Performance of a Computational Phenotyping Algorithm for Sarcoidosis Using Diagnostic Codes in Electronic Medical Records: Case Validation Study From 2 Veterans Affairs Medical Centers.
EMRs
VA
Veterans Affairs
computational phenotype
diagnostic codes
electronic medical records
practice guidelines
sarcoidosis
Journal
JMIR formative research
ISSN: 2561-326X
Titre abrégé: JMIR Form Res
Pays: Canada
ID NLM: 101726394
Informations de publication
Date de publication:
02 Mar 2022
02 Mar 2022
Historique:
received:
29
06
2021
accepted:
24
01
2022
revised:
24
01
2022
pubmed:
27
1
2022
medline:
27
1
2022
entrez:
26
1
2022
Statut:
epublish
Résumé
Electronic medical records (EMRs) offer the promise of computationally identifying sarcoidosis cases. However, the accuracy of identifying these cases in the EMR is unknown. The aim of this study is to determine the statistical performance of using the International Classification of Diseases (ICD) diagnostic codes to identify patients with sarcoidosis in the EMR. We used the ICD diagnostic codes to identify sarcoidosis cases by searching the EMRs of the San Francisco and Palo Alto Veterans Affairs medical centers and randomly selecting 200 patients. To improve the diagnostic accuracy of the computational algorithm in cases where histopathological data are unavailable, we developed an index of suspicion to identify cases with a high index of suspicion for sarcoidosis (confirmed and probable) based on clinical and radiographic features alone using the American Thoracic Society practice guideline. Through medical record review, we determined the positive predictive value (PPV) of diagnosing sarcoidosis by two computational methods: using ICD codes alone and using ICD codes plus the high index of suspicion. Among the 200 patients, 158 (79%) had a high index of suspicion for sarcoidosis. Of these 158 patients, 142 (89.9%) had documentation of nonnecrotizing granuloma, confirming biopsy-proven sarcoidosis. The PPV of using ICD codes alone was 79% (95% CI 78.6%-80.5%) for identifying sarcoidosis cases and 71% (95% CI 64.7%-77.3%) for identifying histopathologically confirmed sarcoidosis in the EMRs. The inclusion of the generated high index of suspicion to identify confirmed sarcoidosis cases increased the PPV significantly to 100% (95% CI 96.5%-100%). Histopathology documentation alone was 90% sensitive compared with high index of suspicion. ICD codes are reasonable classifiers for identifying sarcoidosis cases within EMRs with a PPV of 79%. Using a computational algorithm to capture index of suspicion data elements could significantly improve the case-identification accuracy.
Sections du résumé
BACKGROUND
BACKGROUND
Electronic medical records (EMRs) offer the promise of computationally identifying sarcoidosis cases. However, the accuracy of identifying these cases in the EMR is unknown.
OBJECTIVE
OBJECTIVE
The aim of this study is to determine the statistical performance of using the International Classification of Diseases (ICD) diagnostic codes to identify patients with sarcoidosis in the EMR.
METHODS
METHODS
We used the ICD diagnostic codes to identify sarcoidosis cases by searching the EMRs of the San Francisco and Palo Alto Veterans Affairs medical centers and randomly selecting 200 patients. To improve the diagnostic accuracy of the computational algorithm in cases where histopathological data are unavailable, we developed an index of suspicion to identify cases with a high index of suspicion for sarcoidosis (confirmed and probable) based on clinical and radiographic features alone using the American Thoracic Society practice guideline. Through medical record review, we determined the positive predictive value (PPV) of diagnosing sarcoidosis by two computational methods: using ICD codes alone and using ICD codes plus the high index of suspicion.
RESULTS
RESULTS
Among the 200 patients, 158 (79%) had a high index of suspicion for sarcoidosis. Of these 158 patients, 142 (89.9%) had documentation of nonnecrotizing granuloma, confirming biopsy-proven sarcoidosis. The PPV of using ICD codes alone was 79% (95% CI 78.6%-80.5%) for identifying sarcoidosis cases and 71% (95% CI 64.7%-77.3%) for identifying histopathologically confirmed sarcoidosis in the EMRs. The inclusion of the generated high index of suspicion to identify confirmed sarcoidosis cases increased the PPV significantly to 100% (95% CI 96.5%-100%). Histopathology documentation alone was 90% sensitive compared with high index of suspicion.
