Easy confidence interval formulas for network meta-analysis and adjustment of confidence intervals for a small number of studies

TL;DR

This paper introduces simple confidence interval formulas for network meta-analysis and proposes Bartlett-type and bootstrap adjustments to improve accuracy, especially with small study numbers, validated through simulations.

Contribution

It develops new Bartlett-type adjustment methods for both maximum likelihood and restricted maximum likelihood estimators in network meta-analysis.

Findings

Adjusted confidence intervals maintain nominal confidence levels.

Likelihood ratio-based intervals perform well without bootstrap.

Adjustments are effective in actual network meta-analysis scenarios.

Abstract

We propose simple formulas of confidence intervals for the Wald statistic, likelihood ratio statistic, and score statistic for a network meta-analysis. In addition, we consider resolutions for concerns that network meta-analyses with a small number of studies cannot hold a nominal confidence level. For a bias adjustment in analyses with a small number of studies, a Bartlett-type adjustment is a well-known method. Many Bartlett-type adjustment-type methods are based on maximum likelihood estimators. However, the network meta-analysis often uses the restricted maximum likelihood estimators that have not been extensively discussed in Bartlett-type adjustment. In this paper, we propose a Bartlett-type adjustment method for the Wald statistic, likelihood ratio statistic, and score statistic when nuisance parameters are estimated by not only the maximum likelihood method but also the restricted maximum likelihood method. In addition, we propose a higher-order adjustment by applying the bootstrap method to the Bartlett-type adjusted statistics. Using a computer simulation, we confirmed that the adjusted confidence intervals maintained a nominal confidence level. In addition, we confirmed that the confidence interval of the likelihood ratio test based on the restricted maximum likelihood method performs well without further bootstrap adjustment. Finally, we demonstrated that confidence intervals were adjusted for actual network meta-analysis.