Time to first occurrence of kidney disease progression (KDP) or cardiovascular death is reported as incidence rate of first occurrence of KDP or adjudicated cardiovascular death.

Incidence rate= (Number of patients who experienced the event of first occurrence of KDP or cardiovascular death)*100/(patient years at risk (pt-yrs at risk). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Kidney disease progression was defined as:

• end stage kidney disease (defined as the initiation of maintenance dialysis or receipt of a kidney transplant) OR
• a sustained decline in estimated glomerular filtration rate (eGFR) to <10 mL/min/1.73m^2 OR
• renal death OR
• a sustained decline of ≥40% in eGFR from randomisation.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first hospitalization for heart failure ('as adjudicated') or cardiovascular death ('as adjudicated') is reported as incidence rate of first hospitalization for heart failure or cardiovascular death.

Incidence rate= (Number of patients who experienced the event of first hospitalization for heart failure or cardiovascular death) *100/(patient years at risk (pt-yrs at risk)).

pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to occurrences of all-cause hospitalizations is reported as total number of all-cause hospitalizations (first and recurrent combined).

Randomised set (RS) included all randomised participants, whether treated or not.

Time to death from any cause is reported as incidence rate of death from any cause.

Incidence rate of death from any cause = (Number of patients who experienced the event of death from any cause) * 100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first occurrence of kidney disease progression (KDP) is reported as incidence rate of first occurrence of kidney disease progression.

Incidence rate of first occurrence of kidney disease progression= (Number of patients who experienced the event of first occurrence of kidney disease progression) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Kidney disease progression was defined as:

• end stage kidney disease (defined as the initiation of maintenance dialysis or receipt of a kidney transplant) OR
• a sustained decline in estimated glomerular filtration rate (eGFR) to <10 mL/min/1.73m^2 OR
• renal death OR
• a sustained decline of ≥40% in eGFR from randomisation).

Randomised set (RS) included all randomised participants, whether treated or not.

Time to cardiovascular death ('as adjudicated') is reported as incidence rate of cardiovascular death.

Incidence rate of cardiovascular death= (Number of patients who experienced the event of cardiovascular death) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first occurrence of cardiovascular death ('as adjudicated') or end stage kidney disease is reported as incidence rate of first occurrence of cardiovascular death or end stage kidney disease (ESKD).

Incidence rate of first occurrence cardiovascular death or end stage kidney disease (ESKD)= (Number of patients who experienced the event of first occurrence of cardiovascular death or end stage kidney disease (ESKD)) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

ESKD was defined as the initiation of maintenance dialysis or receipt of a kidney transplant.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first occurrence of kidney disease progression (KDP) or cardiovascular death is reported as incidence of progression of kidney disease or death from cardiovascular causes in the interventional part of the trial and in the post-trial follow-up (non-interventional part).

Incidence rate= (Number of patients who experienced the event of first occurrence of KDP or cardiovascular death)*100/(patient years at risk (pt-yrs at risk). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Kidney disease progression was defined as:

• a sustained decline in eGFR to less than 10 mL/min/1.73m^2 OR
• renal death OR
• sustained decline of more than 40% in eGFR from randomization.

Randomised set (RS): included all randomised participants, whether treated or not.

The time to first occurrence of kidney disease progression in the interventional part of the trial and in the post-trial follow-up (non-interventional part) is reported as the incidence rate of first occurrence of kidney disease progression.

Incidence rate of first occurrence of kidney disease progression= (Number of patients who experienced the event of first occurrence of kidney disease progression) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

Kidney disease progression was defined as:

• end stage kidney disease (defined as the initiation of maintenance dialysis or receipt of a kidney transplant) OR
• a sustained decline in estimated glomerular filtration rate (eGFR) to <10 mL/min/1.73m^2 OR
• renal death OR
• a sustained decline of ≥40% in eGFR from randomisation.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first occurrence of death from any cause or end stage kidney disease (ESKD) in the interventional part of the trial and in the post-trial follow-up (non-interventional part) is reported as incidence rate of first occurrence of death from any cause or ESKD.

Incidence rate of first occurrence of death from any cause or end stage kidney disease (ESKD)= (Number of patients who experienced the event of first occurrence of death any cause or end stage kidney disease (ESKD)) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

ESKD was defined as the initiation of maintenance dialysis or receipt of a kidney transplant.

Randomised set (RS) included all randomised participants, whether treated or not.

Time to first occurrence of end stage kidney disease (ESKD) in the interventional part of the trial and in the post-trial follow-up (non-interventional part) is reported as incidence rate of first occurrence of ESKD.

Incidence rate of first occurrence of end stage kidney disease (ESKD)= (Number of patients who experienced the event of first occurrence of ESKD) *100/(patient years at risk (pt-yrs at risk)). pt-yrs at risk= sum of time at risk [days] over all patients in a treatment group / 365.25.

ESKD was defined as the initiation of maintenance dialysis or receipt of a kidney transplant.

Randomised set (RS) included all randomised participants, whether treated or not.

Mean absolute fluid overload averaged over time in the body composition measurement sub-study. Fluid overload or overhydration was measured using bioimpedance spectroscopy which derives the amount of water in liters (L) in the adipose tissue and lean mass tissues and computed as the difference between expected (based upon weight and body composition) versus measured extracellular water volume, with positive values representing excess fluid.

A mixed model of repeated measures (MMRM) with terms for baseline, age, sex, screening diabetes status, local screening eGFR, local screening UACR, treatment, treatment-by-time interaction and baseline-by-time interaction was used for the analysis. The weighted mean of the values at 2 and 18 months.

Body composition measurement sub-study: all randomized patients who signed the informed consent form to participate in the body composition measurement sub-study, whether treated or not, with at least one valid BCM measurement. Patients with non-valid follow-up measurements were excluded from the analysis.

Kidney cortical T1 mapping using the modified Look-Locker inversion recovery (MOLLI) measured by magnetic resonance imaging (MRI) in the placebo and empagliflozin groups.

A linear regression with terms for age, sex, screening diabetes status, local screening eGFR, local screening UACR was used in the analysis.

Magnetic resonance imaging sub-study: all randomized patients who signed the informed consent form to participate in magnetic resonance imaging sub-study, whether treated or not.