Page 40 - ISMCON souvenir 2021

ISMSCON - 2021

Application of Multivariate Bayesian

Arm-Based Network Meta-Analysis of

Pharmacological Interventions for the

Treatment of Acute Bipolar Mania in Adults

2
1
Palash Kumar Malo , Binukumar B. , Muralidharan K. 3
1 Department of Computational Sciences, CHRIST (Deemed to be University), Ghaziabad;
3
2 Department of Biostatistics, Department of Psychiatry,
National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru.

Email: palash16malo@gmail.com

Abstract

Background: In a network meta-analysis (NMA), multiple treatments can be compared simultaneously
by aggregating evidences from direct as well as indirect treatment comparisons in different randomized
controlled trials (RCTs). Conventional NMA are performed using normal approximation approach and can
be applied for arm-level binary outcome data as well. This study aims to estimate the treatment effects
within a Bayesian framework using binomial likelihood for a multivariate NMAmodel.

Materials & Methods: The dataset consists of 57 RCTs comparing the effect of ten pharmacological drugs
and placebo for acute bipolar mania in adults. The binary outcomes of interest were treatment response
and all-cause-dropouts measured at 3 weeks from the baseline. A binomial distribution was adopted for
the number of events and the probability of event occurrence modelled on the logit scale. Jeffrey’s Beta
prior was considered for the heterogeneity and inconsistency standard deviation parameters. Moreover,
Cholesky and spherical decomposition strategies were adopted for the between-study variance-
covariance matrix. In addition, the deviance information criterion (DIC) indices are computed to determine
the model fit. All results pertain to Markov Chain Monte Carlo (MCMC) simulations and all analyses were
carried out in WinBUGSsoftware.
Results: The estimated common heterogeneity SDs were found to be similar and the DIC values did not
provide any evidence for superiority between the two decomposition strategies. The correlation (95%
credible interval) between the outcomes was estimated as -0.31 (-0.71, -0.02) and -0.37 (-0.73, -0.03) for
the Cholesky and spherical decompositions, respectively. Gelman-Rubin convergence statistics found
stable and Monte Carlo errors for all the parameters were around0.005.
Conclusions: Overall, olanzapine, paliperidone and quetiapine were both significantly more effective
and acceptable than placebo when both the study outcomes were considered simultaneously.
Keywords: Bayesian Network Meta-Analysis; Arm-Based Analysis; Multivariate Network Meta-
Analysis.

38 CONFERENCE SOUVENIR