Most results on this site arrive as hazard ratios, which describe a proportion and not a person. This analysis of the FLOW trial publishes numbers needed to treat, which is the same information in a form somebody can act on [1]. The broader kidney evidence is in the kidney disease evidence.
To prevent one primary kidney outcome over three years, you would treat 22 people with established atherosclerotic cardiovascular disease, 13 with heart failure, or 17 at high cardiovascular risk without established disease. The primary outcome is a composite: a fall in kidney function of 50% or more, an eGFR below 15, dialysis, transplantation, or death from kidney or cardiovascular causes.
Thirteen is a striking figure for a chronic disease outcome. It means that in a heart failure clinic, treating a room of thirteen people for three years prevents one of them from reaching dialysis, transplant or death from these causes — and the people in that room are the ones most likely to get there.
That pattern — dramatic-looking subgroup differences dissolving into null interaction tests — is the same one described in the frailty analysis and the HFpEF sex analysis. It keeps recurring because trials are powered for their whole population and journalism is drawn to the slices.
One boundary is worth marking clearly. FLOW enrolled people with type 2 diabetes and chronic kidney disease, and used semaglutide at 1.0 mg — the diabetes dose, not the 2.4 mg weight product. These numbers describe kidney protection in diabetic kidney disease and do not transfer to someone taking a weight-management dose without either condition, nor to people already on dialysis, which is a separate question covered in the dialysis cohort.