Systematic Reviews for the Health Sciences

Screening and selecting studies is a methodical process. To reduce bias at least 2 members of the review team should be involved in the screening of the articles.

Before you begin screening the results, you need to remove all duplicate articles. Many citation managers assist with this process as well as:

The initial pass through the results is scanning the title and abstract. Apply your inclusion and exclusion criteria to these sections. If in doubt, include the article.

Covidence, Rayyan, SR-Accelerator, and Nested Knowledge allow users to populate the tools with inclusion and exclusion criteria. Users can then apply these criteria during the screening process.

The second pass is a full text screening. Again apply your inclusion and exclusion criteria to each article to determine the articles to be included in your review. If necessary, full text copies can be obtained through:

REMEMBER: To reduce bias at least 2 members of the review team should be involved in the screening of the articles.

Create a consistent and reproducible method to analyze the quality of your included studies. While this assessment can be subjective, you will need to make note of how this process was conducted. There are many tools to assist you with determining bias and potential issues in study methodologies.

Remember not all risk of bias tools will work for all study types! You will need to choose a tool based on your topic and the types of studies included. Use the Quality Assessment & Risk of Bias Tool Repository to explore options.

Covidence includes two templates for quality assessment. The first is based on the Cochrane Risk of Bias version 1 tool, the second can be customized to the needs of each review.

Other information on assessing bias and quality:

You will need to track all information of the search process: search results, duplicates, reasons for exclusion, etc. This information can then be recorded in a PRISMA flow diagram: