A systematic review (SR) is an overview of existing evidence pertinent to a clearly formulated question, which uses pre-specified and standardised methods to identify and critically appraise relevant research, and to collect, report and analyse data from the studies that are included in the review.

Despite the common use of systematic reviews in areas of human health research, formal systematic reviews have rarely been used in food and feed safety and the existing SR methods may not be directly applicable to food safety issues.

This Guidance was developed to assess the possible modification of available systematic review methods for the systematic evaluation of food and feed safety research, and to evaluate the potential use of SR methodology when doing risk assessments (RA) to support decision making in food and feed safety.

To develop this Guidance, the Assessment Methodology Unit of EFSA recruited a working group of EFSA scientific officers and external members with expertise in food and feed safety, systematic reviews (in health care, ecology, veterinary medicine, zoonotic public health, and environmental management), and in information science. This Guidance has been written for those with expertise in various areas of food and feed safety and risk assessment in support of decision making who may not be familiar with the methodology of systematic reviews.

Systematic review methodology can be implemented to answer well-formulated specific questions generated by the risk assessment process or by other analytical frameworks developed in food and feed safety in a transparent, reproducible, and rigorous evidence-based way. However, several aspects must be considered in order to decide whether specific questions obtained by simplifying broad food or feed safety policy problems are suitable for systematic review. A useful means of determining whether a question is answerable by SR is to identify the structure of the question. If the question structure can be specified in such a way that a particular primary research study design can be envisaged that would answer the question, then it is likely that a systematic review would be appropriate.

If a question is suitable for systematic review, it does not necessarily follow that a systematic review would be worthwhile or practically feasible. Considerations include: prioritisation of risk assessment model parameters for which refinement of the parameter estimates is considered most critical; the quantity and quality of available evidence; the source and potential confidentiality of the evidence; the need for transparency and/or for integrating conflicting results; and the availability of resources for carrying out the review.

This Guidance describes a general method for performing systematic reviews, taking into account issues that may be unique to the field of food and feed safety and risk assessment that should be factored into the review process.

This Guidance represents a first step towards the application of systematic review methodology in food and feed safety, and regular updates are foreseen in light of experience and developments both in food and feed safety and systematic review methodology.

    详情见：http://www.efsa.europa.eu/en/scdocs/scdoc/1637.htm