Guidance on ethical principles relating to research on healthy volunteers
Royal College of Physicians (1986)
All research involving healthy volunteers should be approved by an ethics committee.
All studies should be scientifically and ethically justified.
Confidentiality should be maintained.
No study on healthy volunteers should involve more than minimal risk and there should be full disclosure of risks.
There should be no financial inducement or any coercion that might persuade a volunteer to take part in a study against his better judgement. Payment should be related to expenses, inconvenience and discomfort, not risk.
The volunteer should be asked to give permission to the researcher for his general practitioner and, if appropriate, ‘a company or other medical officer’ to be contacted for details of past history. Where appropriate, he should be medically examined and be asked about relevant medical history. He should sign a consent form.
Any significant untoward event occurring during or after a study affecting a volunteer should be communicated to the general practitioner and appropriate medical action taken to safeguard the volunteer’s health.
The sponsor, whether this be a commercial organisation, university, NHS or other institution, should agree to pay compensation for any injury caused by participation in a research study without regard to proof of negligence.
The most recent publication from the Royal College of Physicians: Guidelines on the Practice of Ethics Committees in Medical Research with Human Participants, 4th edition (2007) does not make the same distinctions between research on patients and research on healthy volunteers, but the advice set out above remains valid.
Research on healthy volunteers often takes the form of a drug study. The Association of the British Pharmaceutical Industry (ABPI) in 1988 published its Guidelines for Medical Experiments in Non‐Patient Human Volunteers, which was revised in 2007. The guidance was again revised and published in 2012 as Guidance for Phase I Clinical Trials. Its recommendations are similar to those in the Royal College of Physicians’ report. Where research is proposed on a ‘captive audience’, e.g. medical students or employees, no one should be made to feel under an obligation to volunteer, nor should they be disadvantaged in any way by not volunteering. Volunteers may be rewarded in cash or in kind, but the amount should be reasonable and related to time, inconvenience and discomfort, not risk. The ABPI operates an ex gratia scheme whereby any healthy volunteer in a drug trial mounted by an ABPI member will receive compensation for any injury arising from the trial. The Medical Research Council and the Wellcome Trust have also published guidance on clinical trials.
In an occupational health setting it will be necessary also to obtain the agreement and support of the employer, and probably the trade unions. The OH physician must be very careful to ensure that any volunteers have freely consented to take part. He should make clear at the start his role as a research investigator.
Issues can arise in a workplace setting in relation to whether participation is truly ‘voluntary’. Manipulation of the employee must not be used and clarification of the relationship between the study team and the employer is often important in this regard. Due attention should be paid to the perception of workers and whether the presentation of the research might make them feel that their employment position could be affected adversely or favourably by their decision on whether to participate … Involving workers’ representatives or, in the NHS a Patient and Public Involvement Working Group, at an early stage in study development can be helpful in avoiding some of these pitfalls.
(Faculty of Occupational Medicine Ethics Guidance for Occupational Health Practice (2018))
The most recent edition of the Faculty of Occupational Medicine’s Ethics Guidance for Occupational Health (2018) includes a comprehensive chapter on occupational health research.
Drugs research using patients is regulated by the Medicines Act 1968. The Committee on Safety of Medicines must license human trials by granting a clinical trial certificate (or exemption from the need to hold such a certificate) and will only do so if satisfied with preliminary research and animal tests. Any adverse reactions must be reported to the committee. This does not apply to non‐therapeutic research because it is not regarded as the administration of a medicinal product. Contraception is defined as therapeutic.
Epidemiological research using data only is discussed in Chapter 3. The main legal problem here is compliance with the General Data Protection Regulation (GDPR) and the common law of confidentiality. In my experience, most occupational health research involves an investigation of data and thus confidentiality is of major importance. Consent must be obtained, but the use of identifiable patient information for research without consent may be allowed where it is not possible to use anonymised information and where seeking consent is not practical. Section 251 of the Health and Social Care Act 2006 allows the Secretary of State for Health to make regulations to set aside the common law duty of confidentiality for defined medical purposes. This falls within the remit of the Confidentiality Advisory Group of the Health Research Authority (Chapter 3).
Where data are obtained for one purpose, for example in clinical treatment, it may be used for research or statistical purposes, though a different purpose, by virtue of the second data protection principle in the GDPR. It is good practice for identifiable patient data to be anonymised or pseudonymised. Pseudonymisation is the separation of the patient health data from its identifiers, such that, for example, the researcher knows that a man of 35 working with vibrating tools and living in Manchester has been diagnosed with hand arm vibration syndrome but does not know that his name is John Smith. The OH physician who supplied that information to the researcher has a record of the patient’s identity, although the researcher does not, therefore the data is not completely anonymised and the GDPR applies. The treating physician should make the patient aware that his data may be used in research and offer him the opportunity to opt out if he wishes. Research and statistical data can be kept indefinitely but good practice is that it should be pseudonymised before being archived. Publication of the findings of research should not identify individual research subjects.
2.14 Expert witnesses
Where OH professionals give an opinion as an expert in legal proceedings they cannot be sued for defamation in respect of anything said in court or in documents produced in preparation for legal proceedings. This is for reasons of public policy. It is important that witnesses feel able to speak freely. But in General Medical Council v. Meadow (2007) the Court of Appeal overturned the ruling of a High Court judge that this immunity exempts doctors from fitness to practise proceedings before a professional body. Health professionals can be held guilty of professional misconduct if they give evidence to a court which is so incompetent and lacking an evidence base as to amount to negligence. Professor Roy Meadow, a paediatrician, was struck off the Medical Register for serious professional misconduct in respect of evidence he gave at the criminal trial of Sally Clark, who was convicted of murdering two of her infant children (though the conviction was later overturned when medical evidence which cast doubt on the prosecution’s case and should have been made available to the defence came to light). Professor Meadow misinterpreted the statistics relating to sudden infant death syndrome.
The appeal court also held by a majority that, although Professor Meadow had rightly been convicted by the GMC of professional misconduct, it was not serious enough to justify the striking from the register and his registration was restored.
Until 2011 the common law rule about the liability for negligence of an expert witness, which had stood for 400 years, was as stated by the House of Lords in Watson v. M’Ewan (1905). It was that expert witnesses were immune from liability for negligence in respect of their evidence in court and also their preliminary