Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies

Ning Shang; Atlas Khan; Fernanda Polubriaginof; Francesca Zanoni; Karla Mehl; David Fasel; Paul E. Drawz; Robert J. Carrol; Joshua C. Denny; Matthew A. Hathcock; Adelaide M. Arruda-Olson; Peggy L. Peissig; Richard A. Dart; Murray H. Brilliant; Eric B. Larson; David S. Carrell; Sarah Pendergrass; Shefali Setia Verma; Marylyn D. Ritchie; Barbara Benoit; Vivian S. Gainer; Elizabeth W. Karlson; Adam S. Gordon; Gail P. Jarvik; Ian B. Stanaway; David R. Crosslin; Sumit Mohan; Iuliana Ionita-Laza; Nicholas P. Tatonetti; Ali G. Gharavi; George Hripcsak; Chunhua Weng; Krzysztof Kiryluk

doi:10.1038/s41746-021-00428-1

Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies

Ning Shang, Atlas Khan, Fernanda Polubriaginof, Francesca Zanoni, Karla Mehl, David Fasel, Paul E. Drawz, Robert J. Carrol, Joshua C. Denny, Matthew A. Hathcock, Adelaide M. Arruda-Olson, Peggy L. Peissig, Richard A. Dart, Murray H. Brilliant, Eric B. Larson, David S. Carrell, Sarah Pendergrass, Shefali Setia Verma, Marylyn D. Ritchie, Barbara BenoitVivian S. Gainer, Elizabeth W. Karlson, Adam S. Gordon, Gail P. Jarvik, Ian B. Stanaway, David R. Crosslin, Sumit Mohan, Iuliana Ionita-Laza, Nicholas P. Tatonetti, Ali G. Gharavi, George Hripcsak, Chunhua Weng, Krzysztof Kiryluk^*

^*Corresponding author for this work

Pharmacology

Research output: Contribution to journal › Article › peer-review

36 Scopus citations

Abstract

Chronic Kidney Disease (CKD) represents a slowly progressive disorder that is typically silent until late stages, but early intervention can significantly delay its progression. We designed a portable and scalable electronic CKD phenotype to facilitate early disease recognition and empower large-scale observational and genetic studies of kidney traits. The algorithm uses a combination of rule-based and machine-learning methods to automatically place patients on the staging grid of albuminuria by glomerular filtration rate (“A-by-G” grid). We manually validated the algorithm by 451 chart reviews across three medical systems, demonstrating overall positive predictive value of 95% for CKD cases and 97% for healthy controls. Independent case-control validation using 2350 patient records demonstrated diagnostic specificity of 97% and sensitivity of 87%. Application of the phenotype to 1.3 million patients demonstrated that over 80% of CKD cases are undetected using ICD codes alone. We also demonstrated several large-scale applications of the phenotype, including identifying stage-specific kidney disease comorbidities, in silico estimation of kidney trait heritability in thousands of pedigrees reconstructed from medical records, and biobank-based multicenter genome-wide and phenome-wide association studies.

Original language	English (US)
Article number	70
Journal	npj Digital Medicine
Volume	4
Issue number	1
DOIs	https://doi.org/10.1038/s41746-021-00428-1
State	Published - Dec 2021

Funding

The eMERGE Phase III Network was initiated and funded by the National Human Genome Research Institute (NHGRI) through the following grants: U01HG8680 (Columbia University Health Sciences), U01HG8672 (Vanderbilt University Medical Center), U01HG8657 (Kaiser Permanente Washington Health Research Institute/ University of Washington), U01HG8685 (Brigham and Women’s Hospital), U01HG8666 (Cincinnati Children’s Hospital Medical Center), U01HG6379 (Mayo Clinic), U01HG8679 (Geisinger Clinic), U01HG8684 (Children’s Hospital of Philadelphia), U01HG8673 (Northwestern University), MD007593 (Meharry Medical College), U01HG8676 (Partners Healthcare/Broad Institute), and U01HG8664 (Baylor College of Medicine). This work was also funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Kidney Precision Medicine Project (KPMP grant UH3DK114926), the National Library of Medicine grant R01LM013061, and the Precision Medicine Pilot from the Irving Institute/Columbia CTSA (UL1TR001873). Additional sources of funding included R01DK105124 (K.K.), RC2DK116690 (K.K.), and R01LM006910 (G.H.).

