Combining randomized and non-randomized evidence in clinical research: a review of methods and applications

摘要: Researchers may have multiple motivations for combining disparate pieces of evidence in a meta-analysis, such as generalizing experimental results or increasing the power to detect an effect that single study is not able detect. However, while main question be simple, structure available answer it complex. As consequence, becomes challenge. In this review, we cover statistical methods been used evidence-synthesis different types with same outcome and similar interventions. For methodological literature retrieval area generalized was performed, publications were identified, assessed, grouped classified. Furthermore real applications these medicine identified described. approaches, 39 clinical could identified. A new classification provided, which takes into account: inferential approach, bias modeling, hierarchical structure, use graphical modeling. We conclude discussion pros cons our approach give some practical advice.