P1432THE PRE-DIALYSIS CARE TRAJECTORIES OF ESRD PATIENTS: WHAT TYPES AND WHAT FACTORS OF EMERGENCY DIALYSIS START?

Abstract

Abstract Background and Aims According to the REIN registry, in France in 2017, 30% of incident end-stage renal disease (ESRD) patients started dialysis in emergency. Emergency dialysis start (ES) is associated with higher risks of morbidity and mortality as well as a decreased quality of life for patients. The objectives were to describe groups of patients with similar care trajectories (or “types”) and identify the factors associated with ES. Method Data from the French REIN registry linked with the French national health insurance database (SNDS) were used. Patients aged 18 years old and older who started dialysis in 2015 in France were included. The 2 years pre-dialysis care trajectory was defined using 8 key components of ambulatory and hospital CKD care consumption. Multiple Correspondence Analysis (MCA) on these components was performed before a Hierarchical Clustering on Principal Components (HCPC) in order to identify types of care trajectories. Logistic regression models including both patients’ characteristics and the key CKD healthcare consumption were used to identify factors associated with ES. Results Among the 8856 patients included, 2681 (30.3%) started dialysis in emergency. The HCPC identified 6 types of pre-dialysis care trajectories. ES rate ranged from 11% to 62% depending on the type. Patient’s profiles were also different between types of trajectories, including age and comorbidities (diabetes prevalence ranged from 22% to 56%). For example, the 5th and 6th types of pre-dialysis care trajectories with high ES rates were characterized by low and absent CKD healthcare consumption respectively. They were shared by younger and less comorbid patients (26.7% and 25.7% aged < 45 years old, 66.5% and 59.4% without any cardiovascular diseases respectively). Conversely to the absence of nephrologist follow-up, the absence of general practitioner follow-up was not significantly associated with ES (OR=1.83, CI 95% [1.58-2.13] and OR=0.79, CI 95% [0.61-1.04] respectively). Conclusion Exploiting the strength of the combined use of clinical and healthcare consumption data, our study shows the diversity of pre-dialysis care trajectories of CKD patients. While some of these types of trajectories can be explained by patients’ clinical and demographic characteristics, complementary qualitative approaches are required to fully understand ES. Some types, as well as ES associated factors, highlight the need for earlier CKD screening and nephrologist referral.