Intended for healthcare professionals


Research with children

BMJ 2001; 322 doi: (Published 09 June 2001) Cite this as: BMJ 2001;322:1377

Paediatric practice needs better evidence—gained in collaboration with parents and children

  1. Rosalind L Smyth (r.l.smyth{at}, Brough professor of paediatric medicine
  1. University of Liverpool Institute of Child Health, Alder Hey Children's Hospital, Liverpool L12 2AP

    All those concerned with the clinical care of children have a responsibility to improve that care, and one way of achieving this is by research. The importance of research into normal childhood development and into the methods of promoting good health, together with studies of the aetiology, natural history, diagnosis, and treatment of childhood diseases is well recognised. So too are the perils of relying on information from studies conducted in adults and generalising these to children. However, undertaking clinical research in children presents unique challenges.1

    The most obvious of these are the ethical ones: the need to protect children from harm while respecting their autonomy and to obtain properly informed consent from parents, and, when possible, the children themselves. There are also methodological challenges. Outcome measures, developed and validated in adults, are unlikely to be appropriate or feasible for children and babies. The proportions of children affected by chronic diseases are smaller and the diseases often more heterogeneous than in adults, and diagnostic criteria may be less precise.1

    These formidable practical and ethical difficulties may have contributed to a dearth of high quality paediatric research. A group of community paediatricians investigated the research evidence that existed to support their clinical decisions and found that only 40% of decisions were supported by good quality clinical studies.2 A recent review of randomised controlled trials published in one paediatric journal over 15 years identified only 249, of which 43% were funded by pharmaceutical companies.3 The numbers of children recruited to these trials were generally small (about half recruited less than 40), indicating that they were unlikely to detect small or moderate differences in treatment effects. Many of the trials funded by the pharmaceutical industry will have been conducted for drug regulatory purposes, and in this area the gaps in research have been comprehensively demonstrated. A study of five centres across Europe showed that almost half of all drug prescriptions for children were either for off label indications or used unlicensed drugs.4 In an era in which clinicians are being exhorted to practise evidence based care, paediatricians are faced with a lack of evidence, which is at best confusing and at worst dangerous.

    The low priority given to children's health at a policy level has been contrasted with the importance that the public attaches to it.5 In Britain public concern about issues related to children's health has led to the inquiries into events in Bristol, North Staffordshire, and Alder Hey. Among the issues highlighted has been the need for research to be conducted with greater transparency and to the highest ethical and scientific standards, with fully informed consent. No one would argue with these principles, but the very public scrutiny that children's research has undergone and the prevailing double standards on informed consent to treatment within and outside therapeutic research6 are likely to deter research rather than encourage it. As one paediatrician has noted, “I need permission to give a drug to half of my patients, but not to give it to them all.”7 It would hardly be surprising if researchers and funding bodies were to consider research with children too fraught with problems and hence diverted their energies and resources into other areas. This would add to the already depressed state of paediatric clinical research.8

    So how can we move forward in the interests of children? It is important to note that, rather than opposing research with children, the public has recognised that the information available from paediatric research is limited and needs to be strengthened. There is now a strong emphasis on the need for consumer involvement in clinical research, with research organisations throughout the world working with patient groups to help set their research agendas.9 Within the UK, guidance on consumer involvement in research in the NHS supports several models, including consultation, collaboration, and consumer controlled research.10 The guidance relates mainly to health services research and does not specifically mention involvement of consumers in children's research. As consumers, children are usually represented by their parents, but children may be able to express their own opinions, if these are sought, using appropriate methods to elicit them.11 Collaboration with parents and, when possible, children in the design, approval, and conduct of all types of clinical research will help ensure that it is relevant and conducted to the rigorous standards that the public expects.

    Nevertheless, there are real anxieties that research with children will become increasingly difficult to perform.12 Too often, clinical decisions about the care of children have been based on research conducted in adults, or on no research at all. Acquiescence in this unsatisfactory situation is untenable. We need to develop and strengthen research with children by enlisting the active collaboration of parents and children themselves. It will only be by partnership with the individuals whom we wish to benefit from this process that research with children can achieve its aims.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract