BACKGROUND:The role of chronic kidney disease (CKD) as a risk factor for cognitive impairment independent of their shared antecedents remains controversial. OBJECTIVE:To determine whether kidney damage (indicated by albuminuria) or kidney dysfunction (estimated glomerular filtration rate [eGFR] <60 ml/min/1.73 m2) predict future (12-year) cognitive function independently of their shared risk factors. METHODS:We studied 4,128 individuals from the 1999/00 population-based Australian Diabetes, Obesity, and Lifestyle (AusDiab) Study who returned in 2011/12 for follow-up cognitive function testing. Albuminuria was defined by urinary albumin:creatinine≥3.5 (women) or≥2.5 mg/mmol (men). Kidney dysfunction was indicated by eGFR <60 ml/min/1.73 m2. Cognitive function domains assessed included memory (California Verbal Learning Test [CVLT]) and processing speed (Symbol Digit Modalities Test [SDMT]). RESULTS:Baseline albuminuria and kidney dysfunction were identified in 142 (3.4%) and 39 (0.9%) individuals, respectively, with minimal overlap (n = 7). Those with albuminuria demonstrated concurrently reduced 12-year SDMT (p = 0.084) and CVLT scores (p = 0.005) following adjustment for age, sex, and education. However, only CVLT performance remained worse (p = 0.027) following additional adjustment for myocardial infarction, stroke, and related risk factors (hypertension, diabetes, dyslipidemia, smoking, BMI, physical activity, and alcohol intake). Indeed, these collective covariates were responsible for 47% of the effect of albuminuria on SDMT, but only 21% of its effect on CVLT. Kidney dysfunction was not associated with either SDMT or CVLT performance (p > 0.10). CONCLUSIONS:Albuminuria predicted worse memory function at 12 years follow-up, whereas its effect on processing speed was driven largely by differences in cardiovascular risk. Kidney dysfunction based on eGFR predicted neither cognitive domain.