Kidney Week

Abstract: FR-PO0020

Unmasking Occult CKD in Primary Care: The UC ODISEA Project

Session Information

• Artificial Intelligence: Clinical Prediction, Decision Support, and Digital Kidney Care
  October 23, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Artificial Intelligence, Digital Health, and Data Science

• 300 Artificial Intelligence, Digital Health, and Data Science

Authors

• Guerra-Torres, Xavier E., Section of Nephrology and Hypertension, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Xavier E. Guerra-Torres, MD

• Arenas Moncaleano, Iván, Section of Nephrology and Hypertension, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Iván Arenas Moncaleano, MD

• Miranda Cam, Mauricio Alejandro, Section of Nephrology and Hypertension, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Mauricio Alejandro Miranda Cam, MD

• Monroy Condori, Maribel, Section of Nephrology and Hypertension, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Maribel Monroy Condori, MD

• Morales Montoya, Jorge Luis, Section of Nephrology and Hypertension, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Jorge Luis Morales Montoya, MEd

• Timón Zapata, Jesús, Service of Clinical Analysis, Hospital General Universitario Nuestra Señora del Prado, Talavera de la Reina, Spain

Jesús Timón Zapata, PhD

Background

Chronic kidney disease (CKD) affects 15% of the Spanish population and substantially increases mortality risks. Primary care physicians (PCP) manage most patients, however undiagnosed or uncoded CKD remains alarmingly prevalent. Recorded diagnostic rates often fall below 5% despite the availability of laboratory data. This diagnostic gap limits the delivery of evidence-based therapies. We aim to evaluate CKD diagnostic rates by PCP compared to the Sentinel Unit for Optimization of CKD Diagnosis and Follow-up in Primary Care (UC ODISEA) system, an active nephrologist-led analysis of electronic laboratory databases (ELD).

Methods

A retrospective, cross-sectional study was conducted in the Talavera de la Reina health area (Spain). The Institutional Review Board approved the study protocol.CKD diagnostic rates from standard PCP practice were compared with those from the UC ODISEA system. Appropriate filters were applied to ELD to identify potential subjects. CKD was defined as a persistent eGFR <60 mL/min/1.73m2 or a uACR >30 mg/g for at least three months. Extracted data were manually verified to ensure data integrity. Statistical associations were assessed using McNemar’s test in Jamovi, version 2.7.

Results

We analyzed 428 individuals with a median age of 80 years (23-102). The cohort was 54.4% female, and 90.4% were aged 60 years or older. The UC ODISEA system identified 416 patients (97.2%) with CKD. In contrast, PCP identified only 35 patients (8.2%). UC ODISEA system captured an additional 214 patients (p<0.05), representing a 511.43% improvement in case detection (figure 1). Furthermore, among 306 patients requiring standard of care (SoC) therapy, PCP identified only 22 (7.2%). In contrast, UC ODISEA captured 297 (97.1%) of these eligible patients.

Conclusion

Our data shows a profound discrepancy between PCP documentation and active surveillance. Integrating systems like UC ODISEA into electronic health records may potentially reverse the trajectories of CKD complications.

Comparison of CKD identification rates between the UC ODISEA system and PCP.