Care Trajectories of Patients Initiating Dialysis As Planned Versus Emergency Procedure

Abstract Background Emergency dialysis is defined as a first dialysis session performed within 24 hours after the nephrologist’s assessment due to a life-threatening risk. Starting dialysis in emergency (ES) is associated with poorer outcomes compared to a planned start (PS). The aim of this study was to compare the pre-dialysis care trajectory between ES and PS patients. Methods Patients ≥ 18 years old who started dialysis in 2015 in France were included. Data from REIN were linked with the French national health database (SNDS). Individual characteristics at dialysis initiation, inpatient stays (number, duration and diagnosis), and consultations with general practitioners (GP) and nephrologists during the year before dialysis start were compared between ES and PS patients, as well as with national clinical practice guidelines. Results Among 8964 patients included, 30.6% started dialysis in emergency. These patients had more comorbidities than PS patients (19% of ES patients had 3 or more cardiovascular diseases versus 12.6% of PS patients, p < 0.001). In the year before dialysis start, ES patients had a lower median number of inpatient stays (1 versus 2 for PS patients, p = 0.001) but had a longer mean duration of hospital stay (8.1 days versus 5.8 days for PS patients, p < 0.001). ES patients had less consultations with a nephrologist (2.5 versus 4.5 for PS patients, p < 0.001). However, the number of GP consultations was not different between the two groups (9.7 versus 9.4 for PS patients, p < 0.064). Conclusions Nearly 1 in 3 patients started dialysis in emergency with more comorbidities at baseline than PS patients. Their one year pre-dialysis care trajectory was characterized by less frequent nephrological follow-up and a similar GP follow up, compared with PS patients. These results raise questions regarding access to coordinated care associating GPs and nephrologists. This issue of coordination will be investigated using qualitative methods.