how to use Cochrane Risk of Bias Tool (RoB 2)

The Cochrane Risk of Bias Tool (RoB 2) is a structured method for assessing potential biases in randomized controlled trials (RCTs). Below is a step-by-step guide on how to use it effectively:

Step 1: Understand the Five Domains of Bias

The RoB 2 evaluates bias across five key domains:

1. Randomization process
2. Deviations from intended interventions (effect of assignment to intervention)
3. Missing outcome data
4. Measurement of the outcome
5. Selection of the reported result

Each domain is assessed separately, and an overall risk of bias judgment is made.

Step 2: Gather Necessary Information

Before assessing bias, ensure you have:
✔ The full-text RCT paper
✔ Study protocol (if available, to check selective reporting)
✔ Trial registration details (e.g., ClinicalTrials.gov)

Step 3: Assess Each Domain

For each domain, answer signaling questions (usually Yes/Probably Yes/No/Probably No/No Information) to determine bias risk.

1. Bias arising from the randomization process

• Was allocation sequence random?
• Was allocation concealed (e.g., sealed envelopes, central randomization)?
• Were baseline differences between groups likely due to chance?
• Judgment: Low/Some concerns/High risk

2. Bias due to deviations from intended interventions

• Were participants and personnel blinded?
• Were there unintended crossovers or contamination?
• Were deviations from the protocol balanced between groups?
• Judgment: Low/Some concerns/High risk

3. Bias due to missing outcome data

• Was dropout balanced between groups?
• Were reasons for missing data explained?
• Was analysis performed using intention-to-treat (ITT) principles?
• Judgment: Low/Some concerns/High risk

4. Bias in measurement of the outcome

• Were outcome assessors blinded?
• Could outcome measurement differ between groups?
• Judgment: Low/Some concerns/High risk

5. Bias in selection of the reported result

• Were all pre-specified outcomes reported?
• Was there evidence of selective reporting (e.g., omitted unfavorable results)?
• Judgment: Low/Some concerns/High risk

Step 4: Determine Overall Risk of Bias

• Low risk: All domains rated as low.
• Some concerns: At least one domain has “some concerns” but no high risk.
• High risk: At least one domain is high risk, or multiple domains have concerns.

Step 5: Document and Justify Your Judgments

• Use the RoB 2 Excel tool (available from Cochrane) to record responses.
• Provide justifications for each judgment (e.g., “Allocation concealment was unclear”).

Example Assessment

Overall Risk of Bias: High (due to selective reporting).

Additional Tips

• Use the official RoB 2 guidance for complex cases.
• For cluster RCTs or crossover trials, use specialized RoB 2 variants.
• Train multiple reviewers to ensure consistency (calculate inter-rater reliability).

Would you like a worked example of an RCT appraisal using RoB 2?