Integrating electronic health data records to develop and validate a predictive model of hospital-acquired acute kidney injury in non-critically ill patients.

acute kidney injury electronic health data records hospital-acquired prediction risk score

Journal

Clinical kidney journal
ISSN: 2048-8505
Titre abrégé: Clin Kidney J
Pays: England
ID NLM: 101579321

Informations de publication

Date de publication:
Dec 2021
Historique:
received: 24 11 2020
accepted: 20 04 2021
entrez: 24 12 2021
pubmed: 25 12 2021
medline: 25 12 2021
Statut: epublish

Résumé

Models developed to predict hospital-acquired acute kidney injury (HA-AKI) in non-critically ill patients have a 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. The study set was 36 852 non-critically ill hospitalized patients admitted from January to 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. This model was then externally validated in 21 545 non-critical patients admitted to the validation centre in the period from June 2017 to December 2018. The incidence of AKI in the study set was 3.9%. Among chronic comorbidities, the highest odds ratios (ORs) were conferred by chronic kidney disease, urologic disease and liver disease. Among acute complications, the highest ORs were associated with acute respiratory failure, anaemia, systemic inflammatory response syndrome, circulatory shock and major surgery. The model showed an area under the curve (AUC) of 0.907 [95% confidence interval (CI) 0.902-0.908), a sensitivity of 82.7 (95% CI 80.7-84.6) and a specificity of 84.2 (95% CI 83.9-84.6) to predict HA-AKI, with an adequate goodness-of-fit for all risk categories (χ 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.

Sections du résumé

BACKGROUND BACKGROUND
Models developed to predict hospital-acquired acute kidney injury (HA-AKI) in non-critically ill patients have a 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 METHODS
The study set was 36 852 non-critically ill hospitalized patients admitted from January to 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. This model was then externally validated in 21 545 non-critical patients admitted to the validation centre in the period from June 2017 to December 2018.
RESULTS RESULTS
The incidence of AKI in the study set was 3.9%. Among chronic comorbidities, the highest odds ratios (ORs) were conferred by chronic kidney disease, urologic disease and liver disease. Among acute complications, the highest ORs were associated with acute respiratory failure, anaemia, systemic inflammatory response syndrome, circulatory shock and major surgery. The model showed an area under the curve (AUC) of 0.907 [95% confidence interval (CI) 0.902-0.908), a sensitivity of 82.7 (95% CI 80.7-84.6) and a specificity of 84.2 (95% CI 83.9-84.6) to predict HA-AKI, with an adequate goodness-of-fit for all risk categories (χ
CONCLUSIONS CONCLUSIONS
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.

Identifiants

pubmed: 34950463
doi: 10.1093/ckj/sfab094
pii: sfab094
pmc: PMC8690094
doi:

Types de publication

Journal Article

Langues

eng

Pagination

2524-2533

Informations de copyright

© The Author(s) 2021. Published by Oxford University Press on behalf of ERA-EDTA.

Références

Clin J Am Soc Nephrol. 2016 Nov 7;11(11):1935-1943
pubmed: 27633727
Kidney Int. 2012 Mar;81(5):477-85
pubmed: 22157656
Lancet. 2016 May 14;387(10032):2017-25
pubmed: 27086173
Stat Med. 2008 Jan 30;27(2):157-72; discussion 207-12
pubmed: 17569110
J Am Soc Nephrol. 1999 Mar;10(3):581-93
pubmed: 10073609
Am J Nephrol. 2012;35(4):349-55
pubmed: 22473149
PLoS One. 2017 Jan 5;12(1):e0169341
pubmed: 28056039
Can J Kidney Health Dis. 2016 Feb 26;3:9
pubmed: 26925245
Ulster Med J. 2014 Sep;83(3):149-57
pubmed: 25484464
Mayo Clin Proc. 2009 May;84(5):410-6
pubmed: 19411437
Med Intensiva. 2016 Aug-Sep;40(6):374-82
pubmed: 27388683
Clin Kidney J. 2019 Nov 07;14(1):309-316
pubmed: 33564433
Nature. 2019 Aug;572(7767):116-119
pubmed: 31367026
Kidney Blood Press Res. 2017;42(5):761-773
pubmed: 29136619
Med Intensiva. 2017 Mar;41(2):116-126
pubmed: 28190602
Health Serv Res. 2005 Oct;40(5 Pt 2):1620-39
pubmed: 16178999
Intensive Care Med. 2015 Aug;41(8):1411-23
pubmed: 26162677
Stat Med. 1995 Oct 15;14(19):2161-72
pubmed: 8552894
Ann Intern Med. 2015 Jan 6;162(1):W1-73
pubmed: 25560730
Kidney Int Rep. 2017 Jul;2(4):519-529
pubmed: 28845471
J Am Soc Nephrol. 2005 Nov;16(11):3365-70
pubmed: 16177006
BMJ Open. 2017 Sep 27;7(9):e016591
pubmed: 28963291
Clin J Am Soc Nephrol. 2015 Sep 4;10(9):1510-8
pubmed: 26231194
Intensive Care Med. 2017 Jun;43(6):855-866
pubmed: 28466146
BMJ Open. 2017 Mar 8;7(3):e013511
pubmed: 28274964
Clin Nutr. 2003 Jun;22(3):321-36
pubmed: 12765673
Clin J Am Soc Nephrol. 2013 Sep;8(9):1482-93
pubmed: 23744003

Auteurs

Alfons Segarra (A)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Jacqueline Del Carpio (J)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Maria Paz Marco (MP)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Elias Jatem (E)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Jorge Gonzalez (J)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Pamela Chang (P)

Department of Nephrology, Arnau de Vilanova University Hospital, Lleida, Spain.

Natalia Ramos (N)

Department of Nephrology, Vall d'Hebron University Hospital, Barcelona, Spain.

Judith de la Torre (J)

Department of Nephrology, Vall d'Hebron University Hospital, Barcelona, Spain.

Joana Prat (J)

Department of Development, Parc Salut Hospital, Barcelona, Spain.

Maria J Torres (MJ)

Department of Informatics, Vall d'Hebron University Hospital, Barcelona, Spain.

Bruno Montoro (B)

Department of Hospital Pharmacy, Vall d'Hebron University Hospital, Barcelona, Spain.

Mercedes Ibarz (M)

Laboratory Department, Arnau de Vilanova University Hospital, Lleida, Spain.

Silvia Pico (S)

Laboratory Department, Arnau de Vilanova University Hospital, Lleida, Spain.

Gloria Falcon (G)

Technical Secretary and Territorial Management of Lleida-Pirineus, Lleida, Spain.

Marina Canales (M)

Technical Secretary and Territorial Management of Lleida-Pirineus, Lleida, Spain.

Elisard Huertas (E)

Informatic Unit of the Catalonian Institute of Health-Territorial Management, Lleida, Spain.

Iñaki Romero (I)

Territorial Management Information Systems, Catalonian Institute of Health, Lleida, Spain.

Nacho Nieto (N)

Department of Informatics, Vall d'Hebron University Hospital, Barcelona, Spain.

Classifications MeSH