Red, yellow, green: can a traffic light system help systematic reviews?

SIR–The article by Novak et al.1 has systematically graded evidence on all interventions for children with cerebral palsy (CP) using the GRADE2 system paired with a traffic light system that interprets the evidence grading in colour categories of green (do it), yellow (+ve probably do it; −ve probably do not do it), and red (do not do it).3 This article is creating ‘buzz’ (both negative and positive) with some negative comments voiced by health professionals providing red or yellow coded interventions, as well as questions about the validity of using this methodology to make recommendations on clinical practice. Interestingly, I have seen this article used in the first month since its publication on three occasions: (1) by an academic colleague who wished to succinctly summarize the evidence for constraint therapy and used the ‘bubble’ figures at a workshop for clinicians; (2) by a new occupational therapist working with children with CP who wanted to review the evidence for interventions; and (3) by me when asked by a journalist to comment on the use of ‘therasuits’ as an intervention for CP. It is rare to see an article get so much ‘quick’ pick-up and I interpret this as a sign that it fills a gap in our literature. I wish to share my perspectives as both an academician contributing to evidence as well as a clinician providing service for children with CP. I have taken the liberty of borrowing the authors' Evident Alert Traffic Light System and have organized my thoughts in the following categories: Green – go – positive aspects; Yellow – proceed with caution; and Red – stop – negative aspects (fatal flaws) of the article. Green (positive). (1) Provision of a summary of the state of evidence on interventions for children with CP allowing easy access for all audiences. (2) Utilization of the GRADE system which incorporates a clinical context into grading the evidence (e.g. balancing of benefits with risks). (3) A relative comparison of the effectiveness of different interventions. (4) Use of the traffic light system which allows easy interpretability. This is a particular advantage to busy clinicians who wish to have a quick overview in a manner that is easy to remember. (5) Categorization of interventions according to the International Classification and Functioning, Disability and Health framework, which helps readers to understand the primary intent of the intervention (e.g. activity, participation, environment). (6) The use of a helicopter perspective on all interventions which identifies where we need to go to move the field forward. Yellow (caution). (1) Given the complexity of the GRADE system and the ability to provide a grade even though the quality of evidence may be low, further details on the expert panel would be helpful. Was the expert panel multidisciplinary enough to interpret evidence with adequate expertise? (2) Enhanced discussion on the limits to published evidence to further guide readers. For example, greater details on the challenges of doing randomized clinical trials for certain interventions (e.g. surgical), and the potential bias of industry driven research could be provided. (3) For some interventions the authors interpret evidence outside of the CP literature (e.g. anticonvulsants) to make recommendations for CP. The authors need to ensure that this is applied (where applicable) to all interventions that are graded. (4) A helicopter view has the potential to miss details required for clinical context and hence clinical interpretation should be made in a conservative manner. For example, hip surveillance received a green rating but an orthopedic intervention to manage hip subluxation received a yellow rating. One of the criteria for surveillance is the need for established effective interventions. Certainly the expert clinician would accept orthopaedic interventions to maintain the integrity of the hip joint as the current criterion standard of treatment. This example highlights the difference between evidence-based practice where clinicians provide interventions that have an established evidence base, versus my strong preference for evidence-informed practice where clinicians develop treatment recommendations by integrating knowledge of evidence with clinical expert opinion to establish a well-rounded practice base. The DELPHI methodology is useful for this.4 Red (fatal flaws). None identified. Overall, this article represents a step forward for evidence-informed practice, representing an innovative and accessible way of comparing evidence for interventions. This methodology has many ‘green’ and some yellow ‘cautionary’ aspects, but no ‘red’ fatal flaws. Further work on ways to incorporate expert opinion using methodology such as the DELPHI approach may help to strengthen this article's method for making recommendations on clinical service provision. Both the article and the discussion that is ensuing will help to move the field of evidence-informed practice forward.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.