Kidney function phenotypes among adults undergoing public-sector creatinine testing in South Africa, 2018–2024: a national retrospective repeated cross-sectional study of routinely collected laboratory data

South Africa lacks routine national surveillance of kidney function. We described laboratory-reported estimated glomerular filtration rate (eGFR) phenotypes among adults undergoing creatinine testing in the public health sector. We analysed National Health Laboratory Service records from adults aged ≥ 18 years in all nine provinces during 2018–2024. The primary analysis used laboratory-reported Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) values during stable reporting in 2020 Q2–2024 Q4, selecting the first valid result per linked person-quarter. Numerical matching suggested the 2009 equation without its Black-race multiplier, but laboratory-specific implementation metadata were unavailable. We described GFR categories, standardised temporal estimates for age and sex and geographic estimates for age, sex and year, and assessed agreement with Modification of Diet in Renal Disease (MDRD) classifications using Cohen’s kappa. Sensitivity analyses varied result selection. Among 83 390 060 raw records, 80 289 377 remained after cleaning. The primary analysis included 47 821 563 person-quarters from 24 839 672 linked adults. eGFR < 60 and < 30 mL/min/1.73 m² occurred in 6 066 893 (12.69%) and 1 299 488 (2.72%) person-quarters. Selecting the lowest quarterly value gave 13.22% and 3.03%. Standardised proportions < 30 were highest in Limpopo (3.45%), Northern Cape (3.00%) and Free State (2.94%), and lowest in Western Cape (2.18%). Agreement was 95.68% (κ = 0.8292) at < 60 and 99.81% (κ = 0.9645) at < 30. Standardised quarterly proportions < 30 ranged from 2.37% to 3.07%, while those < 60 increased overall with fluctuation. Reduced eGFR was common among tested adults, with standardised geographic variation. These results describe laboratory-reported phenotypes, not community chronic kidney disease prevalence or validation against measured GFR. Changes in testing and clinical case mix may affect trends; future surveillance requires clinical linkage, complementary biomarkers and measurement metadata. Not applicable.