Article type
Abstract
Background: To inform the largest audience, trial authors may pursue several publishing strategies or develop secondary publications, resulting in trials with multiple associated records. Managing these trials within systematic reviews (SRs) presents unique challenges. While the objective of evidence syntheses is to encompass all available evidence, optimizing time and effort to prevent duplication of information is crucial.
Objectives: To describe the process of managing trials with several associated records in the context of a living systematic review and network meta-analysis (LSRNMA); highlighting specific considerations for such records at different stages of the review.
Methods: Through ongoing discussions, members of the core team involved in this LSRNMA reflected on their experiences managing trials with associated records. We identified the path through which associated records travel through the SRNMA process and highlighted key lessons learnt and considerations at each stage. We will share specific examples of how these lessons are applied in practice.
Results: Until February 2024, we have included 891 trials of which 353 (39.6%) have at least one associated record (in addition to the original record). From these, 136/353 (38.5%) have three or more associated records which may include pre-prints, conference abstracts, post-hoc analyses, fragmented publications, interim analyses, or duplicates. When managing studies with associated records, we identified key considerations at the following review stages: (1) Protocol Development - establish explicit and unambiguous eligibility criteria to avoid including associated records that do not address the SR question. (2) Study Selection - implement a record linking stage prior to data extraction to identify whether any included record is associated with another and to what degree (i.e., true duplicate vs. additional data). (3) Data Extraction - develop an approach for consolidating extractions from several associated records into one extraction. (4) Data Analysis - review included data for any sign of duplicated data prior to analysis.
Conclusions: Inclusion of associated records within an SR provides a more comprehensive overview of trial data; however, they necessitate additional considerations to maximize efficiency during the SR process whilst also ensuring the accuracy of SR results.
Objectives: To describe the process of managing trials with several associated records in the context of a living systematic review and network meta-analysis (LSRNMA); highlighting specific considerations for such records at different stages of the review.
Methods: Through ongoing discussions, members of the core team involved in this LSRNMA reflected on their experiences managing trials with associated records. We identified the path through which associated records travel through the SRNMA process and highlighted key lessons learnt and considerations at each stage. We will share specific examples of how these lessons are applied in practice.
Results: Until February 2024, we have included 891 trials of which 353 (39.6%) have at least one associated record (in addition to the original record). From these, 136/353 (38.5%) have three or more associated records which may include pre-prints, conference abstracts, post-hoc analyses, fragmented publications, interim analyses, or duplicates. When managing studies with associated records, we identified key considerations at the following review stages: (1) Protocol Development - establish explicit and unambiguous eligibility criteria to avoid including associated records that do not address the SR question. (2) Study Selection - implement a record linking stage prior to data extraction to identify whether any included record is associated with another and to what degree (i.e., true duplicate vs. additional data). (3) Data Extraction - develop an approach for consolidating extractions from several associated records into one extraction. (4) Data Analysis - review included data for any sign of duplicated data prior to analysis.
Conclusions: Inclusion of associated records within an SR provides a more comprehensive overview of trial data; however, they necessitate additional considerations to maximize efficiency during the SR process whilst also ensuring the accuracy of SR results.