Table 1. A summary of the RCTs bias using the Cochrane assessment tool.

RCT=Randomized Controlled Trial

COCHRANE APPRAISAL	Random Sequence Generation - Selection bias	Allocation of concealment - Selection bias	Blinding of both participants and evaluators - Performance bias	Blinding of assessment during outcome collection - Detection bias	Incomplete outcome data - Attrition bias	Selective reporting - Reporting bias	Other bias / Comments
Wanner et al. 2016 [14] EMPA-REG	LOW RISK	HIGH RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	Small sample size to generalise to the broader population
Neal et al. 2017 [10] CANVAS	LOW RISK	LOW RISK	HIGH RISK	LOW RISK	LOW RISK	LOW RISK	Few clinical events, small proportion of participants to generalise
Wiviott et al. 2018 [7] DECLARE TIMI-58	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	Premature discontinuation of thousands of patients from each study group - statistical data possibly affected
Perkovic et al. 2019 [15] CREDENCE	LOW RISK	LOW RISK	LOW RISK	LOW RISK	HIGH RISK	LOW RISK	Trial stopped early - potential limited power
Phrommintikul et al. 2019 [16]	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	Small sample size
Bonora et al. 2019 [17]	LOW RISK	LOW RISK	LOW RISK	LOW RISK	UNCLEAR	LOW RISK	Small sample size
Petrie et al. 2020 [18] DAPA-HF	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	N/A
Fuchigami et al. 2020 [19]	LOW RISK	LOW RISK	LOW RISK	HIGH RISK	LOW RISK	LOW RISK	Small sample size, poor genetic diversity (all Japanese), short follow up
Packer et al. 2020 [20] EMPEROR-REDUCED	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	LOW RISK	N/A
Cannon et al. 2020 [21] VERTIS-CV	LOW RISK	LOW RISK	LOW RISK	HIGH RISK	LOW RISK	LOW RISK	N/A