Thursday, Aug 7: 10:30 AM - 12:20 PM
0132
Invited Paper Session
Music City Center
Room: CC-104C
Access to high-quality, affordable, and appropriate health products is a cornerstone for advancing universal health coverage, addressing health emergencies effectively, and promoting the overall well-being of populations. In the quest to achieve these goals, the evaluation of alternative treatment options plays a pivotal role in comparative effectiveness research (CER) and health technology assessment (HTA).
When direct head-to-head randomized controlled trials (RCTs) are not feasible due to practical, ethical, or logistical constraints, indirect treatment comparisons (ITC) become a critical tool in healthcare decision making. ITC allows to assess and compare the relative effectiveness and safety of different treatment options by synthesizing evidence from various sources. Indirect treatment comparisons (including network meta-analysis and population-adjusted indirect comparisons) leverage both direct evidence from trials where treatments are directly compared and indirect evidence from trials where treatments are compared through a common comparator. This approach often involves the aggregation of study-level summary data and, when available, individual patient-level data. By combining these data sources, using non-standard and rigorous statistical techniques, ITC provides a comprehensive view of the relative benefits and risks associated with different health interventions.
As the field of statistics, data science and artificial intelligence (AI) evolves, innovative methodologies and causal inference methods are emerging to enhance the accuracy and reliability of ITC. In line with the 2025 JSM theme, "Statistics, Data Science, and AI Enriching Society", this session is designed to explore emerging topics and innovative statistical methodologies that are shaping the future of ITC studies. Specifically, we will delve into the latest advancements in statistical techniques, data integration methods, and AI-driven approaches that are enhancing the design and analysis of ITC studies. By addressing these cutting-edge developments, we aim to improve the precision and applicability of ITC, ultimately leading to more informed and effective healthcare decisions. Equally importantly, the session is intended to raise awareness among the general statistical community about the challenges and opportunities of this important tool of evidence generation.
World-renowned leaders and scientists from academia and the pharmaceutical industry will present and discuss the state-of-the-science approaches to drive innovation in this area of profound impact. Specifically, Dr. Weili He from Abbvie will discuss some of the challenges associated with the existing approaches, with particular emphasis on innovative approaches in handling missing-data in the application of matching-adjusted indirect comparison. Dr. Joseph Cappelleri from Pfizer, who has published extensively on the topic, will present a network meta-interpolation approach adjusting effect modification in network meta-analysis using subgroup analyses. Thirdly, Dr. Lifeng will present a novel procedure involving nonparametric Bayesian approach to treatment ranking in network meta-analysis. Finally, Dr. Demissie Alemayehu, will provide a critical review of the papers in relation to their importance in comparative effectiveness research and health technology assessment.
Indirect Treatment Comparisons
comparative effectiveness research
health technology assessment
matching-adjusted indirect comparison
network meta-analysis
missing-data
Applied
Yes
Main Sponsor
Section on Statistics in Epidemiology
Co Sponsors
ENAR
International Chinese Statistical Association
Presentations
In the support of Health Teachnology Assessment (HTA) submission, we often need to conduct indirect treatment comparisons (ITC). One common type of ITC is population-adjusted indirect comparisons (PATC), in which the individual patient data (IPD) in one trial and the aggregate data (AgD) in the other trial are used to adjust for between-trial differences in the distribution of variables that influence outcome \cite{phillippo2016nice}. The most popular PATC method is the Matching-Adjusted Indirect Comparison (MAIC). However, the literature is lacking on how to handle missing data in either the outcome variable or the variables that influence the outcome in the application of MAIC. In this talk, we propose some methods for handling missing data in either the outcome variable or the variables that influence the outcome when applying MAIC.
Keywords
Matching-adjusted indirect comparison (MAIC)
Aggregate data (AgD)
Patient-level data (IPD)
HTA
Individual patient data
Effect modification (EM) may cause bias in network meta-analysis (NMA). Existing population adjustment NMA methods use individual patient data to adjust for EM but disregard available subgroup information from aggregated data in the evidence network. Additionally, these methods often rely on the shared effect modification (SEM) assumption. In this presentation, we propose Network Meta-Interpolation (NMI): a method using subgroup analyses to adjust for EM that does not assume SEM. NMI balances effect modifiers across studies by turning treatment effect (TE) estimates at the subgroup and study-level into TE and standard errors at EM values common to all studies. In an extensive simulation study, we simulate two evidence networks consisting of four treatments, and assess the impact of departure from the SEM assumption, variable EM correlation across trials, trial sample size, and network size. NMI was compared to standard NMA, network meta-regression (NMR) and Multilevel NMR (ML-NMR) in terms of estimation accuracy and credible interval (CrI) coverage. In the base case non-SEM dataset, NMI achieved the highest estimation accuracy with root mean squared error (RMSE) of 0.228, followed by standard NMA (0.241), ML-NMR (0.447), and NMR (0.541). In the SEM dataset, NMI was again the most accurate method with RMSE of 0.222, followed by ML-NMR (0.255). CrI coverage followed a similar pattern. NMI's dominance in terms of estimation accuracy and CrI coverage appeared to be consistent across all scenarios. NMI represents an effective option for NMA in the presence of study imbalance and available subgroup data.
Keywords
network meta-analysis, subgroup analyses, research synthesis, population adjustment methods, effect modification, network meta-interpolation
Network meta-analysis is a powerful tool to synthesize evidence from independent studies and compare multiple treatments simultaneously. A critical task of performing a network meta-analysis is to offer ranks of all available treatment options for a specific disease outcome. Frequently, the estimated treatment rankings are accompanied by a large amount of uncertainty, suffer from multiplicity issues, and rarely permit possible ties of treatments with similar performance. These issues make interpreting rankings problematic as they are often treated as absolute metrics. To address these shortcomings, we formulate a ranking strategy that adapts to scenarios with high-order uncertainty by producing more conservative results. This improves the interpretability while simultaneously accounting for multiple comparisons. To admit ties between treatment effects in cases where differences between treatment effects are negligible, we also develop a Bayesian nonparametric approach for network meta-analysis. The approach capitalizes on the induced clustering mechanism of Bayesian nonparametric methods, producing a positive probability that two treatment effects are equal. We demonstrate the utility of the procedure through numerical experiments and a network meta-analysis designed to study antidepressant treatments.
Keywords
multiple comparisons
network meta-analysis
non-parametric Bayesian approach
stick-breaking process
spike and slab
treatment ranking