Article Text

Download PDFPDF
Developing clinical guidelines with scant evidence approaches taken and lessons learnt
  1. Claire Friedemann Smith1,
  2. Rosa Nieto-Hernandez1,
  3. Asmaa Abdelhamid2
  1. 1 Department of Research and Policy, Royal College of Paediatrics and Child Health, London, UK
  2. 2 Norwich Medical School, University of East Anglia, Norwich, UK
  1. Correspondence to Dr Claire Friedemann Smith, Department of Research and Policy, Royal College or Paediatrics and Child Health, London, WC1X 8SH, UK; cfriedemann86{at}

Statistics from

Request Permissions

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.


It is estimated that 400 children suffer from an arterial ischaemic stroke (AIS) or haemorrhagic stroke (HS) in the UK each year with up to 40% dying from it, and approximately 60% of those surviving the initial attack going on to have recurrent strokes.1 Although in recent years the recognition of stroke as a disease that affects children has increased, many still experience delays in diagnosis or misdiagnosis.2 3 The view that stroke is an illness of the middle-to-late adult years may have also hindered research into stroke in childhood. The need for a clinical guideline was clear; it would provide a practical guide that should aid clinical decision-making and improve patient care,4 but when there are gaps in the literature, creating guidelines presents a challenge.

Methods of guideline development

A guideline development group (GDG) was established that consisted of clinical experts from specialties relevant to the topic of the guideline, parent representatives whose children had suffered a stroke and systematic reviewers from the Royal College of Paediatrics and Child Health, following accepted practice.5 The scope of the guideline was drafted, and after stakeholder consultation, specific clinical questions were formed.

Systematic review

Clinical questions were grouped into subtopics and review protocols were written. Searches were carried out in five databases, MEDLINE, Embase, PsycInfo, Cochrane Library and CINAHL, between 1995 and 2016, these were screened and data extracted from relevant studies. Our expectation was that the literature on rehabilitation following stroke in childhood would not be sufficient to answer all of our clinical questions and so our inclusion and exclusion criteria allowed for the inclusion of studies carried out in children with similar deficits, for example, as a result of acquired brain injury. Despite this, we still found that 36% of all our clinical questions had two or fewer studies on which to base recommendations.

The …

View Full Text


  • Contributors CFS and AA were systematic reviewers at the RCPCH (to April 2017 and September 2016, respectively) and were involved in every stage of the development of the guideline. RNH oversaw the development of every stage of the guideline. CFS drafted this manuscript and RNH and AA edited it. All authors approve this manuscript. CFS, RNH and AA have several years’ experience of carrying out systematic reviews for the RCPCH, the Cochrane Collaboration and numerous other UK universities. The authors declare no conflict of interest. All members of the GDG were required to disclose any conflicts of interest and any conflicts declared will be available in Appendix 2a of the guideline which will be launched in May 2017. CFS is the named guarantor of this work.

  • Funding This work was funded by The Stroke Association. The funder had no influence over the methods used or conclusions drawn in the guideline. The researchers carried out the research independently of the funders. This paper discusses work which was funded by the Stroke Association and developed by the Royal College of Paediatrics and Child Health (RCPCH), which was launched in May 2017. The funders played no role in the conception, writing or content of this paper but have given their approval for it to be submitted for publishing.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.