Safety Commission Strategic Plan for 1999/2002. (1999) HSE Books.
2. Mainstream Research Market 1999/2000. (1999) HSE.
3. World Health Organisation. International Ethical Guidelines for Biomedical Research Involving Human Subjects. (1993) WHO, Geneva.
4. Royal College of Physicians of London. Research on Healthy Volunteers (1986) Journal of the Royal College of Physicians, 20 ?
5. Royal College of Physicians of London. Guidelines on the Practice of Ethics Committees in Medical Research Involving Human Subjects (1996) RCP, London.
6. The Faculty of Occupational Medicine. Guidance for Ethics for Occupational Physicians (1999) 5th edition, FOM, London.
7. Ethics Committee Review of Multi-centre Research (1997) NHS Executive HSG(97)23.
8. Occupational Exposure Limits Eh30/99. 1999 (HMSO)
9. Beauchamp, T.L. and Childress, J.F. Principles of Biomedical Ethics. 4th edition (1994) (New York: Oxford University Press).
10. Briefing Pack for Research Ethics Committee Members. (1997) Department of Health.
11. Local Research Ethics Committees (1991) HSG(91)5 Department of Health
12. Ethical Problems of Futile research. Grimley Evans, J J. Med. Ethics (1997) 23;5-6
13. Beauchamp, T.L. and Childress, J.F. op.cit., p143.
14. Chatterton v Gerson 1981 QB 432 at 443.
15. Halushka v University of Saskatchewan 1965 52 WWR 605
16. Zimmer v Ringrose 1981 28 AR 69
17. Sidaway v Bethlem Royal Hospital Governors 1985 AC 871
18. Brazier,M. Medicine, Patients and the Law 2nd edition 1992 pp418-419
19. Responses by four Local Research Ethics Committees to submitted proposals. Kent,G J. Med. Ethics. (1999) 25:274-277
20. Genetic Research and Ethics. Advisory Committee on Genetic Testing (Feb. 1999) Bull. Med. Eth. 21-24
21. What's the price of a research subject? approaches to payment for research participation. Dickert, N and Grady, C. N.E.J.M. (1999) 341:198-203