Figure 3.

The association between initiation of sodium-glucose cotransporter 2 inhibitors, compared with dipeptidyl peptidase 4 inhibitors, and the change in eGFR over time. In the intention-to-treat analysis, follow-up continued until the occurrence of an effectiveness outcome (for the specific end point), end of data availability, death, or September 2021. The as-treated analysis's follow-up criteria were those used by the intention-to-treat analysis, added by censoring at study drug discontinuation or the initiation of the comparator drug. In the as-treated analysis, follow-up was extended by a grace period of 90 days after the last drug dispensation. A subgroup of patients without evidence of cardiovascular or kidney disease was defined by the lack of the following: evidence of cardiovascular disease, eGFR <60 ml/min per 1.73 m², or urine albumin-to-creatinine ratio ≥30 mg/g. Mixed models for repeated measures were used to describe the change in eGFR over time. We defined time windows as follows: every 3 months in the first year, every 6 months between years 1 and 3, and each year thereafter. At each time window, we considered only the eGFR measurement closest to the end of the period for each patient. We calculated the P value of the change in eGFR between groups at each time point using a mixed-effect model with repeated measures.