Using Directed Acyclic Graphs in Matching-Adjusted Indirect Comparison and Simulated Treatment Comparison: Systematic Review and Methodological Insights

Author(s)

Jen-Yu Amy Chang, MSc, RPh, PhD1, Sarah Ren, MSc, PhD1, Nicholas Latimer, MSc, PhD1, Ruth Wong, BSc, MSc, PhD1, Min-Hua Jen, PhD2, Uwe Siebert, MPH, MSc, ScD, MD3, Kate Ren, PhD1.
1University of Sheffield, Sheffield, United Kingdom, 2Eli Lilly, Uxbridge, United Kingdom, 3Department of Public Health, Health Services Research and Health Technology Assessment, UMIT TIROL - University for Health Sciences and Technology, Hall in Tirol, Austria.
OBJECTIVES: Population-adjusted indirect comparisons (PAICs), such as matching-adjusted indirect comparison (MAIC) and simulated treatment comparison (STC), are increasingly used in health technology assessment (HTA) where head-to-head evidence is unavailable. Directed acyclic graphs (DAGs) may improve the transparency and rigour of these methods by clarifying causal assumptions and informing covariate selection. This review examined the extent of DAG application in MAIC and STC to highlight methodological practices and gaps.
METHODS: We systematically searched PubMed, Embase, and Web of Science through June 2025. Two reviewers independently screened studies. Eligible articles reported DAG use in relation to MAIC or STC. Extracted data included disease areas, outcome types, DAG development processes, and reported benefits or limitations.
RESULTS: Of 103 records screened, 26 full-texts were reviewed and four met inclusion criteria. One early study (2012) presented conceptual DAGs for anchored indirect comparisons involving binary outcomes. Three recent studies (2024) applied DAGs in practice: one used a post-hoc DAG to explore bias in an unanchored oncology MAIC with survival outcomes; two evaluated DAGs in hybrid external control designs combining internal and external placebo arms in rare diseases. Only one study explicitly referenced MAIC; the rest discussed addressing transportability in analyses conceptually similar to MAIC/STC. DAGs were used to identify baseline covariates, visualise bias pathways, and highlight risks of over-adjustment. Two studies further applied DAGs to discuss issues arising from post-baseline confounding, including differences in follow-up schedules and censoring rules, which may persist despite adjustment for baseline prognostic factors affecting trial inclusion. While expert input on covariate selection was commonly cited, only one study described a formal consensus process, with limited detail on how DAGs supported these discussions.
CONCLUSIONS: Despite their potential, DAG use in PAICs remains limited. Existing applications demonstrate their value, but clearer methodological guidance is needed to support consistent and proactive use in HTA.

Conference/Value in Health Info

2025-11, ISPOR Europe 2025, Glasgow, Scotland

Value in Health, Volume 28, Issue S2

Code

MSR214

Topic

Health Technology Assessment, Methodological & Statistical Research

Topic Subcategory

Confounding, Selection Bias Correction, Causal Inference

Disease

No Additional Disease & Conditions/Specialized Treatment Areas, Oncology, Rare & Orphan Diseases

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on ispor.org. Update my browser now

×