Specifying the ‘PICO’ for each synthesis: current practice and development of the InSynQ (Intervention Synthesis Questions) checklist and guide
2Centre for Health Communication & Participation, La Trobe University, Australia, Australia
3EPPI-Centre, UCL Social Research Institute, University College London, UK
4Cochrane Australia, School of Public Health and Preventive Medicine, Monash University, Australia, Australia
Background: Methods guidance prompts systematic review authors to define their questions and specify criteria for including studies in the review. Less emphasis has been given to specifying criteria for each synthesis (the 'PICO [population, intervention, comparator and outcome] for each synthesis’), although choices about the structure of the synthesis have important implications for review findings, particularly in reviews addressing multiple research questions.
Methods: We examined the extent to which PICO criteria were specified for syntheses (comparisons) in 100 randomly sampled systematic reviews published in 2018 from the Health Evidence and Health Systems Evidence databases. We used evidence from this study to draft the InSynQ (Intervention Synthesis Questions) checklist and guide for development and reporting of synthesis questions in systematic reviews of interventions. We conducted two online webinars and a public evaluation survey to invite feedback and refine the final version.
Results: Authors commonly group studies for synthesis based on population, interventions, outcomes and methodological characteristics, but the use of these groups is often incompletely reported. For example, of 41 systematic reviews that identified or used intervention groups as the basis for comparisons, 28 (68%) mentioned the groups in the background or methods; 12 (29%) defined the groups in enough detail to replicate allocation of studies to groups; 6 (15%) provided a rationale; and 24 (59%) stated that the groups would be used to define comparisons. Having identified aspects of reporting that are often incomplete, we developed the draft InSynQ checklist and guide (http://InSynQ.info). Revisions made in response to feedback led to the final 11-item checklist with refined wording and an additional item on consumer and stakeholder involvement.
Conclusions: Complete reporting of synthesis questions and the PICO for each synthesis helps ensure that the questions addressed in a review are clear and supports transparent decisions about the eligibility of studies for each synthesis. The InSynQ checklist and guide is designed to help authors develop and report synthesis questions completely. Further feedback is invited to inform future versions.
Patient, public and/or consumer involvement: Evaluation of the draft checklist was open to the public. These findings promote clear communication of research questions of importance to consumers and transparent reporting of reviews.