4.2. Evidence retrieval, quality assessment and grading of the evidence

Systematic review teams at the London School of Hygiene & Tropical Medicine and the University of Sydney were asked to prepare standard protocols before embarking on the review. These protocols were then evaluated and endorsed by the Guideline Steering Group and the guideline development methodologist. Seven systematic reviews were conducted to inform the development of these guidelines. Where earlier systematic reviews were identified, if these reviews were eligible for inclusion, primary research studies were selected and included in the present reviews (see Online annexes).

The quality of evidence and strength of the recommendations were assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach (74). This approach to developing recommendations rates the certainty of evidence for all the critical outcomes identified in the PICO question as “high”, “moderate”, “low” or “very low”, based on a set of criteria: study design limitations (risk of bias), inconsistency, imprecision, indirectness and publication bias. Further, in GRADE, the direction (whether in favour of or against an intervention) and strength (whether conditional or strong) of the recommendations reflects the panel’s degree of confidence as to whether the desirable effects of the recommendations outweigh the undesirable effects (52). The guideline development process also considered resource use and cost implications of implementing the recommendations, from a public health perspective. In situations in which strong recommendations (for or against) were suggested on the basis of low or very low confidence in the estimate of effect, members of the Guideline Development Group followed the five paradigmatic situations offered by the GRADE methodology to justify such pairings; these paradigmatic situations are described elsewhere (52, 53).

Book navigation