Step 4: Study Selection

Your search may produce a large number of potentially relevant articles and these need to be assessed for relevance against the pre-specified criteria set out in your protocol.

If you have results from searches in several different databases, you will need to remove any duplicate publications prior to screening. This can be done using reference software such as EndNote and in the future, through SyRF. The unique results from your database searches can then be uploaded to SyRF to screen for study selection: Launch SyRF screening application.

 

Study selection typically has 2 stages:

Stage 1: Using the study title and abstract, a decision is made to include or exclude based on your pre-specified criteria. It is important to be over-inclusive at this stage to prevent the incorrect removal of relevant papers. Excluded studies should be either clearly not relevant, or clearly fail on one or more of the inclusion criteria.

Stage 2: Studies that have made it through the first stage, or cases where a definite decision cannot be made, are screened to assess if they qualify for inclusion. The full paper should be obtained at this stage to make a more detailed assessment. Full text PDFs can be automatically retrieved with a reference management software such as EndNote. For studies that are excluded, it can be useful to keep a record of why they did not meet the inclusion criteria.

 

Even if there are detailed inclusion and exclusion criteria, the study selection stage can still involve a degree of subjectivity. Therefore, the reliability is increased if all papers are independently assessed by more than one researcher. Disagreements between reviewers can be discussed with reference to the protocol or an additional reviewer can be consulted. Keep an eye out for multiple reporting, where several published studies refer to the same experimental results. If these are not taken into consideration, bias can be introduced into the results of your systematic review.

 

The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram or similar chart can be useful to visualise and report the study selection process. More information can be found on the PRISMA website.

 

References & Resources:

Centre for Reviews and Dissemination University of York, UK

PRISMA website

 

  < <  Move back to Step 3   Move on to Step 5  > >