Table 4.

Assessment of bias using the cochrane risk of bias tool for randomized controlled trials for Peek et al. [43]

Domain	Source of bias	Support for judgement	Review authors’ judgement
Selection bias	Random sequence generation	“Patients were enrolled from three types of centres: the ECMO centre at Glenfield Hospital, Leicester, which treated all patients who were randomly allocated for consideration to receive ECMO; tertiary intensive care units (conventional treatment centres); and referral hospitals, which sent patients to the conventional treatment centres if they were randomly allocated to receive continued conventional management.”	Low risk
	Allocation concealment	“Patients were randomly allocated by minimisation in a 1:1 ratio to conventional management by intermittent positive-pressure ventilation or high-frequency oscillatory ventilation, or both, or consideration for treatment by ECMO. Minimisation factors were type of centre; age; hours of high-pressure or high FiO2 ventilation; presence of hypoxia or hypercarbia; diagnostic group; and number of organs failed.”	High risk
Performance bias	Blinding of participants and personnel	Blinding of participants and personnel was not described	Unclear risk
	Blinding of outcome assessment	Blinding of outcome assessment was not described	Unclear risk
Attrition bias	Incomplete outcome data	“Consequently, the number of patients with missing data are lower than for other components of EQ-5D, and other follow-up and economic assessments.”	High risk
Reporting bias	Selective reporting	The primary and secondary outcomes are identifiable in the published report	Low risk
Other bias	Other source bias	None were identified	Low risk