Risk of Hospital Readmission among People with CKD: A Population-Based Cohort Study

People with CKD are at high risk of unplanned readmission and death due to multimorbidity and health care complexity. Lower eGFR is strongly associated with 30-day unplanned readmission or death, with congestive heart failure as the most common cause. Targeted, risk-based discharge strategies are needed to reduce readmissions and improve outcomes in CKD populations.

People with CKD have higher risk of repeated hospitalization and mortality than the general population. Given many unplanned readmissions may be avoidable, examining risks and reasons for readmission across CKD categories may inform targeted interventions and safer discharge planning. Thus, we estimated the association between eGFR and both readmission and mortality and identified diagnoses for readmission.

Adults discharged from hospitals in Alberta, Canada, from 2005 to 2021 were included. Preadmission eGFR was categorized into eight groups (in ml/min per 1.73 m 2 ): ≥60 (G1-2), 45-59 (G3a), 30-44 (G3b), 15-29 (G4), <15 not on dialysis (G5ND), on dialysis (G5D), prevalent kidney transplant recipients (G1T-5T), and unknown. The primary outcome was unplanned readmission or death within 30 days of discharge. Multivariable logistic regression, accounting for multiple admissions per participant, was used to estimate unadjusted and adjusted odds ratios (aOR) of the primary outcome, using G1-2 as the reference. The results were stratified by age and sex. The type and frequency of each unplanned readmission was also determined.

The cohort comprised 2,992,810 admissions among 1,249,248 participants (median [interquartile range] admission per participant: 1 [1-3] admission; age: 61 [47-74] years). Among lower eGFR categories, heart failure was the most common readmission diagnosis, followed by AKI. Relative to G1-2, more impaired eGFR groups had significantly higher odds of the primary outcome, with the highest adjusted odds experienced by the G1T-5T group (aOR, 1.47; 95% confidence interval, 1.12 to 1.93) and G5ND group (aOR, 1.40; 95% confidence interval, 1.34 to 1.47). Stratified analyses showed that individuals younger than 65 years and females with reduced eGFR had higher odds compared with counterparts in the G1-2 reference group.

People with severe CKD had the highest risk of hospital readmission, suggesting the need to test interventions in this group to target volume and hemodynamic related readmissions such as postdischarge AKI and heart failure.