ASJC Scopus subject areas

Health Information Management
Health Informatics
Medicine (miscellaneous)
Computer Science Applications

Access to Document

10.1038/s41746-021-00428-1

Cite this

Shang, N., Khan, A., Polubriaginof, F., Zanoni, F., Mehl, K., Fasel, D., Drawz, P. E., Carrol, R. J., Denny, J. C., Hathcock, M. A., Arruda-Olson, A. M., Peissig, P. L., Dart, R. A., Brilliant, M. H., Larson, E. B., Carrell, D. S., Pendergrass, S., Verma, S. S., Ritchie, M. D., ... Kiryluk, K. (2021). Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies. npj Digital Medicine, 4(1), Article 70. https://doi.org/10.1038/s41746-021-00428-1

@article{fe796a8fe4304a8baacca4120872d467,

title = "Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies",

abstract = "Chronic Kidney Disease (CKD) represents a slowly progressive disorder that is typically silent until late stages, but early intervention can significantly delay its progression. We designed a portable and scalable electronic CKD phenotype to facilitate early disease recognition and empower large-scale observational and genetic studies of kidney traits. The algorithm uses a combination of rule-based and machine-learning methods to automatically place patients on the staging grid of albuminuria by glomerular filtration rate (“A-by-G” grid). We manually validated the algorithm by 451 chart reviews across three medical systems, demonstrating overall positive predictive value of 95% for CKD cases and 97% for healthy controls. Independent case-control validation using 2350 patient records demonstrated diagnostic specificity of 97% and sensitivity of 87%. Application of the phenotype to 1.3 million patients demonstrated that over 80% of CKD cases are undetected using ICD codes alone. We also demonstrated several large-scale applications of the phenotype, including identifying stage-specific kidney disease comorbidities, in silico estimation of kidney trait heritability in thousands of pedigrees reconstructed from medical records, and biobank-based multicenter genome-wide and phenome-wide association studies.",

author = "Ning Shang and Atlas Khan and Fernanda Polubriaginof and Francesca Zanoni and Karla Mehl and David Fasel and Drawz, {Paul E.} and Carrol, {Robert J.} and Denny, {Joshua C.} and Hathcock, {Matthew A.} and Arruda-Olson, {Adelaide M.} and Peissig, {Peggy L.} and Dart, {Richard A.} and Brilliant, {Murray H.} and Larson, {Eric B.} and Carrell, {David S.} and Sarah Pendergrass and Verma, {Shefali Setia} and Ritchie, {Marylyn D.} and Barbara Benoit and Gainer, {Vivian S.} and Karlson, {Elizabeth W.} and Gordon, {Adam S.} and Jarvik, {Gail P.} and Stanaway, {Ian B.} and Crosslin, {David R.} and Sumit Mohan and Iuliana Ionita-Laza and Tatonetti, {Nicholas P.} and Gharavi, {Ali G.} and George Hripcsak and Chunhua Weng and Krzysztof Kiryluk",

note = "Funding Information: The eMERGE Phase III Network was initiated and funded by the National Human Genome Research Institute (NHGRI) through the following grants: U01HG8680 (Columbia University Health Sciences), U01HG8672 (Vanderbilt University Medical Center), U01HG8657 (Kaiser Permanente Washington Health Research Institute/ University of Washington), U01HG8685 (Brigham and Women{\textquoteright}s Hospital), U01HG8666 (Cincinnati Children{\textquoteright}s Hospital Medical Center), U01HG6379 (Mayo Clinic), U01HG8679 (Geisinger Clinic), U01HG8684 (Children{\textquoteright}s Hospital of Philadelphia), U01HG8673 (Northwestern University), MD007593 (Meharry Medical College), U01HG8676 (Partners Healthcare/Broad Institute), and U01HG8664 (Baylor College of Medicine). This work was also funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Kidney Precision Medicine Project (KPMP grant UH3DK114926), the National Library of Medicine grant R01LM013061, and the Precision Medicine Pilot from the Irving Institute/Columbia CTSA (UL1TR001873). Additional sources of funding included R01DK105124 (K.K.), RC2DK116690 (K.K.), and R01LM006910 (G.H.). Publisher Copyright: {\textcopyright} 2021, The Author(s).",

year = "2021",

month = dec,

doi = "10.1038/s41746-021-00428-1",

language = "English (US)",

volume = "4",

journal = "npj Digital Medicine",

issn = "2398-6352",

publisher = "Nature Publishing Group",

number = "1",

}

Shang, N, Khan, A, Polubriaginof, F, Zanoni, F, Mehl, K, Fasel, D, Drawz, PE, Carrol, RJ, Denny, JC, Hathcock, MA, Arruda-Olson, AM, Peissig, PL, Dart, RA, Brilliant, MH, Larson, EB, Carrell, DS, Pendergrass, S, Verma, SS, Ritchie, MD, Benoit, B, Gainer, VS, Karlson, EW, Gordon, AS, Jarvik, GP, Stanaway, IB, Crosslin, DR, Mohan, S, Ionita-Laza, I, Tatonetti, NP, Gharavi, AG, Hripcsak, G, Weng, C & Kiryluk, K 2021, 'Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies', npj Digital Medicine, vol. 4, no. 1, 70. https://doi.org/10.1038/s41746-021-00428-1

TY - JOUR

T1 - Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies

AU - Shang, Ning

AU - Khan, Atlas

AU - Polubriaginof, Fernanda

AU - Zanoni, Francesca

AU - Mehl, Karla

AU - Fasel, David

AU - Drawz, Paul E.

