Table 2.

 Example of risk of bias table from a Cochrane review¹⁴

Bias	Authors’ judgment	Support for judgment
Random sequence generation (selection bias)	Low risk	Quote: “Randomization was one to one with a block of size 6. The list of randomization was obtained using the SAS procedure plan at the data statistical analysis centre”
Allocation concealment (selection bias)	Unclear risk	The randomisation list was created at the statistical data centre, but further description of allocation is not included
Blinding of participants and researchers (performance bias)	High risk	Open label
Blinding of outcome assessment (detection bias)	High risk	Open label
Incomplete outcome data (attrition bias)	Low risk	Losses to follow-up were disclosed and the analyses were conducted using, firstly, a modified intention to treat analysis in which missing=failures and, secondly, on an observed basis. Although the authors describe an intention to treat analysis, the 139 participants initially randomised were not all included; five were excluded (four withdrew and one had lung cancer diagnosed). This is a reasonable attrition and not expected to affect results. Adequate sample size of 60 per group was achieved
Selective reporting (reporting bias)	Low risk	All prespecified outcomes were reported
Other bias	Unclear risk	No description of the uptake of the therapeutic drug monitoring recommendations by physicians, which could result in performance bias