Kidney Week

Abstract: TH-PO023

Integrating Electronic Health Data Records to Develop and Validate a Predictive Model of Hospital-Acquired AKI in Non-Critically Ill Patients

Session Information

• AI, Digital Health, Data Science - I
  November 02, 2023 | Location: Exhibit Hall, Pennsylvania Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Augmented Intelligence, Digital Health, and Data Science

• 300 Augmented Intelligence, Digital Health, and Data Science

Authors

• Segarra, Alfonso, Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

Alfonso Segarra,

• Del Carpio Salas, Jacqueline, Universitat Autonoma de Barcelona, Barcelona, Catalunya, Spain

Jacqueline Del Carpio Salas,

• Ramos, Natalia, Hospital Universitari Vall d'Hebron, Barcelona, Catalunya, Spain

Natalia Ramos,

• González, Jorge, Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

Jorge González,

• Montoro Ronsano, Jose Bruno, Hospital Universitari Vall d'Hebron, Barcelona, Catalunya, Spain

Jose Bruno Montoro Ronsano,

• Pico Fornies, Silvia, Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

Silvia Pico Fornies,

• Canales Navarro, Marina, Institut Catala de la Salut, Lleida, Catalunya, Spain

Marina Canales Navarro,

• Marco, María Paz, Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

María Paz Marco,

• Jatem, Elias A., Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

Elias A. Jatem,

• Chang, Pamela Janet, Hospital Universitari Arnau de Vilanova, Lleida, Catalunya, Spain

Pamela Janet Chang,

Background

Models developed to predict hospital-acquired acute kidney injury(HA-AKI)in non-critically ill patients have low sensitivity,do not include dynamic changes of risk factors and do not allow the establishment of a time relationship between exposure to risk factors and AKI.We developed and externally validated a predictive model of HA-AKI integrating electronic health databases and recording the exposure to risk factors prior to the detection of AKI.

Methods

The study set 36852 non-critically ill hospitalized patients admitted from January-December 2017.Using stepwise logistic analyses,including demography,chronic comorbidities and exposure to risk factors prior to AKI detection,we developed a multivariate model to predict HA-AKI. Externally validated in 21545 non-critical patients admitted to the validation centre from June 2017-December-2018.

Results

Incidence of AKI in study set was 3.9%.Among chronic comorbidities,the highest ORs chronic kidney disease,urologic disease and liver disease.Among acute complications,the highest ORs acute respiratory failure,anaemia,systemic inflammatory response syndrome,circulatory shock and major surgery. Model showed AUC 0.907,sensitivity 82.7 and specificity 84.2 to predict HA-AKI.In validation set,incidence of AKI 3.2%.Model showed AUC 0.905,sensitivity 81.2 and specificity 82.5 to predict HA-AKI and had an adequate goodness-of-fit.

Conclusion

By using electronic health data records,our study provides a model that can be used in clinical practice to obtain an accurate dynamic and updated assessment of the individual risk of HA-AKI during the hospital admission period in non-critically ill patients.

Schematic representation of the interrelation between electronic databases performed to obtain updated clinical information during hospital stay.

Funding

• Commercial Support – Amgen S.A and Menarini S.A