Background: A sizeable proportion of patients with chronic kidney disease (CKD) are reported to be frail. Here we examined the safety andefficacy of dapagliflozin in patients with CKD by frailty level.Methods: Adults with CKD, with/without type 2 diabetes, with an estimated glomerular filtration rate (eGFR) of 25–75 mL/min/1.73 m2 , andurinary albumin-to-creatinine ratio 200–5 000 mg/g were randomized to dapagliflozin (10 mg/day) or placebo. The primary endpoint was a com-posite of sustained ≥50% eGFR decline, end-stage kidney disease (ESKD), or death from kidney or cardiovascular (CV) causes.Results: Frailty index (FI), assessed by Rockwood cumulative deficit approach, was calculable in 4 303/4 304 (99.9%) patients: 1 162 (27.0%)in not-to-mildly frail (FI ≤0.210), 1 642 (38.2%) in moderately frail (FI 0.211–0.310), and 1 499 (34.8%) in severely frail categories (FI >0.311).Dapagliflozin reduced the risk of the primary composite endpoint across all FI categories (hazard ratios [95% confidence interval {CI}]: 0.50[0.33–0.76], 0.62 [0.45–0.85], and 0.64 [0.49–-0.83], respectively; p-interaction = 0.67). Results were similar for secondary outcomes includ-ing kidney composite outcome (sustained ≥50% eGFR decline, ESKD or death from kidney cause; p-interaction = 0.44), CV endpoint (heartfailure hospitalization or CV death; p-interaction = 0.63), and all-cause mortality (p-interaction p = .42). Results were consistent when usingFI as a continuous variable. Occurrence of serious adverse events was numerically lower in patients receiving dapagliflozin versus placeboin all FI categories (16.9% vs 20.1%, 26.3% vs 30.7%, and 42.9% vs 47.8%, in not-to-mildly, moderately, and severely frail categories,respectively).Conclusions: The relative benefit of dapagliflozin for all outcomes was consistent across all frailty categories, with no difference in associatedsafety.