Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The concept of researchers needing to obtain explicit, informed consent from research subjects is a comparatively recent phenomenon of the mid-twentieth century. Resistance to the paternalistic medicine practised up to this point evolved as a result of increasing concern about the rights of the individual and as a reaction to the horrendous, unethical research undertaken in Nazi Germany and Japan during the Second World War.1 As a result of the war crimes trials, the Nuremberg Code was written: this was the first internationally accepted ethical code for human biomedical experimentation. Emphasis on the importance of voluntary informed consent to research was outlined in item 1 of the Code which deemed it “essential” in research on human beings.
The next major ethical code to be developed in this area was the World Medical Association’s Declaration of Helsinki of 1964 which has subsequently been revised several times between 1975 and 1989. This further emphasises the rights of the individual by stating that: “concern for the interests of the subject must always prevail over the interests of science and society.”
The value of the individual is embedded in the ethical principle—respect for autonomy—and is one of the four principles of biomedical ethics expanded as a common morality theory by Beauchamp and Childress.2 For autonomy to be preserved within the context of giving informed consent for research, the following conditions must be met: consent must be given freely (or voluntarily); sufficient information must be given to allow understanding of the issues involved (including the risks) for an informed decision to be made; and the person giving consent must be mentally competent to provide consent.
The legal position concerning consent to medical treatment is based in common law. Legally, for an adult, any unauthorised “touching,” including medical procedures, would constitute battery unless “the patient has an essential understanding of what is proposed and consents to it.”3 If insufficient information is disclosed a doctor might be liable in negligence.
Obtaining informed consent for research
The aim of medical research and randomised controlled trials in particular, is to provide rigorous and critical assessment of the claims of medical knowledge. Arguably, clinicians have an ethical obligation to participate in the development of medical science through randomised controlled trials to be able to answer important medical questions on the best methods of treating their patients. But the procedure of obtaining informed consent for clinical trials, although aimed at the protection of the individual and appealing to the notion of respect for autonomy, poses some difficulties.
The equipoise condition of randomised controlled trials involves the need to admit uncertainty as to which is the best treatment. Such apparent indecision can lead the patient to doubt the doctor’s ability, undermining the doctor–patient relationship, particularly when clinicians are inexperienced and/or lack time to explain the issues. There are times when in order to satisfy entry criteria, obtaining informed consent can be a rapid and brutal process. The supplying of information about complex medical arguments can give rise to misunderstanding, especially for those who are poorly educated and emotionally stressed. Such emotional stress provokes debate about whether the patient is capable of making rational decisions and providing informed consent in this situation. The procedure may also “coerce” people into participating in research—this may be at the level of persuasive argument, or the desire for a particular treatment that is unavailable ordinarily, or the fact that the requesting doctor is the person also treating the patient, or merely pressure on the clinician’s time. Thus the concept of voluntariness may be undermined.
Furthermore, the informed consent procedure is often considered to be a legal or bureaucratic necessity by patients, for the benefit of the doctors, rather than being an ethical process in respect of the autonomy of the individual.
For certain categories of people, deemed incompetent to provide such consent, including babies, young children, or those who are mentally handicapped, obtaining informed consent presents ethical and legal difficulties. For consent to be valid, as stated above, the patient must be competent to make a free and adequately informed decision. Clearly this is not possible for neonates and thus proxy consent for medical treatment of children is allowed. Proxy consent can be given by someone who has parental responsibility under the Children Act 1989—that is, married parents, single mothers, and anyone else who acquires parental responsibility by legal process.
The legal principle governing any decision which affects a child must be consistent with the statutory requirement laid down by section 1(1) of the Children Act. This stipulates that the paramount consideration must be the welfare of the child. It follows that research in children is permitted only if the child is subject to no more than minimal risk as a result of participation. Unless the child stands to benefit him or herself, the research is considered unacceptable.4Department of Health guidelines are consistent with the statutory requirements and state:
“Those acting for the child can only legally give their consent provided that the intervention is for the benefit of the child. If they are responsible for allowing the child to be subjected to any risk (other than one so insignificant as to be negligible) which is not for the benefit of that child, it could be said that they were acting illegally.”5
Randomised controlled trials represent something of a grey area in terms of whether the intervention will benefit the individual young child for whom proxy consent is requested. A situation of equipoise is prerequisite for the ethical condition of a randomised controlled trial which provides no certain benefit to the individual patient and, in fact, could harm the child as a result of potential side effects.
A further problem is that parents of babies on neonatal units are often vulnerable. Their consent is requested while their baby is under the care of the clinician making the request. The parents are highly anxious about the welfare of their newborn baby and the mother may still be recovering from giving birth. They often experience feelings of powerlessness, their medical knowledge may be poor, and they may not understand complex medical arguments put to them at a time of stress. Whether or not they agree to their baby being entered into a trial, they may experience guilt as to whether they made the correct decision if the baby dies. This is compounded by the fact that medical staff often have limited time to provide adequate information about the research for which consent to the baby’s participation is required. Thus even with respect to proxy consent it can be argued that the neonatal intensive care situation hinders consent being given freely, with adequate information and with competence to make an informed decision.
