Interesting study to compare change in HbA1c, all-cause mortality, micro- and macro-vascular complications, patient-reported symptoms and quality of life and specific adverse outcomes reported for Canagliflozin / SGLT2s including renal impairment, diabetic ketoacidosis, volume depletion, fournier’s gangrene and lower limb amputation relevant to diabetes reported in published articles versus those in CSRs. The authors seem to only be requesting access to the CSRs for purposes of their meta-analyses, in order to compare effect estimates generated from published data and from CSRs. Is that accurate? The authors should clarify if they are requesting access only to the CSRs, or instead to the IPD and CSRs in order to compare: published reports to IPD; published reports to CSRs; and/or IPD to CSRs.
In addition, the authors should clarify how they will interpret their comparisons. For a comparison of published reports to CSRs, for all of the many outcomes they’ve specified, do the numbers need to be an exact match to be considered equivalent, or within a certain % of one another, etc.?