Systematic Reviews

Performing a meta-analysis is a common way to quantitatively compare the studies in your review. However, differences in study design may make it impossible to perform a meta-analysis. This is OK; not all systematic reviews include a meta-analysis. You should decide whether you plan to include one at the outset of you review and, if so, factor study design into your inclusion-exclusion criteria. Just as the assistance of a librarian may be useful in designing your search protocol, the services of a statistician may be valuable in developing a framework for quantitative analysis. 

If you are not conducting a quantitative analysis of the studies in your review, the alternative is a narrative approach. You will describe and compare the studies included in your review. As in all other aspects of the review process, it is important to minimize bias. Although this process is not as formal as a meta-analysis, there are general guidelines you can use to help your structure your analysis. This framework for narrative analysis is adopted from the Centre for Reviews and Dissemination's book, Systematic Reviews : CRD’s Guidance for Undertaking Reviews in Health Care. 

• Develop a theory of how the intervention works, why and for whom.
• Develop a preliminary synthesis of findings of included studies.
• Explore relationships within and between studies.
• Assess the robustness of the synthesis.