What does proxy consent really represent? For it to embody the the views of the baby is clearly unrealistic as the baby has recently been born and is not in a position to have an opinion about research and whether or not it would like to participate in it. At best it represents parental discretion and family values.
As it can be argued that in the neonatal context it is questionable as to whether the parents’ consent is really informed, the appropriateness of burdening them with the additional responsibility of decision making at a time of psychological stress warrants further consideration. One approach, which recognises the stress the parents are under, but also the need to balance this with the importance of randomised controlled trials in the improvement of clinical care, aims to protect parents from having to make extra decisions. Modi considers that it should be legitimate to randomise a baby without parental consent where “a trial sets out to compare two treatment strategies, each of which is regarded as acceptable clinical practice and each of which individually might be implemented without parental consent.”6
In reality this situation is rare, as it is unusual for both arms of a trial to encompass treatments which are acceptable clinical practice. Randomised controlled trials compare the best standard treatment with a new treatment which it is hoped might be more effective. The new treatment, however, might be worse, or as effective, or even better, but it may also be associated with unacceptable adverse events. It could be argued that parents need to be informed of such risks even at the expense of raising anxiety in the short term.
English law also requires consent to medical treatment (except in an emergency). Failure to inform the patient of the nature and purpose of the intended procedure exposes them to liability in trespass.7 A doctor is also required to disclose such risks as a reasonable body of responsible medical opinion would disclose, to render medical treatment lawful should claims against negligence be made.8
A second alternative approach might be to charge an independent, informed group of experts to assess the merits and shortfalls of proposed research without the emotional and stressful burden of being a parent of a sick neonate. A body which currently examines randomised controlled trials on behalf of potential trial participants is the Local Research Ethics Committee (LREC). This should ensure that the research has scientific merit, protects the subjects from harm, and maintains respect for the subject’s autonomy. Should this body be charged with scrutinising research proposals more extensively on behalf of subjects beyond a local level? Certainly review of neonatal research by LRECs should provide a vital safeguard against unethical research. But LRECs have their drawbacks. Many have heavy workloads9for unpaid members, most of whom, albeit committed to their duty to the LREC, have other demands on their time. This usually means that there is insufficient time for the review process and delays in dealing with research proposals. In multicentre research it is common for LRECs, because of the differing personalities, skills, and ethical philosophies among their membership, to proffer differing conclusions and recommendations.10 To a certain degree this is inevitable, but it is frustrating for the researcher and leads the sceptic to question the relevance of local ethical review.
Training LREC members and setting up central regional ethical committees to examine multicentre research are among the recommendations which have been made to promote efficiency and improve ethical review of research proposals.11 The Department of Health is currently considering how to improve the efficiency of the research ethical review system for multicentre research while maintaining the input and independence of LRECs. A further step which might be contemplated could be an assembly of specialist regional ethical committees (for neonatal research specifically), either local or central. This would have the advantage of fewer applications to deal with from every specialty and would provide expert knowledge in the field under review. The LREC, however, cannot know about the situation of the individual baby, its family and circumstances, and so cannot be in a position to shoulder the entire responsibility for the sanctioning of research; it has neither parental responsibility nor the ability to consider what would best serve that child’s welfare.
A different approach to the problem of adequate critical assessment and protection of the subjects in neonatal randomised controlled trials is to increase the involvement of the general public in research, and promote education about these trials.12-15 An increase in the general acceptability of randomised controlled trials would encourage recruitment and an informed public would be in a better position to give their consent if they knew enough about randomised controlled trials in general. This, however, is a long term approach and provides neither immediate nor complete solutions.
The question remains
The complex area of informed consent for neonatal randomised controlled trials is currently being studied in a European Union funded study coordinated by the University of Leeds, entitled “Is obtaining informed consent for neonatal research an elaborate ritual?—a European Study” (EURICON). Leading neonatologists, medical ethicists, legal experts and a European medical sociologist (26 people from 11 European countries) are collaborating to address this problem. In summary, the objectives of this research are to: investigate the extent to which European countries differ in the legal and ethical guidelines regarding the obtaining of informed consent for research on neonatal units; analyse the possibility of harmonisation; seek to develop proposals to that effect; determine the validity of the consent process from the viewpoint of the clinicians and also parents of babies requested to provide proxy consent while on a neonatal unit (by interviews); examine the European legal and ethical guidelines in the light of any practical difficulties in applying these guidelines.
The existing position with regard to informed consent for neonatal research is problematic. Does the requirement to obtain informed consent lead to low patient recruitment into clinical trials and therefore delays in answering important questions which might benefit children in general? The requirement to request informed consent has the consequence that vulnerable, stressed parents are burdened with sometimes complex information and asked to make a decision about their newborn baby under very difficult circumstances.16 In this situation it is arguable whether the proxy consent requested is even valid in terms of representing the view of the baby, is adequately informed, or freely given with emotional competence and with the welfare of the child as paramount.
I acknowledge the contributions of the following people in the development of this article: Peter Allmark, Andrew Bridgman, Julia Brown, Christopher Megone, Aurora Plomer.