What is the Cochrane RoB 2 tool?

RoB 2 is the revised tool recommended by Cochrane for assessing the risk of bias in randomized parallel-group trials included in systematic reviews. It evaluates five distinct bias domains using structured signalling questions.

What are the five domains of RoB 2?

The five domains are: (1) Bias arising from the randomization process, (2) Bias due to deviations from intended interventions, (3) Bias due to missing outcome data, (4) Bias in measurement of the outcome, and (5) Bias in selection of the reported result.

What answers can you give to RoB 2 signalling questions?

Signalling questions are answered with: Yes (Y), Probably Yes (PY), Probably No (PN), No (N), or No Information (NI). Certain questions also permit Not Applicable (NA).

How is the overall risk of bias determined in RoB 2?

Overall risk of bias is: 'Low risk of bias' if all five domains are judged as low risk; 'Some concerns' if at least one domain has some concerns but no domain is high risk; and 'High risk of bias' if at least one domain is judged as high risk or multiple domains have some concerns that substantially lower confidence in the result.

Do I have to use the official Cochrane Excel macro spreadsheets?

No. While Cochrane publishes macro-enabled Excel spreadsheets (.xlsm), many reviewer teams find them prone to macro errors and difficult to use on modern operating systems. EvidenceFlow includes built-in algorithmic RoB 2 assessments directly in your review workflow.

CRITICAL APPRAISAL GUIDE

Cochrane Risk of Bias 2 (RoB 2) Tool: Step-by-Step Practical Guide

The Cochrane RoB 2 tool is the gold standard for critical appraisal of randomized trials in systematic reviews. This comprehensive guide explains the 5 bias domains, signalling question logic, and how to reach consistent domain and overall judgments without struggling with legacy macro-enabled spreadsheets.

Published October 12, 2026

Perform dual-blind RoB 2 appraisal in EvidenceFlow

Automatic domain scoring, reviewer conflict resolution, and PRISMA traffic light summaries.

Start Free Review →

1. Overview of the 5 RoB 2 Domains

Unlike the original RoB 1 tool (which relied on subjective reviewer impressions), RoB 2 introduces structured decision algorithms anchored by signalling questions across five core domains:

Domain	Focus Area	Key Considerations	Questions
Domain 1	Randomization process	Allocation sequence generation, concealment, baseline imbalances.	1.1, 1.2, 1.3
Domain 2	Deviations from intended interventions	Blinding of participants and carers, adherence, protocol deviations.	2.1 – 2.7
Domain 3	Missing outcome data	Attrition rates, reasons for dropout, sensitivity of findings to missingness.	3.1 – 3.4
Domain 4	Measurement of the outcome	Blinding of outcome assessors, objectivity of diagnostic criteria.	4.1 – 4.5
Domain 5	Selection of the reported result	Pre-registration in clinical trial registries, selective reporting of analysis.	5.1 – 5.3

2. How to Answer Signalling Questions

For each signalling question within a domain, reviewers must choose one of five standardized responses:

Yes (Y): Clear evidence in the trial report or registry protocol that the criterion was met.
Probably Yes (PY): Direct evidence is incomplete, but contextual trial details strongly imply the criterion was met.
Probably No (PN): Contextual details suggest the safeguard was absent or breached.
No (N): Definitive evidence that the safeguard was not used (e.g., explicit unblinded allocation).
No Information (NI): The published paper and supplementary files omit the detail entirely.

3. The Three Judgment Levels

Once signalling questions are answered, the algorithm assigns one of three levels to the domain:

Low Risk of Bias: The study is comparable to a well-conducted randomized trial; bias is unlikely to alter the clinical direction of findings.

Some Concerns: The trial has limitations or incomplete reporting that raise doubts, but lacks fatal methodological flaws.

High Risk of Bias: Significant methodological flaws that substantially diminish confidence in the estimate of effect.

4. Calculating Overall Risk of Bias

The overall judgment is determined according to strict rules:

Low overall risk: The trial is judged at low risk across all five domains.
Some concerns overall: The trial is judged to have some concerns in at least one domain, but no domain is judged at high risk.
High overall risk: The trial is judged at high risk in at least one domain, OR has some concerns across multiple domains that collectively compromise confidence.

5. Modern RoB 2 vs Legacy Macro Spreadsheets

Cochrane's official Excel template relies on VBA macros that frequently fail on macOS, trigger antivirus blocks, and make dual-reviewer reconciliation laborious. EvidenceFlow incorporates the official 2019 Cochrane algorithms natively, allowing independent reviewers to score studies, detect disagreements automatically, and export clean summary tables alongside your data extraction forms.

Start appraising studies in EvidenceFlow

Streamline screening, RoB 2 appraisal, extraction, and meta-analysis in one connected platform.

Create Free Account