#1458 Prevalence of smoking and smoking effect on the CKD-related mortality of French dialysis patients

Abstract Background and Aims Smoking has been described in the literature as a factor associated with kidney failure, an increased risk of cancer in patients with chronic kidney disease (CKD) and lower access to transplantation. However, few studies have investigated its role in the mortality associated with CKD in France. This study aims to investigate the factors associated with smoking prevalence and its role on the CKD-related mortality among French dialysis patients. Method Thanks to the French Epidemiologic and Information Network in Nephrology (REIN) 29,281 patients aged between 18 and 85 years old who started dialysis between 2017 and 2019 were included in the study and followed until the end of 2022. Patient's characteristics and neighbourhood social deprivation level (the European deprivation index (EDI)) were extracted. A joint modelling imputation was performed due the presence of missing data in smoking and EDI. A multinomial logistic regression with a random effect of department was used to assess the factors associated with smoking prevalence. Survival analysis was conducted, using the excess hazard modelling framework with a random effect to account for the presence of territorial inequalities between French departments in terms of mortality. Results Among the 29,281 included patients, 11,425 (39.02 %) died during the follow-up time. Regarding the smoking status, 5,375 patients had missing data (18.3 %). Of the remaining 23,906 complete data, 3,619 (15.1 %) patients were current smokers, 7,966 (33.3 %) were former smokers and 12,321 (51.5 %) were non-smokers. The odds of being a former smoker and current smoker compared to non-smoker were lower for female, respectively, (OR: 0.34 [0.32, 0.37]) and (OR: 0.2 [0.18, 0.21]). The odds of being current smoker was not associated with social deprivation and decreased with age. The CKD-related mortality increased for former smokers (EHR 1.24 [1.15, 1.32]) and for current smokers (EHR 1.51 [1.38, 1.64]). The standard error of random effects was significant, indicating heterogeneity in CKD-related mortality between departments. Conclusion To the best of our knowledge, this study is the first to quantify the smoking effect on the CKD-related mortality of dialyzed patients. Further analysis will be conducted to distinguish heterogeneity in excess mortality from expected mortality aiming to better understand territorial inequalities in the mortality of dialyzed patients.