Meta-analysis is the quantitative, scientific synthesis of research results. Fixed-effect and random-effect models are two popular statistical models for meta-analysis. The selection of the appropriate model is crucial. In this paper, we introduce some noval views of models and explain key assumptions, hypothesis, and interpretation of each model. We conclude with a discussion of factors to consider in model selection, and provide a recommendation on selection of appropriate statistical models for traditional meta-analysis.
The assumption of fixed-effects model is based on that the true effect of the each trial is same. However, the assumption of random-effects model is based on that the true effect of included trials is normal distributed. The total variance is equal to the sum of within-trial variance and between-trial variance under the random-effects model. There are many estimators of the between-trial variance. The aim of this paper is to give a brief introduction of the estimators of between-trial variance in trial sequential analysis for random-effects model.
Dose-response meta-analysis is being increasingly applied in evidence production and clinical decision. The research method, synthesizing certain dose-specific effects across studies with the same target question by a certain types of weighting schedule to get a mean dose-response effect, is to reflect the dose-response relationship between certain exposure and outcome. Currently, the most popular method for dose-response meta-analysis is based on the classical "two-stage approach", with the advantage that it allows fixed- or random-effect model, according to the amount of heterogeneity in the model. There are two types of random-effect model available for dose-response meta-analysis, that is, the generally model and the coefficient-correlation-adjusted model. In this article, we briefly introduce two models and illustrate how they are applied in Stata software, which is expected to provide theoretical foundation for evidence-based practice.
