From: Use of network meta-analysis in systematic reviews: a survey of authors
A clear research question |
Search details—including terms, databases, period, language restrictions |
Inclusion/exclusion criteria including whether study quality was assessed and by what method |
Assumptions The main issue for MTCs to be valid is the consistency assumption. Very often there is not enough information on the included studies to determine whether the consistency assumption is actually reasonable. Assessment that the studies are comparable. e.g. consistent endpoint definitions, differences in patient populations or study settings are not expected to influence the treatment effect. Explicitly state and discuss assumptions of the analysis. |
Detailed statistical analysis plan, including software; Bayesian/Frequentist method; random/fixed effects model; code; co-variates; handling of missing data; pooling of data; choice of priors; assessment of heterogeneity; and assessment of consistency sensitivity analysis of the impact of analyst assumptions is also very important. |