AU - Carrol, Robert J.

AU - Denny, Joshua C.

AU - Hathcock, Matthew A.

AU - Arruda-Olson, Adelaide M.

AU - Peissig, Peggy L.

AU - Dart, Richard A.

AU - Brilliant, Murray H.

AU - Larson, Eric B.

AU - Carrell, David S.

AU - Pendergrass, Sarah

AU - Verma, Shefali Setia

AU - Ritchie, Marylyn D.

AU - Benoit, Barbara

AU - Gainer, Vivian S.

AU - Karlson, Elizabeth W.

AU - Gordon, Adam S.

AU - Jarvik, Gail P.

AU - Stanaway, Ian B.

AU - Crosslin, David R.

AU - Mohan, Sumit

AU - Ionita-Laza, Iuliana

AU - Tatonetti, Nicholas P.

AU - Gharavi, Ali G.

AU - Hripcsak, George

AU - Weng, Chunhua

AU - Kiryluk, Krzysztof

N1 - Funding Information: The eMERGE Phase III Network was initiated and funded by the National Human Genome Research Institute (NHGRI) through the following grants: U01HG8680 (Columbia University Health Sciences), U01HG8672 (Vanderbilt University Medical Center), U01HG8657 (Kaiser Permanente Washington Health Research Institute/ University of Washington), U01HG8685 (Brigham and Women’s Hospital), U01HG8666 (Cincinnati Children’s Hospital Medical Center), U01HG6379 (Mayo Clinic), U01HG8679 (Geisinger Clinic), U01HG8684 (Children’s Hospital of Philadelphia), U01HG8673 (Northwestern University), MD007593 (Meharry Medical College), U01HG8676 (Partners Healthcare/Broad Institute), and U01HG8664 (Baylor College of Medicine). This work was also funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Kidney Precision Medicine Project (KPMP grant UH3DK114926), the National Library of Medicine grant R01LM013061, and the Precision Medicine Pilot from the Irving Institute/Columbia CTSA (UL1TR001873). Additional sources of funding included R01DK105124 (K.K.), RC2DK116690 (K.K.), and R01LM006910 (G.H.). Publisher Copyright: © 2021, The Author(s).

PY - 2021/12

Y1 - 2021/12

N2 - Chronic Kidney Disease (CKD) represents a slowly progressive disorder that is typically silent until late stages, but early intervention can significantly delay its progression. We designed a portable and scalable electronic CKD phenotype to facilitate early disease recognition and empower large-scale observational and genetic studies of kidney traits. The algorithm uses a combination of rule-based and machine-learning methods to automatically place patients on the staging grid of albuminuria by glomerular filtration rate (“A-by-G” grid). We manually validated the algorithm by 451 chart reviews across three medical systems, demonstrating overall positive predictive value of 95% for CKD cases and 97% for healthy controls. Independent case-control validation using 2350 patient records demonstrated diagnostic specificity of 97% and sensitivity of 87%. Application of the phenotype to 1.3 million patients demonstrated that over 80% of CKD cases are undetected using ICD codes alone. We also demonstrated several large-scale applications of the phenotype, including identifying stage-specific kidney disease comorbidities, in silico estimation of kidney trait heritability in thousands of pedigrees reconstructed from medical records, and biobank-based multicenter genome-wide and phenome-wide association studies.

AB - Chronic Kidney Disease (CKD) represents a slowly progressive disorder that is typically silent until late stages, but early intervention can significantly delay its progression. We designed a portable and scalable electronic CKD phenotype to facilitate early disease recognition and empower large-scale observational and genetic studies of kidney traits. The algorithm uses a combination of rule-based and machine-learning methods to automatically place patients on the staging grid of albuminuria by glomerular filtration rate (“A-by-G” grid). We manually validated the algorithm by 451 chart reviews across three medical systems, demonstrating overall positive predictive value of 95% for CKD cases and 97% for healthy controls. Independent case-control validation using 2350 patient records demonstrated diagnostic specificity of 97% and sensitivity of 87%. Application of the phenotype to 1.3 million patients demonstrated that over 80% of CKD cases are undetected using ICD codes alone. We also demonstrated several large-scale applications of the phenotype, including identifying stage-specific kidney disease comorbidities, in silico estimation of kidney trait heritability in thousands of pedigrees reconstructed from medical records, and biobank-based multicenter genome-wide and phenome-wide association studies.

UR - http://www.scopus.com/inward/record.url?scp=85104344566&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85104344566&partnerID=8YFLogxK

U2 - 10.1038/s41746-021-00428-1

DO - 10.1038/s41746-021-00428-1

M3 - Article

C2 - 33850243

AN - SCOPUS:85104344566

SN - 2398-6352

VL - 4

JO - npj Digital Medicine

JF - npj Digital Medicine

IS - 1

M1 - 70

ER -

Medical records-based chronic kidney disease phenotype for clinical care and “big data” observational and genetic studies

Abstract

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this