B-285 Underdiagnosis and Gaps in Guideline-Recommended Testing for Chronic Kidney Disease in an Academic Health System: A Population-Based Race, Ethnicity, Language, and Sex Analysis

Abstract

Chronic kidney disease (CKD) is underdiagnosed, and adherence to National Kidney Foundation (NKF) screening recommendations among high-risk patients remains suboptimal. NKF guidelines recommend annual assessment with both estimated glomerular filtration rate (eGFR) and urine albumin-to-creatinine ratio (uACR) among patients with diabetes or hypertension. We evaluated CKD diagnosis, risk burden, and testing gaps using the NKF CKD framework, stratified by race, ethnicity, language, and sex variables. We conducted a retrospective population-based analysis of 407,894 adults aged 18–85 years with at least one clinical encounter in the Oregon Health & Science University (OHSU) health system between January 1 and December 31, 2025. CKD diagnosis was identified using ICD-10 codes. High-risk status was defined as diabetes and/or hypertension. Guideline-concordant testing was defined as completion of both eGFR and uACR within 12 months. Laboratory-identified CKD was defined as an eGFR <60 mL/min/1.73 m² on at least two measurements separated by =90 days in the absence of a documented CKD diagnosis. Overall, 5.01% (20,450/407,894) of adults had a documented CKD diagnosis, with prevalence increasing markedly with age (0.73% ages 18–35; 2.90% ages 35–65; 10.88% ages 65–85). Documented CKD prevalence differed by sex, with lower prevalence among males (4.12%, 9,746/236,766) compared with females (6.26%, 10,701/170,911). Race, and language information was largely disclosed but not complete. Race was documented for 93.84% of patients (382,776), while 5.01% (20,450) declined disclosure and 1.14% (4,668) had unknown race. Ethnicity was disclosed for 88.29% (360,124) patients, with 11.71% (47,770) declining or missing. Language preference was undocumented for 4.8% (19,664) patients. Across race, and language strata, patients with unknown demographic information consistently ranked lowest for guideline-concordant testing completion (Table 1). Among patients without a CKD diagnosis (n=387,444), 24.0% (92,946) had diabetes and/or hypertension, yet only 11.30% (10,501/92,946) completed guideline-concordant co-testing (eGFR and uACR) within 12 months. Completion was lowest among patients with hypertension alone (2.28%, 1,301/57,134). Additionally, 3,141 patients met laboratory criteria for CKD without a documented diagnosis; most were male (59.97%,884/3,141) and aged =65 years (70.8%, 2,225/3,141). Among all patients with diagnosed or laboratory-identified CKD (n=23,591), only 16.19% (3,820) had uACR measured in the prior 12 months. Substantial gaps persist in CKD recognition and guideline-recommended testing, particularly among high-risk patients. Incomplete documentation of race, and language is common, and lower completion of guideline-concordant testing is consistently observed among these patients. System-wide strategies incorporating automated laboratory triggers, standardized uACR ordering, and improved demographic data capture may improve early CKD detection and advance equity in kidney care.