Inconsistency identification in Network Meta-Analysis via Stochastic Search Variable Selection

Date & Time
Monday, September 4, 2023, 11:55 AM - 12:05 PM
Location Name
Session Type
Oral presentation
Network meta-analysis
Oral session
Network Meta-analysis
Seitidis G1, Nikolakopoulos S2, Ntzoufras I3, Mavridis D1
1Department of Primary Education, University of Ioannina, Greece
2Department of Biostatistics, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, The Netherlands
3Department of Statistics, Athens University of Economics and Business, Greece

Background: The reliability of the results of a network meta-analysis (NMA) lies in the plausibility of the key assumption of transitivity, which implies that the effect modifiers’ distribution is similar across treatment comparisons. Transitivity is statistically manifested through the consistency assumption, which suggests that direct and indirect evidence are in agreement. Several methods have been suggested to evaluate consistency. A common approach for testing network consistency suggests adding inconsistency factors to the NMA model.
Objectives: To evaluate the consistency assumption of NMA in a Bayesian framework.
Methods: We describe each inconsistency factor with a candidate covariate whose inclusion on the model relies on variable selection techniques. Our proposed method, Stochastic Search Inconsistency Factor Selection (SSIFS), evaluates the consistency assumption both locally and globally by applying the stochastic search variable selection method to determine whether the inconsistency factors should be included in the model. The posterior inclusion probability of each inconsistency factor quantifies how likely a specific comparison is to be inconsistent. We use posterior model odds or the median probability model to decide on the importance of inconsistency factors.
Results: Differences between direct and indirect evidence can be incorporated into the inconsistency detection process. A key point of our proposed approach is the construction of a reasonable “informative” prior concerning network consistency. The prior is based on the elicitation of information-derived historical data from 201 published network meta-analyses. The performance of our proposed method is evaluated in two published network meta-analyses.
Conclusions: SSIFS is a novel Bayesian method that evaluates the consistency assumption both globally and locally. The proposed method is publicly available in an R package called ssifs, developed and maintained by the authors of this work. Patient, public, and/or healthcare consumer involvement: -