"A little learning is a dangerous thing": A call for better understanding of the term 'systematic review'

Abstract

Systematic reviews are becoming a widely accepted gold standard in evidence synthesis for evidence-based and -informed policy and practice. Many organisations exist to coordinate the registration, conduct and publication of systematic reviews across a range of disciplines, including medicine, international development, and environmental management and biodiversity conservation. As the term 'systematic review' becomes more widely recognised, however, there is a risk that stakeholders may have only partial understanding of the rigorous methods required to produce a reliable systematic review. Here, we highlight one such example from the field of education and international development, where a World Bank report claimed to 'systematically review' six 'systematic reviews' that found divergent results. We critically appraise the six included reviews and the World Bank report itself using an a priori quality assessment tool. Our analysis shows that none of the six included reviews are classifiable as systematic reviews according to widely accepted criteria. We also find that the World Bank report failed to use true systematic review methods to synthesise the included reviews findings. Our study demonstrates the risks associated with partial understanding of the added value associated with systematic reviews and highlights a need for improved awareness of what systematic reviews are.

Keywords: Evidence; Evidence synthesis; Evidence-based policy; Grey literature; Risk of bias; Science policy.