Mapping studies for inclusion in a Cochrane Overview of physical activity




Poster session 5


Thursday 27 October 2016 - 10:30 to 11:00


All authors in correct order:

Bennett, A1, Baker P1, Soares J2
1 Queensland University of Technology, Australia
2 Centers for Disease Control and Prevention, USA
Presenting author and contact person

Presenting author:

Philip Baker

Contact person:

Abstract text
Background: The abundance of systematic reviews (SR) in the literature that investigate interventions for increasing physical activity levels makes a challenging process for overview authors’ to select, describe the effects and scope of the SR to synthesize the evidence into a single convenient source that enables public health decision makers to apply evidence-based practices.

Objectives: The purpose of this work was to describe the process to minimise redundancy and overlap in the overview’s summary to facilitate public health decision makers' actions to support evidence-informed decisions.

Methods: We searched the Health registry database to identify 'strong' SR that investigated interventions for increasing physical activity (PA) levels and mapped the included studies contained in eligible SR. For each class, or type of intervention for the outcome (e.g. school-based interventions for PA), we selected the most current from the strongest SR that comprehensively described the intervention and the outcomes. We examined the studies contained in each review to avoid overlap, and succinctly summarised the current body of evidence from the SR. We used the fewest number of SR required to summarise the evidence from each intervention approach.

Results: We identified 80 SR. Forty-three SR were eliminated due to duplication. The mapping process was applied to the remaining 37 eligible SR, which contained 635 studies. The mapping process identified 493 studies (78%) that were unique to only one included review. The reviews eliminated through mapping were generally older, contained fewer relevant studies, or were narrow in focus compared to the retained SR.

Conclusions: Based on this work we conclude that using the process of mapping studies contained within the SR made it possible to identify the breadth of the interventions and outcomes; and the number of times a primary study has been included in high quality SR. In addition, through explicitly mapping the primary studies, the number of SR required was substantially reduced.