RoB 2: Risk of Bias in Randomized Trials
A result-specific risk-of-bias appraisal tool for randomized trials, with judgments tied to the outcome, estimand, and relevant trial design rather than a numeric study-quality score.
On this page
What this is
A result-specific risk-of-bias appraisal tool for randomized trials, with judgments tied to the outcome, estimand, and relevant trial design rather than a numeric study-quality score. RoB 2 appraises bias in a particular result through domains covering randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selection of the reported result. Choose the relevant assignment/adherence effect and the appropriate parallel, cluster, or crossover guidance. The exact official tool and algorithms govern the judgment; the practical domains here do not replace them.
When to use
Use for a randomized result in an evidence synthesis, appraisal, or design critique. A pragmatic or registry-based randomized trial remains randomized and should not automatically be assessed with ROBINS-I because it uses routine data. Record outcome, time, effect of interest, and which analysis result is being assessed.
What it requires and practical domains
The analyst or appraiser should keep the question, population, outcome or model result, source documents, version, and judgments connected. Use the domains below to organize that evidence. They are a condensed implementation crosswalk; consult the linked official source for exact instrument text, permissions, and decision rules. Retain disagreements and missing information rather than converting them into favorable answers.
When NOT to use and limitations
Do not apply RoB 2 to an observational intervention estimate, award a quality percentage by adding domains, infer low bias from a CONSORT checklist, or assume randomization prevents bias from missingness, measurement, and selective reporting. A completed form documents an assessment; it does not create missing study evidence, guarantee future performance, or authorize a clinical, regulatory, or reimbursement conclusion. Applicability remains a separate question from internal validity and reporting completeness.
How it maps to this catalog
Follow the related concepts to implement definitions, analysis, diagnostics, and interpretation. For this entry the companion routes are pragmatic-trial, registry-trial, cluster-randomized, intention-to-treat-analysis-rwe, consort. Preserve the distinction between the guidance/tool and the methods that implement it; source-backed concepts explain those methods, while this resource organizes reporting or appraisal evidence.
Evidence record
For each applicable domain record the exact evidence location, relevant study/model version, judgment and rationale, uncertainty, unresolved issue, responsible reviewer, and review date. Do not invent approvals, validation results, or completed analyses to fill a field. Store the actual operational specification separately from the assessment, so a favorable label cannot silently replace the underlying evidence.
Checklist
- Identify the specific randomized result, outcome, time point, and effect of interest.
- Select the official design-specific RoB 2 version and record its date.
- Evaluate each relevant domain with supporting trial evidence and signaling answers.
- Document the algorithm-guided and final judgments, including justified overrides.
- Use the result-specific appraisal in synthesis without treating it as a numerical quality weight.