CONCLUSIONS
CONCLUSIONS
ICD codes are reasonable classifiers for identifying sarcoidosis cases within EMRs with a PPV of 79%. Using a computational algorithm to capture index of suspicion data elements could significantly improve the case-identification accuracy.
Identifiants
pubmed: 35081036
pii: v6i3e31615
doi: 10.2196/31615
pmc: PMC8928044
doi:
Types de publication
Journal Article
Langues
eng
Pagination
e31615Subventions
Organisme : NLM NIH HHS
ID : T15 LM007442
Pays : United States
Organisme : NCATS NIH HHS
ID : TL1 TR001871
Pays : United States
Informations de copyright
©Mohamed I Seedahmed, Izabella Mogilnicka, Siyang Zeng, Gang Luo, Mary A Whooley, Charles E McCulloch, Laura Koth, Mehrdad Arjomandi. Originally published in JMIR Formative Research (https://formative.jmir.org), 02.03.2022.
Références
Ann Am Thorac Soc. 2016 Aug;13(8):1244-52
pubmed: 27509154
Ann Am Thorac Soc. 2017 Dec;14(Supplement_6):S437-S444
pubmed: 29073361
Pharmacoepidemiol Drug Saf. 2006 Apr;15(4):245-52
pubmed: 16374899
Eur Respir J. 2014 Oct;44(4):985-93
pubmed: 25142485
Lung. 2017 Dec;195(6):713-715
pubmed: 28993879
Ann Am Thorac Soc. 2017 Dec;14(Supplement_6):S415-S420
pubmed: 29048937
Curr Opin Pulm Med. 2019 Sep;25(5):484-496
pubmed: 31365383
Curr Protoc Hum Genet. 2019 Jan;100(1):e80
pubmed: 30516347
BMC Med Inform Decis Mak. 2019 Apr 4;19(Suppl 3):78
pubmed: 30943974
Am J Respir Crit Care Med. 2020 Apr 15;201(8):e26-e51
pubmed: 32293205
Int J Cancer. 2009 Jun 1;124(11):2697-700
pubmed: 19230028
N Engl J Med. 1988 Feb 11;318(6):352-5
pubmed: 3123929
Ann Am Thorac Soc. 2015 Oct;12(10):1561-71
pubmed: 26193069
Am J Epidemiol. 1997 Feb 1;145(3):234-41
pubmed: 9012596
J Am Med Inform Assoc. 2009 May-Jun;16(3):371-9
pubmed: 19261943
Br Med J. 1961 Nov 4;2(5261):1165-72
pubmed: 14497750
BMC Med Inform Decis Mak. 2020 Feb 3;20(1):16
pubmed: 32013925
Front Immunol. 2020 Jul 14;11:1443
pubmed: 32760396
AMIA Annu Symp Proc. 2012;2012:606-15
pubmed: 23304333
AMIA Annu Symp Proc. 2018 Apr 16;2017:912-920
pubmed: 29854158
Am J Manag Care. 2002 Jan;8(1):37-43
pubmed: 11814171
PLoS One. 2015 Aug 24;10(8):e0136651
pubmed: 26301417
Sci Rep. 2019 Aug 14;9(1):11862
pubmed: 31413285
PLoS One. 2012;7(9):e44818
pubmed: 22984568
J Am Med Inform Assoc. 2018 Feb 1;25(2):150-157
pubmed: 28645207
Contemp Clin Trials Commun. 2018 Jul 10;11:107-112
pubmed: 30035242
Can Respir J. 2012 Mar-Apr;19(2):e5-9
pubmed: 22536584
Chest. 2011 Jan;139(1):144-50
pubmed: 20595459
Ann Am Thorac Soc. 2017 Jun;14(6):880-887
pubmed: 28355518
Chest. 2015 Feb;147(2):438-449
pubmed: 25188873
Front Med (Lausanne). 2020 Oct 29;7:568020
pubmed: 33195314
Curr Opin Pulm Med. 2020 Sep;26(5):527-534
pubmed: 32701677
Am J Respir Crit Care Med. 2001 Nov 15;164(10 Pt 1):1885-9
pubmed: 11734441
J Am Med Inform Assoc. 2021 Jun 12;28(6):1207-1215
pubmed: 33638343
Clin Exp Rheumatol. 2019 Nov-Dec;37(6):1052-1064
pubmed: 31498063
F1000Prime Rep. 2014 Oct 01;6:89
pubmed: 25374667
Am Rev Respir Dis. 1973 Apr;107(4):615-20
pubmed: 4697670