A rapid, fit-for-purpose QES for a WHO guideline: processes and challenges
Anke Rohwer, Centre for Evidence-based Health Care, Stellenbosch University
Celeste Naude, Centre for Evidence-based Health Care, Division of Biostatistics and Epidemiology, Department of Global Health, Faculty of Medicine and Health Sciences, Stellenbosch University
Background: The evidence-to-decision (EtD) criteria for World Health Organization (WHO) guideline development requires the explicit consideration of acceptability, feasibility and equity implications of interventions. We were commissioned by WHO to produce a qualitative evidence synthesis (QES) on these phenomena, for a broad range of interventions, within a tight timeline.
Objectives: To describe the processes and challenges to produce a rapid QES for a WHO guideline.
Methods: Our approach was informed by a rapid Cochrane QES and the need for a ‘fit-for-purpose’ product. To ensure completion in a rapid but rigorous manner, authors continuously discussed and reflected on their approach and met with the WHO team regularly. Eligible interventions mirrored those in the systematic review of effectiveness, commissioned for the same guideline. We limited perspectives to those of critical stakeholders; searched only MEDLINE (PubMed), supplementing this with targeted searching for ‘siblings’ of relevant trials; and included a ‘pre-screening’ step before full-text screening. Data extraction, analysis and synthesis was done iteratively using best-fit framework synthesis.
Results: The broad range of interventions resulted in the inclusion of >80 studies. Inclusion of trial siblings enabled us to discern evidence that was ‘directly’ and ‘indirectly’ applicable to interventions included in the related effectiveness review. To produce ‘fit-for-purpose’ evidence, we presented findings per intervention category and for each phenomenon of interest. This approach resulted in the richness of some pertinent cross-cutting perspectives not being captured, and some interventions having very sparse data. The best-fit framework approach was suitable for this accelerated synthesis; though construct-level coding enabled systematic analysis, it was time-intensive. Constant communication and extensive time spent on calibration and synchronous discussions by at least three authors was key to successful QES completion. WHO was positive about the aptness of the QES evidence for the multiple EtD frameworks.
Conclusions: Pragmatic processes and decisions, continuous communication and a dedicated team ensured the timeous completion of the QES. This was not without challenges and methodological limitations, particularly due to the composite question. Decisions about which elements should be prioritised, and which accelerated, in rapid QES are challenging.
Patient, public and/or healthcare consumer involvement: No direct involvement.