CLUSTER searching approach to inform evidence syntheses: a methodological review

Tsang, Anthony ORCID: https://orcid.org/0000-0002-2735-7515 and Maden, Michelle (2021) CLUSTER searching approach to inform evidence syntheses: a methodological review. Research Synthesis Methods, 12 (5). pp. 576-589. ISSN 1759-2879

Preview

Published Version Available under License Creative Commons Attribution Non-commercial. Download (1MB) | Preview

Abstract

Background The CLUSTER model of searching was proposed as a systematic method of searching for studies for reviews of complex interventions. The method has not been evaluated before. Aim This methodological review identified and evaluated a sample of evidence syntheses that have used CLUSTER. Methods A forward citation search on the seed CLUSTER publication was conducted on Web of Science Core Collection using six journal citation indexes and Google Scholar in December 2020. Reviews which used the CLUSTER method were eligible for inclusion. A narrative synthesis was used to describe the types of evidence syntheses that used CLUSTER searching, the extent to which the CLUSTER approach has been operationalised within evidence syntheses and whether the value, benefits and limitations of CLUSTER were assessed by the reviewers. Findings A total of sixteen reviews were identified and eligible for synthesis. Six different review types that used CLUSTER were identified with realist reviews being the most prominent. The evaluation of complex interventions was the most common review topic area. The use of CLUSTER varied among reviews with the retrieval of sibling studies being the most common reason. ‘Citations’ and ‘Lead authors’ were the most followed elements of CLUSTER. Conclusions Evidence suggests that CLUSTER has been adopted for use in reviews of complex interventions. Its usage varied among the included reviews. It is imperative that future reviewers diligently report the elements and steps of CLUSTER that were utilised in order to provide a reproducible and transparent search strategy that can be reported with similar transparency to bibliographic database searches.