Table 2.

Potential bias

Criteria for potential bias	Explanation
RCT	The performance of a randomized controlled trial is required.
Suitable allocation to groups	A precise description of the randomization sequence generating procedure is required (such as computer-generated lists).
Allocation to groups suitably concealed	Information on how allocation to groups is blinded is required (for example centrally by telephone or using sealed opaque envelopes).
Assessment of endpoints blinded	Information on who (patient and/or researcher) assessed which endpoint under blinding conditions (without knowing the group to which the patient had been allocated) is required.
Reasons given for any data loss	The requirement is either no data loss or, if data loss is reported, identification of all patients whose data could not be fully evaluated after randomization and the reasons for this (for example patients who dropped out before the beginning of treatment or during follow-up).
Adequate ITT analysis	Evaluation using the number of randomized patients as the size of the population is required