A guide to prospective meta-analysis

Abstract

In a prospective meta-analysis (PMA), studies are identified and determined to be eligible for inclusion before the results of the studies related to the PMA research question are known PMAs are applicable to high priority research questions where limited previous evidence exists and where new studies are expected to emerge Compared with standard systematic review and meta-analysis protocols, key adaptations should be made to a PMA protocol, including search methods to identify planned and ongoing studies, details of studies that have already been identified for inclusion, core outcomes to be measured by all studies, collaboration management, and publication policy A systematic search for planned and ongoing studies should precede a PMA, including a search of clinical trial registries and medical literature databases, and contacting relevant stakeholders in the specialty PMAs are ideally conducted by a collaboration or consortium, including a central steering and data analysis committee, and representatives from each individual study Usually PMAs collect individual participant data, but PMAs of aggregate data are also possible. PMAs can include interventional or observational studies PMAs can enable harmonised collection of core outcomes, which can be particularly useful for rare but important outcomes, such as adverse side effects Adaptive forms of PRISMA (preferred reporting items for systematic reviews and meta-analyses) and quality assessment approaches such as GRADE (grading of recommendations assessment, development, and evaluation) should be used to report and assess the quality of evidence for a PMA. The development of a standardised set of reporting guidelines and PMA specific evidence rating tools is highly desirable

Understanding, comparing and learning from the four EPOCH early childhood obesity prevention interventions: A multi‐methods study

“…Direct access to investigators and facilitators enabled us to conduct interviews and compare and contrast lessons from the four trials. Prospective meta‐analysis is an emerging collaborative approach, 11 and this study is an important example of how this approach can enable new insights that progress the research agenda and implementation of interventions.…”

“…Through the EPOCH Collaboration, a core set of outcomes was identified to enable a prospective meta‐analysis to be undertaken 11 . On average, the EPOCH interventions were effective in reducing body mass index (BMI) z‐scores at 18 to 24 months of age by 0.12 standard deviations (95% confidence interval 0.02 to 0.22, P = .17), which translates into a 2% reduction in obesity prevalence 12 .…”

Interventions commenced by early infancy to prevent childhood obesity—The EPOCH Collaboration: An individual participant data prospective meta‐analysis of four randomized controlled trials

“…Whilst the intervention designs varied across trials in terms of mode of delivery and intensity, these reflect differences in real‐world settings and practices, and we believe the assessed groups and outcomes were broadly similar enough to warrant synthesis by meta‐analysis (see details in Table ). It is a strength of PMA methodology that broad clinical and public health questions can be addressed using prospectively planned and agreed data synthesis methods, akin to a multicentre RCT design, but can include trials that assess interventions delivered in local, real‐world settings, where characteristics of the included participants, and the way in which the interventions are delivered, can vary substantially between trials …”

Quantifying the advantages of conducting a prospective meta-analysis (PMA): a case study of early childhood obesity prevention

“…(5,6) Prospective meta-analysis (PMA) is an approach that addresses many of these concerns and the potential sources of bias associated with traditional retrospective meta-analysis. (7,8) For a PMA, eligible studies are identi ed and the PMA hypotheses, selection criteria and analysis methods are speci ed before any results of the included studies related to the PMA research question are known. (8) Therefore, hypotheses, selection criteria and outcomes for the PMA cannot be in uenced by the results of the individual studies (since they are speci ed before the results are known), and publication bias and selective outcome reporting are reduced.…”

“…(10) This increases the external validity of PMA, while maintaining the bene ts of attaining a complete combined dataset with key outcomes available for all included studies. (9) All these bene ts of PMA have been outlined by multiple sources, (3,4,(8)(9)(10)(11)(12) yet, to date they have not been analysed quantitatively.…”