Background: Research gaps prevent systematic reviewers from making conclusions and, ultimately, limit our ability to make informed health care decisions. Methods for conducting a systematic review are well-defined but there has been no explicit process for the identification of research gaps from systematic reviews.

Objective: To develop and evaluate a framework for the identification and characterisation of research gaps from systematic reviews.

Methods: We developed a framework that uses PICOS (Population, Intervention, Comparison, Outcomes, Setting) to describe the gaps and categorizes the reasons for the gaps as: (A) insufficient or imprecise information, (B) biased information, (C) inconsistent or unknown consistency results, and/or (D) not the right information. We evaluated the framework by: (1) applying the framework to existing systematic reviews, and (2) asking Evidence-based Practice Centers (EPCs) to use the framework and provide feedback on usability and usefulness of the framework.

Results: Our application of the framework to 50 systematic reviews identified about 600 unique research gaps. Key issues emerging from this evaluation included the need to clarify instructions for dealing with multiple comparisons (lumping versus splitting) and need for guidance on use of the framework retrospectively. We received evaluation forms from seven EPCs, applying the framework in eight projects. Challenges identified by the EPCs led to revisions in the instructions including guidance for teams to decide a priori whether to limit the use of the framework to questions for which strength of evidence has been assessed, and the level of detail needed for the characterization of the gaps.

Conclusions: Our team developed and evaluated a framework. Future research is needed to evaluate the relative costs and benefits of using this framework, for review authors and for users of the systematic reviews.