Evidence Synthesis Methodologies in the Health Sciences

Chapter 5 of the Cochrane Handbook for Systematic Reviews of Interventions provides a thorough overview of the data extraction step of the evidence synthesis process and should be consulted as you prepare for this process. Covidence also has a helpful guide to data extraction. As with screening and quality assessment, data extraction for each study should be done by at least two team members.

You should plan to extract only data that is relevant to answering the question you've posed. Refer back to your protocol, where you will have included details of your data extraction plan.

Data to be extracted may include:

• Information about the study (author(s), year of publication, title, DOI).
• Demographics (age, sex, ethnicity, diseases / conditions, other characteristics related to the intervention / outcome).
• Methodology (study type, participant recruitment / selection / allocation, level of evidence, study quality).
• Intervention (quantity, dosage, route of administration, format, duration, time frame, setting).
• Outcomes (quantitative and/or qualitative).

If you plan to synthesize data, you will want to collect additional information such as sample sizes, effect sizes, dependent variables, reliability measures, pre-test data, post-test data, follow-up data, and statistical tests used.

Extraction templates and approaches should be determined by the needs of the specific review.  For example, if you are extracting qualitative data, you will want to extract data such as theoretical framework, data collection method, or role of the researcher and their potential bias.

• Examine one or more articles that meet your inclusion criteria and consider what data elements you would want to capture from it.
• Look at evidence syntheses on your topic and see what data they extracted. See if those articles published their extraction form or other tools, possibly in their supplemental materials.
• Train the review team on the extraction categories and what type of data would be expected. Consider developing a manual or guide to establish standards.
• Pilot the extraction / coding form to ensure data extractors are recording similar data. Revise the extraction form if needed.
• Discuss any discrepancies in coding throughout the process.
• Document everything, including any changes to the process or extraction form and the reasoning behind them.