Skip to content

R0002/2026-03-13/C005/SRC01/E01
Research R0002 — Research Standards for AI-Assisted Writing
Run 2026-03-13
Claim C005
Source SRC01
Evidence SRC01-E01
Type Factual

RoB 2 Assesses Five Domains of Bias

URL: Not captured — experimental run

Extract

The RoB 2 tool assesses five domains of bias in randomized controlled trials:

  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
  5. Bias in selection of the reported result

Neither conflict of interest nor funding source is included as a domain. The original RoB tool's "other bias" domain (where COI and funding were commonly assessed) was removed in RoB 2.

Relevance to Hypotheses

Hypothesis Relationship Strength
H1 Supports Strong — confirms five domains and COI absence
H2 Supports Strong — same factual basis as H1
H3 Contradicts Strong — no COI mechanism anywhere in the five domains

Context

This is the definitive primary source for the RoB 2 domain structure. The five domains are clearly enumerated with no ambiguity. The absence of COI/funding is structural — there is no domain, sub-domain, or signaling question that addresses it.

Notes

The Cochrane documentation focuses on the tool's capabilities, not on what it excludes. The COI absence is documented by omission — it is not mentioned because it is not part of the tool.