Exploring the Impact of Data Availability on Time-Course Model-Based Network Meta-Analysis
Author(s)
Becky Hooper, MSc, Sarah Walsh, MSc, Jenna Ellis, MSc.
EVERSANA, Burlington, ON, Canada.
EVERSANA, Burlington, ON, Canada.
OBJECTIVES: Network meta-analysis (NMA) synthesizes evidence from clinical trials by combining direct and indirect comparisons across a connected network. However, standard NMA focuses on a single timepoint per trial, even as studies increasingly report outcomes at multiple timepoints. Time-course model-based NMA (MBNMA) has been proposed to enhance standard NMA by incorporating longitudinal data, enabling the use of all available evidence. This approach can help mitigate the heterogeneity and inconsistency that may arise when comparing trials which assess outcomes at different timepoints and may provide deeper insights into pharmacological benefits, such as time to onset of action or prolonged treatment effects. The objective of this work is to develop a more robust understanding of the applicability and potential challenges of time-course MBNMA by exploring the impact of limited data availability.
METHODS: Using simulated data, time-course MBNMA was applied to describe the temporal relationship between treatment and a binary response outcome. Scenarios involving common data availability issues were analyzed for their impact on outcome measures.
RESULTS: Varying levels of data availability influenced the outcomes observed across scenarios. Differences in response patterns, such as the timing and magnitude of treatment effects, were observed which illustrate the potential implications of data availability on analysis results.
CONCLUSIONS: Time-course MBNMA provides a framework for incorporating multiple timepoints, facilitating a more nuanced evaluation of treatment effects within a network. The analysis highlights how data availability may affect the utility and interpretability of this method, offering insights into its potential strengths and challenges in addressing longitudinal treatment effects.
METHODS: Using simulated data, time-course MBNMA was applied to describe the temporal relationship between treatment and a binary response outcome. Scenarios involving common data availability issues were analyzed for their impact on outcome measures.
RESULTS: Varying levels of data availability influenced the outcomes observed across scenarios. Differences in response patterns, such as the timing and magnitude of treatment effects, were observed which illustrate the potential implications of data availability on analysis results.
CONCLUSIONS: Time-course MBNMA provides a framework for incorporating multiple timepoints, facilitating a more nuanced evaluation of treatment effects within a network. The analysis highlights how data availability may affect the utility and interpretability of this method, offering insights into its potential strengths and challenges in addressing longitudinal treatment effects.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
MSR114
Topic
Methodological & Statistical Research
Disease
No Additional Disease & Conditions/Specialized Treatment Areas