Do changes in health reveal the possibility of undiagnosed pancreatic cancer? Development of a risk-prediction model based on healthcare claims data.
Journal
PloS one
ISSN: 1932-6203
Titre abrégé: PLoS One
Pays: United States
ID NLM: 101285081
Informations de publication
Date de publication:
2019
2019
Historique:
received:
12
03
2019
accepted:
04
06
2019
entrez:
26
6
2019
pubmed:
27
6
2019
medline:
12
2
2020
Statut:
epublish
Résumé
Early detection methods for pancreatic cancer are lacking. We aimed to develop a prediction model for pancreatic cancer based on changes in health captured by healthcare claims data. We conducted a case-control study on 29,646 Medicare-enrolled patients aged 68 years and above with pancreatic ductal adenocarcinoma (PDAC) reported to the Surveillance Epidemiology an End Results (SEER) tumor registries program in 2004-2011 and 88,938 age and sex-matched controls. We developed a prediction model using multivariable logistic regression on Medicare claims for 16 risk factors and pre-diagnostic symptoms of PDAC present within 15 months prior to PDAC diagnosis. Claims within 3 months of PDAC diagnosis were excluded in sensitivity analyses. We evaluated the discriminatory power of the model with the area under the receiver operating curve (AUC) and performed cross-validation by bootstrapping. The prediction model on all cases and controls reached AUC of 0.68. Excluding the final 3 months of claims lowered the AUC to 0.58. Among new-onset diabetes patients, the prediction model reached AUC of 0.73, which decreased to 0.63 when claims from the final 3 months were excluded. Performance measures of the prediction models was confirmed by internal validation using the bootstrap method. Models based on healthcare claims for clinical risk factors, symptoms and signs of pancreatic cancer are limited in classifying those who go on to diagnosis of pancreatic cancer and those who do not, especially when excluding claims that immediately precede the diagnosis of PDAC.
Sections du résumé
BACKGROUND AND OBJECTIVE
Early detection methods for pancreatic cancer are lacking. We aimed to develop a prediction model for pancreatic cancer based on changes in health captured by healthcare claims data.
METHODS
We conducted a case-control study on 29,646 Medicare-enrolled patients aged 68 years and above with pancreatic ductal adenocarcinoma (PDAC) reported to the Surveillance Epidemiology an End Results (SEER) tumor registries program in 2004-2011 and 88,938 age and sex-matched controls. We developed a prediction model using multivariable logistic regression on Medicare claims for 16 risk factors and pre-diagnostic symptoms of PDAC present within 15 months prior to PDAC diagnosis. Claims within 3 months of PDAC diagnosis were excluded in sensitivity analyses. We evaluated the discriminatory power of the model with the area under the receiver operating curve (AUC) and performed cross-validation by bootstrapping.
RESULTS
The prediction model on all cases and controls reached AUC of 0.68. Excluding the final 3 months of claims lowered the AUC to 0.58. Among new-onset diabetes patients, the prediction model reached AUC of 0.73, which decreased to 0.63 when claims from the final 3 months were excluded. Performance measures of the prediction models was confirmed by internal validation using the bootstrap method.
CONCLUSION
Models based on healthcare claims for clinical risk factors, symptoms and signs of pancreatic cancer are limited in classifying those who go on to diagnosis of pancreatic cancer and those who do not, especially when excluding claims that immediately precede the diagnosis of PDAC.
Identifiants
pubmed: 31237889
doi: 10.1371/journal.pone.0218580
pii: PONE-D-19-07228
pmc: PMC6592596
doi:
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Langues
eng
Sous-ensembles de citation
IM
Pagination
e0218580Subventions
Organisme : NCI NIH HHS
ID : R21 CA220073
Pays : United States
Organisme : NIDDK NIH HHS
ID : U01 DK108314
Pays : United States
Déclaration de conflit d'intérêts
The authors have declared that no competing interests exist.
Références
Ann Oncol. 2014 Oct;25(10):2065-72
pubmed: 25057164
Med Care. 2002 May;40(5):395-404
pubmed: 11961474
Clin Gastroenterol Hepatol. 2014 Oct;12(10):1635-44.e5; quiz e103
pubmed: 24509242
Eur J Cancer. 2011 Sep;47(13):1928-37
pubmed: 21458985
Br J Cancer. 2005 Jun 6;92(11):2076-83
pubmed: 15886696
J Natl Cancer Inst. 2019 Jan 1;111(1):1-2
pubmed: 29917095
PLoS One. 2016 Jan 11;11(1):e0146473
pubmed: 26752291
PLoS One. 2013 Sep 13;8(9):e72311
pubmed: 24058443
BMJ Open. 2014 Nov 19;4(11):e005720
pubmed: 25410605
Ann Intern Med. 2019 Jan 1;170(1):51-58
pubmed: 30596875
Cancer Epidemiol. 2016 Dec;45:145-150
pubmed: 27810486
Gastroenterology. 2018 Sep;155(3):730-739.e3
pubmed: 29775599
Am J Epidemiol. 2011 May 1;173(9):1059-68
pubmed: 21430193
Health Serv Res. 2005 Apr;40(2):517-37
pubmed: 15762905
PLoS Med. 2007 Oct 16;4(10):e296
pubmed: 17941714
Biometrics. 2001 Mar;57(1):120-5
pubmed: 11252586
Med Care. 2002 Aug;40(8 Suppl):IV-3-18
pubmed: 12187163
Gastroenterology. 2005 Aug;129(2):504-11
pubmed: 16083707
Ann Oncol. 2012 Feb;23(2):374-82
pubmed: 21536662
Stat Med. 1996 Feb 28;15(4):361-87
pubmed: 8668867
Cancer Res. 2014 Jun 1;74(11):2913-21
pubmed: 24840647
Lancet Oncol. 2009 Jan;10(1):88-95
pubmed: 19111249
Br J Cancer. 2012 Jun 5;106(12):1940-4
pubmed: 22617126
Gastroenterology. 2017 Mar;152(4):840-850.e3
pubmed: 27923728
J Clin Epidemiol. 2001 Aug;54(8):774-81
pubmed: 11470385
Health Syst Transit. 2015;17(5):1-126
pubmed: 27049966
Ann Oncol. 2012 Nov;23(11):2964-70
pubmed: 22767586
Am J Epidemiol. 2011 Oct 1;174(7):860-70
pubmed: 21821540
Pancreas. 2016 Aug;45(7):986-91
pubmed: 26692445
Epidemiology. 2010 Jan;21(1):128-38
pubmed: 20010215
Am J Epidemiol. 2015 Jul 1;182(1):26-34
pubmed: 26049860
Med Care. 2002 Aug;40(8 Suppl):IV-55-61
pubmed: 12187169