A unified framework for bias assessment of primary analytic research in evidence synthesis

Abstract

The increasing application of research synthesis into policy and practice has generated an increasing interest in methodological quality assessment to judge the trustworthiness of synthesised results. Bias has been demonstrated to impact the magnitude of effect estimates and therefore identifying research that is of lower quality and consequently at risk of being biased is crucial for clinical decision making and the development of healthcare policy. There are numerous issues that impact on methodological quality assessment in practice, including defining what the assessment tools should consist of, how the assessment should be used, and the pitfalls that should be avoided. In this body of research, key concepts surrounding methodological quality assessment are comprehensively evaluated, culminating in a recommended unified framework for future application in health care through a well-defined methodological quality assessment tool for analytic study designs, called the MethodologicAl STandards for Epidemiological Research (MASTER) scale. Show more