Quality Assessment in Meta-analysis

Abstract

Quality assessment of primary studies to evaluate the reliability of study results is an essential and mandatory part of meta-analyses. It refers to the internal validity of a study and is described more precisely as assessing the risk of bias. Potential biases derive from selection of participants, data collection, analysis and selective reporting of study results. Quality assessment tools systematically collect information about study characteristics that may lead to bias in order to estimate the overall risk of bias. There are numerous tools available; they can be classified into checklists, scales and component ratings. Focusing on tools for assessing randomized controlled studies, an overview of covered elements of six selected generic tools is given. The Cochrane Collaboration’s tool is described in more detail because it incorporates some important features. Practical aspects of conducting quality assessments are discussed including the meaning and importance of detailed and precise guidance.