Presentation (CTI)

OBJECTIVES: Transitivity is a core assumption underlying network meta-analysis (NMA) that requires the distribution of effect modifiers to be sufficiently similar across treatment comparisons to support valid indirect comparisons. This study aimed to review existing methodological guidance for evaluating transitivity and to develop a structured preliminary checklist synthesising these recommendations into a practical assessment tool.
METHODS: A targeted literature review was conducted to identify published methodological articles, guidelines, and expert commentaries addressing the assessment of transitivity in NMAs. Broad methodological and conceptual keywords (e.g., ‘network meta-analysis’, ‘NMA’, ‘methodological’, ‘guideline’, ‘transitivity’, ‘effect modifier’, ‘comparability’, and ‘consistency’) were used without temporal restrictions to maximise identification of relevant methodological guidance. Key insights were systematically extracted and recorded in Microsoft Excel^®, then reviewed and categorised into five domains. Through iterative internal discussions, these domains were refined and synthesised into a preliminary structured checklist to support the systematic assessment, documentation, and transparent reporting of transitivity assumptions in NMAs.
RESULTS: The preliminary checklist incorporates five key assessment domains: study population, intervention details, outcome measurement, study design, and analytical exploration. These domains inform an overall transitivity judgement to determine whether the assumption is sufficiently supported for valid indirect comparisons within an NMA. The structured format facilitates systematic documentation of study-level characteristics and identification of potential sources of intransitivity that may affect indirect comparisons. The checklist is intended to support proactive evaluation and transparent reporting of transitivity, thereby enhancing the methodological rigour of NMAs.
CONCLUSIONS: Existing methodological guidance acknowledges the critical role of transitivity in NMAs; however, no standardised, comprehensive checklist currently exists for its assessment. This preliminary checklist synthesises recommendations from multiple methodological sources into a structured assessment tool. Future work will focus on refinement through expert consultation, user feedback, and pilot testing to develop a rigorous, transparent, and broadly applicable checklist for evaluating transitivity in NMAs.