The impact of early ART initiation (versus deferring) on kidney function has not been studied. START was a randomised comparison of immediate versus deferred ART initiation among HIV-positive persons with CD4+ (cells/mm3) counts >500. Serum creatinine and urine dipstick protein were measured at Months 0, 1, 4, 8 and 12, and annually thereafter. The two arms were compared for changes in eGFR (mL/min/1.73 m2, calculated by CKD-EPI equation), over time using longitudinal mixed models. Of 4685 START participants, 4629 (2294 in immediate and 2335 deferred arm) were included. Median baseline CD4+ and eGFR were 651 and 111.5, respectively. ART was initiated in 2271 participants (99.0%) in the immediate and 1127 (48.3%) in the deferred arm, accounting for >94% and >19% of follow-up time, respectively. Overall, 89% started ART using a tenofovir-based regimen. Over 2.1 years median follow-up, mean eGFR was 0.56 (95% CI 0.003-1.11) higher in the immediate versus deferred arm, which was more prominent after adjustment for current tenofovir or bPI use (1.85, 95% CI 1.21-2.50) and in Black participants (30.1% overall) (3.90, 95% CI 2.84-4.97) versus non-Blacks (1.05, 95% CI 0.33-1.77) (P < 0.001 for interaction). Relative risk for proteinuria in the immediate versus deferred arm was 0.74 (95% CI 0.55-1.00) (P = 0.049). In the short-term, immediate ART initiation was associated with a modestly higher eGFR and lower proteinuria risk versus deferring ART (more pronounced in Black participants). Whether this early benefit translates into a lower risk of CKD requires further follow-up.