Skip to Main Content

Systematic Reviews: Selecting Studies

The Basic Process

At least two team members must independently evaluate all the studies to prevent bias.

There are two rounds of review. 

  1. TITLE/ABSTRACTS:  In the first review, members vote on whether to include/exclude each study based solely on the titles and abstracts. There is no need to record the reasons why studies were excluded. After both team members have evaluated the studies, votes are compared and differences are settled in a pre-specified way. 
  2. FULL-TEXT:  In the second round of reviews, members evaluate the remaining studies by the full-text articles. Members must record the reasons for excluding studies based on the pre-determined inclusion/exclusion criteria they chose when planning the review. After both team members have evaluated the studies, votes are compared and differences are settled in a pre-specified way. 

In general, it's possible to review at least 1-2 abstracts a minute (60-120 an hour). 

Usually only a very small percentage of the initial documents retrieved are included in the final systematic review. Common estimates range from 1-3%.

Establish Inclusion/Exclusion Criteria

Criteria will vary depending on your research question.

Recommended Inclusion Criteria: for reviews about treatment, diagnosis, etiology, and more.   

Remove Duplicate Citations

You'll have a lot of duplicate records in your search results because many articles are listed in more than one database. There are a number of tools to help you: 

SR Accelerator Deduplicator (free) 

Deduklick (small fee) 

Software for Selecting Studies

  • Covidence easily sharable among several reviewers and allows for independent title/abstract screening, full-text screening, data extraction, and risk of bias assessment. It is now required by reviewers doing Cochrane systematic reviews. The library has purchased access for MD Anderson.  
  • Rayyan - easily shareable among several reviewers and allows independent title/abstract screening and data extraction. Rayyan also has a free mobile app (iOS/Android). Low cost. 
  • DistillerSR - allows for the management of the entire systematic review process with multiple independent reviewers. Requires a monthly subscription fee.
  • SR-Accelerator - a free software suite with many functions.

Record Selection Process