Cochrane Methods

Methodology Register

Background: As the number of systematic reviews increases, investment in empirical updating methodology research is becoming imperative. Updating systematic reviews is multi-factorial yet no model exists as to how to move the process forward. Objectives: To present an overview of a proposed provisional model of 'when' to update systematic reviews as put forth by the international Expert Working Group on Updating Systematic Reviews. Methods: An international meeting funded by the Canadian Agency for Drugs and Technologies in Health (CADTH) on issues pertaining to updating systematic reviews was convened in Ottawa, March 2006. Small group and plenary sessions resulted in a preliminary framework. Results: The model consists of six elements. Context acknowledges the impact of the situation or setting in which decisions are made. Defining 'update' helps frame methodological research clarifying actions along the update continuum and includes discriminating key updating dimensions (e.g. new data, outcomes, interventions etc.). Surveillance / monitoring entails searches and expert opinion and research on the optimal approach to each. Triggers may include new evidence (exhaustive evidence or the presence of key markers such as the rate of new studies, an increase of subjects or number of harmful events), new methods, or may be time-based. Triggers lead to a decision regarding Action (to undertake a full or partial update, leave the review as it is, or withdraw it). Thus we put forward that circumstances and definitions influence monitoring. In turn monitoring potentiates triggers that prompt an assessment and an eventual updating decision. Research is the final model component that makes clear what data and techniques we need to inform the updating process more effectively. Conclusions: This is a first step to explain the process by which people make choices across a wide set of domains related to updating, and provides a framework for empirical research concerning updating.