Same disease, different outcomes: a retrospective cohort study of COVID-19-associated AKI across Brazil's dual-tiered healthcare system.

INTRODUCTION: Acute kidney injury (AKI) is a critical complication of COVID-19, yet disparities in outcomes between public and private healthcare systems remain underexplored. Brazil's two-tiered healthcare system offers a unique setting to evaluate how resource allocation impacts AKI outcomes. This study compares AKI epidemiology in COVID-19 patients treated in two Brazilian hospitals, one publicly governed and the other of privately governed. METHODS: This retrospective cohort study analyzed 2,333 ICU patients with RT-PCR-confirmed COVID-19 (public 1,041; private 1,292, March 2020-April 2022). AKI was defined per KDIGO criteria and recovery was classified using ADQI guidelines. Multivariate logistic regression and Cox models were used to assess predictors of AKI incidence and mortality. To account for competing risk of in-hospital death, a Fine-Gray model was used to evaluate renal recovery. RESULTS: AKI incidence was high in both settings (private, 80.4%; public, 78.8%). Despite comparable baseline characteristics, public hospital patients had significantly higher mortality rates (46.7% vs. 31.3%, p < 0.001). After adjusting for confounders, public hospital admission remained an independent predictor of AKI incidence (OR 1.279, 95% CI 1.012 - 1.620) and mortality (HR 1.675, 95% CI 1.435-1.956). While crude recovery rates appeared higher in public hospitals, competing risk analysis revealed significantly lower recovery probability (SHR 0.650, 95% CI 0.554-0.762). DISCUSSION: Despite comparable AKI incidence, public hospital patients had higher mortality and lower renal recovery, likely reflecting resource disparities. These findings underscore the need to address cost-effectiveness and equity between the public and private sectors healthcare systems.