1 of 22

ISEE LAB Instructional Modules: Covidence (Data Extraction Phase)

PRESENTED BY Joanna Samhouri

2 of 22

Covidence: Data Extraction

  • This module introduces data extraction in Covidence
  • You will learn how to create extraction templates and extract data consistently
  • Accurate data extraction is critical for high-quality evidence synthesis

2

3 of 22

What Is Data Extraction?

  • Data extraction is the process of systematically collecting key information from included studies
  • It occurs after full-text screening is complete
  • Extracted data are used for synthesis, analysis, and reporting

3

4 of 22

Why Is Data Extraction Important?

  • Ensures consistency across reviewers
  • Reduces errors and bias
  • Supports transparent and reproducible reviews
  • Directly informs results and conclusions

4

5 of 22

When Does Data Extraction Happen?

  • After full-text screening is finalized
  • Once inclusion decisions are confirmed
  • Before synthesis or analysis begins

5

6 of 22

Getting Started in Covidence

  • Log in to Covidence and open your review
  • Navigate to the Data Extraction tab
  • Confirm that all included studies are ready for extraction

6

7 of 22

Creating a Data Extraction Template

  • Templates define what information will be collected
  • Templates should align with the research question and protocol
  • Templates are created before extraction begins

7

8 of 22

Common Data Fields to Include

  • Study identification (author, year, journal)
  • Study design and setting
  • Population characteristics
  • Intervention or exposure details
  • Outcomes and key findings

8

9 of 22

Implementation Research–Specific Fields

  • Implementation outcomes (e.g., acceptability, feasibility)
  • Contextual factors
  • Implementation strategies
  • Barriers and facilitators

9

10 of 22

How to Build the Template in Covidence

  • Use the review settings to edit extraction fields
  • Choose appropriate field types (text, numeric, dropdown)
  • Use clear and consistent field labels
  • Pilot the template before full extraction

10

11 of 22

Piloting the Extraction Template

  • Test the template on 2–3 studies
  • Check for missing or unclear fields
  • Refine the template as needed
  • Finalize before full data extraction

11

12 of 22

Extracting Data in Covidence

  • Open an included study
  • Enter information into each extraction field
  • Use the full text to verify details
  • Save progress regularly

12

13 of 22

Best Practices for Data Extraction

  • Extract data carefully and consistently
  • Do not infer information that is not reported
  • Flag unclear data for discussion
  • Communicate questions to your section lead

13

14 of 22

Duplicate Data Extraction (Two Reviewers)

  • Some reviews require two independent reviewers to extract data
  • Each reviewer completes extraction separately
  • This approach increases accuracy and reduces bias
  • Common in systematic reviews and high-rigor projects

14

15 of 22

How Duplicate Extraction Works in Covidence

  • Both reviewers extract data for the same study
  • Covidence tracks each reviewer’s entries
  • Differences are flagged automatically

15

16 of 22

Resolving Conflicts: Consensus

  • Conflicts occur when reviewers extract different information
  • Reviewers compare entries side-by-side
  • Discussions focus on what is reported in the article
  • Decisions should be evidence-based, not assumptions

16

17 of 22

Consensus Best Practices

  • Refer back to the full text when resolving conflicts
  • Document agreed-upon decisions clearly
  • Escalate unresolved issues to the section lead if needed

17

18 of 22

Completing Extraction for Export

  • Ensure all required fields are complete
  • Confirm conflicts have been resolved
  • Check for consistency across studies
  • Verify alignment with the extraction template

18

19 of 22

Exporting Extracted Data

  • Navigate to the Export section in Covidence
  • Select the appropriate export format (e.g., Excel)
  • Use exported data for synthesis and analysis

19

20 of 22

Preparing for Analysis

  • Review extracted data for completeness
  • Export data to Excel if needed
  • Ensure fields align with planned synthesis

20

21 of 22

Key Takeaways

  • A well-designed template supports high-quality extraction
  • Consistency and clarity are essential
  • Data extraction directly shapes your review findings

21

22 of 22

THANK YOU FOR WATCHING!��Q&A�

“

22