Epidemiology of Community-Acquired Acute Kidney Injury Among US Veterans

Abstract

Community-acquired acute kidney injury (CA-AKI) develops outside of the hospital and is the most common form of AKI globally. National estimates of CA-AKI in the United States are absent due to limited availability of laboratory data. This study leverages national data from the Veterans Health Administration (VA) to estimate incidence and risk factors of CA-AKI. Retrospective cohort study using national VA administrative and laboratory data to assess cumulative CA-AKI incidence. VA primary care patients in 2013-2017 with recorded outpatient serum creatinine (Scr) and no history of chronic kidney disease≥stage5. Sociodemographics, comorbidities, medication use, and health care utilization. Annual incidence of CA-AKI defined as a≥1.5-fold relative increase in Scr on either a subsequent outpatient Scr or inpatient Scr obtained within≤24 hours of admission. We calculated the relative change in Scr within 12 months of an outpatient Scr value. A Cox model was used to estimate the association between CA-AKI and baseline characteristics, accounting for repeated measurements. Of approximately 2.5 million eligible veterans each year, the cumulative incidence of CA-AKI was approximately 2% annually. Only 27% of CA-AKI was detected at hospital admission. In adjusted analyses, high health care utilization, chronic illness, cancer, rural location, female sex, and use of renin-angiotensin aldosterone system inhibitors or diuretics were associated with increased CA-AKI risk (all, HR>1.20). Limited generalizability of results outside a veteran population, lack of a standardized definition for CA-AKI, and possibility of surveillance bias and misclassification. CA-AKI affects 1 of every 50 US veterans annually. With less than a third of CA-AKI observed in the inpatient hospital setting, reliance on inpatient evaluation of AKI suggests significant underrecognition and missed opportunities to prevent and manage the long-term consequences of AKI.