FP344SOCIOECONOMIC STATUS AND RISK OF CHRONIC KIDNEY DISEASE: LONGITUDINAL DATA FROM THE PREVEND STUDY

INTRODUCTION AND AIMS: The association between low socioeconomic status (SES) and chronic kidney disease (CKD) is well-known, but its underlying mechanisms are not fully understood. We therefore examined the associations of SES with the longitudinal kidney outcomes, incident CKD and eGFR decline, and we explored potential mediators of these associations. METHODS: Longitudinal data from the community-based PREVEND Study was used. SES (low, medium, high) was measured by educational level and income. Kidney function (eGFR) was assessed at baseline and at four follow-up examinations using the serum creatinine and cystatin-C based CKD-EPI equation. Incident CKD was defined as eGFR<60mL/min/1.73m2 during follow-up in those free of CKD at baseline. In Cox regression and linear mixed effects models, we included potential confounders (age, sex, baseline eGFR). In univariable causal mediation analyses, we explored potential mediators (BMI, smoking, alcohol use, diabetes, hypertension, hypercholesterolemia, cardiovascular disease). RESULTS: Included were 5,953 participants (median age 47 years [IQR 38-58], 52% males), of whom 171 had CKD at baseline. At baseline, those with lower SES had a less favorable renal risk profile. During 11 years of follow-up [IQR 4.4-12], 309 participants experienced incident CKD. Lower education, not income, was associated with higher rates of incident CKD (low vs high education; Hazard Ratio [95%CI]= 1.40 [1.02 to 1.91], ptrend=0.016) and accelerated eGFR decline (low vs high education; β [95%CI]= -0.17 [-0.24 to -0.10] mL/min/1.73m2 per year, ptrend<0.001), after adjustment for confounders. After introducing potential mediators, education was no longer significantly associated with incident CKD. However, the association with eGFR decline remained unchanged. The association between education and incident CKD was significantly mediated by hypertension (average proportion mediated 7% [1% to 38%], p=0.05, conditioned on age, sex, and baseline eGFR) and smoking (15% [4% to 74%], p=0.03). Results of sensitivity analyses with incident CKD defined as de novo eGFR<60mL or urinary albumin excretion ≥30mg/24h were similar, but in addition identified BMI as a potential mediator (27% [13%-110%], p=0.01). Mediation analysis with eGFR decline as outcome yielded no additional insights. CONCLUSIONS: In this general population cohort, lower education, not income, was associated with elevated CKD risk. Our data suggest that the modifiable risk factors, BMI, smoking, and hypertension underlie this association. These findings may provide support for targeted preventive policies.