MO486: Comparison of Hospitalization Lengths of Stay Among a National Cohort of Advanced CKD Patients Treated With Conservative Management Vs. Dialysis

Abstract

Abstract BACKGROUND AND AIMS While dialysis has been the prevailing treatment paradigm in advanced CKD patients progressing to ESRD, this treatment approach may lead to a decline in physical function, loss of independence, and greater healthcare utilization among certain subgroups. We sought to compare the impact of dialysis versus conservative dialysis-free management on hospitalization lengths of stay (LOS) in advanced CKD patients. METHOD We examined a national cohort of advanced CKD patients (≥2 eGFRs <25 mL/min/1.73 m2 separated by ≥90 days) treated with conservative management (CM) versus dialysis over 1 January 2007—30 June 2020 from the OptumLabs® Data Warehouse (OLDW), which contains de-identified administrative claims, including medical and pharmacy claims and enrollment records for commercial and Medicare Advantage enrollees, as well as electronic health record data. In primary analyses, patients were categorized according to receipt of CM, defined as those who did not receive dialysis within 2-years of the index eGFR (1st eGFR <25 mL/min/1.73 m2), versus receipt of dialysis. In secondary analyses, we examined finer gradations of the timing of dialysis, defined as late versus early transition (eGFRs <15 versus ≥15 mL/min/1.73 m2 respectively, at the time of dialysis initiation). We compared LOS among patients treated with CM versus dialysis who were hospitalized within 2 years of their index eGFR using linear mixed effects models that separately considered a fixed age of the cohort (65-years old), with varying times of hospitalization from the index eGFR date, as well as a fixed time of hospitalization from the index eGFR date (12 months) with varying age. RESULTS Among 169 479 advanced CKD patients who were hospitalized within 2-years of their index eGFR, there were a total of 620 168 hospitalizations over this time period. In primary analyses that considered a fixed age of the cohort, dialysis patients experienced longer average LOS versus those treated with CM, with differences attenuating over time (∆ of + 1.5-days for hospitalizations 1-month from the index eGFR; Fig. 1A, upper panel). In secondary analyses, differences in average LOS were even greater for those treated with early dialysis versus CM (∆ of + 2-days for hospitalizations 1-month from the index eGFR, Fig. A, lower panel). In primary analyses that considered a fixed time of hospitalization from the index eGFR, dialysis patients >20-years old had longer average LOS versus those treated with CM, with differences increasing with older age (∆ of + 1.25-days for hospitalizations for patients 70-years old; Fig. 1B, upper panel). In secondary analyses, differences in average LOS were even greater for those treated with early dialysis versus CM (∆ of + 1.87-days for patients 70-years old, Fig. 1B, lower panel). CONCLUSION In a nationally representative CKD cohort, compared with dialysis, those treated with CM as an alternative patient-centered treatment strategy had shorter LOS across varying time points and age groups. Further studies are needed to examine the comparative effectiveness of CM versus dialysis transition on CKD outcomes.