“…A trial of 1402 participants with type 1 DM (inTandem3) and a short trial of 1250 people hospitalized with COVID-19 (DARE-19) provided only small numbers of clinical outcomes and so were not included in meta-analyses (Supplemental Methods provide more details/results) [ 17 , 30 ]. Data for the remaining eleven trials were extracted from their primary publications [ [2] , [3] , [4] , [5] , [7] , [8] , [9] , [10] , [11] , [12] , [13] ] and eleven subsidiary peer-reviewed publications [ 14 , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , 31 ]. A total of 77,541 participants were included in meta-analyses: four HF trials randomized 15,684 participants [7] , [8] , [9] , [10] , four type 2 DM high-ASCVD risk trials randomized 42,568 participants [2] , [3] , [4] , [5] , and three CKD trials randomized 19,289 participants [11] , [12] , [13] .…”