A Computable Phenotype Improves Cohort Ascertainment in a Pediatric Pulmonary Hypertension Registry - 24/08/17

Doi : 10.1016/j.jpeds.2017.05.037

Alon Geva, MD, MPH ^1,^2,³, Jessica L. Gronsbell, BA ⁴, Tianxi Cai, ScD ⁴, Tianrun Cai, MD ⁵, Shawn N. Murphy, MD, PhD ^6,^7,⁸, Jessica C. Lyons, MS ⁸, Michelle M. Heinz, BS ¹, Marc D. Natter, MD ^1,⁹, Nandan Patibandla, MS ¹⁰, Jonathan Bickel, MD, MS ^1,^9,¹⁰, Mary P. Mullen, MD, PhD ^9,¹¹, Kenneth D. Mandl, MD, MPH ^1,^8,^9,^*
for the
Pediatric Pulmonary Hypertension Network and National Heart, Lung, and Blood Institute Pediatric Pulmonary Vascular Disease Outcomes Bioinformatics Clinical Coordinating Center Investigators^*
A list of additional member of the PPHNet and NHLBI Pediatric Pulmonary Vascular Disease Outcomes Bioinformatics Clinical Coordinating Center is available at www.jpeds.com (Appendix).
Steven Abman, MD ¹, Ian Adatia, MBChB, FRCP(C) ², Eric D. Austin, MD, MSCI ³, Jeffrey Feinstein, MD ⁴, Jeff Fineman, MD ⁵, Brian Hanna, MD ⁶, Rachel Hopper, MD ⁶, Dunbar Ivy, MD ¹, Roberta Keller, MD ⁵, Usha Krishnan, MD ⁷, Thomas Kulik, MD ⁸, Mary Mullen, MD, PhD ⁸, Usha Raj, MD ⁹, Erika Berman Rosenzweig, MD ⁷
¹ University of Colorado, Denver, CO
² Stollery Children's Hospital, Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB, Canada
³ Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, TN
⁴ Stanford University, Stanford, CA
⁵ University of California, San Francisco, San Francisco, CA
⁶ Children's Hospital of Philadelphia, Philadelphia, PA
⁷ Columbia University, New York-Presbyterian, Morgan Stanley Children's Hospital, New York, NY
⁸ Boston Children's Hospital, Boston, MA
⁹ University of Illinois, Chicago, Chicago, IL

¹ Computational Health Informatics Program, Boston Children's Hospital, Boston, MA
² Division of Critical Care Medicine, Department of Anesthesiology, Perioperative, and Pain Medicine, Boston Children's Hospital, Boston, MA
³ Department of Anesthesia, Harvard Medical School, Boston, MA
⁴ Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA
⁵ Division of Rheumatology, Immunology and Allergy, Brigham and Women's Hospital, Boston, MA
⁶ Department of Research Information Services and Computing, Partners Healthcare, Boston, MA
⁷ Department of Neurology, Massachusetts General Hospital, Boston, MA
⁸ Department of Biomedical Informatics, Harvard Medical School, Boston, MA
⁹ Department of Pediatrics, Harvard Medical School, Boston, MA
¹⁰ Information Services Department, Boston Children's Hospital, Boston, MA
¹¹ Department of Cardiology, Boston Children's Hospital, Boston, MA

^*Reprint requests: Kenneth D. Mandl, MD, MPH, Computational Health Informatics Program, Boston Children's Hospital, 300 Longwood Ave, 1 Autumn 535, Mail Stop BCH3187, Boston, MA 02115.Computational Health Informatics ProgramBoston Children's Hospital300 Longwood Ave, 1 Autumn 535, Mail Stop BCH3187BostonMA02115

Abstract

Objectives

To compare registry and electronic health record (EHR) data mining approaches for cohort ascertainment in patients with pediatric pulmonary hypertension (PH) in an effort to overcome some of the limitations of registry enrollment alone in identifying patients with particular disease phenotypes.

Study design

This study was a single-center retrospective analysis of EHR and registry data at Boston Children's Hospital. The local Informatics for Integrating Biology and the Bedside (i2b2) data warehouse was queried for billing codes, prescriptions, and narrative data related to pediatric PH. Computable phenotype algorithms were developed by fitting penalized logistic regression models to a physician-annotated training set. Algorithms were applied to a candidate patient cohort, and performance was evaluated using a separate set of 136 records and 179 registry patients. We compared clinical and demographic characteristics of patients identified by computable phenotype and the registry.

Results

The computable phenotype had an area under the receiver operating characteristics curve of 90% (95% CI, 85%-95%), a positive predictive value of 85% (95% CI, 77%-93%), and identified 413 patients (an additional 231%) with pediatric PH who were not enrolled in the registry. Patients identified by the computable phenotype were clinically distinct from registry patients, with a greater prevalence of diagnoses related to perinatal distress and left heart disease.

Conclusions

Mining of EHRs using computable phenotypes identified a large cohort of patients not recruited using a classic registry. Fusion of EHR and registry data can improve cohort ascertainment for the study of rare diseases.

Trial registration

ClinicalTrials.gov: NCT02249923.

Le texte complet de cet article est disponible en PDF.

Keywords : pulmonary hypertension, pediatrics, registry, computer-based model, bioinformatics

Abbreviations : AUC, CDH, COD, EHR, i2b2, ICD-9, NILE, NLP, PH, PPHNet, RV, RVOTO

Plan

Methods

Results

Discussion

Supported by the National Institutes of Health (Grants NHLBI U01HL121518, NHLBI L40HL133929, NIGMS R01 GM104303, PCORI CDRN130604608, and NICHD T32HD040128). The authors declare no conflicts of interest.

Export

Vol 188

P. 224 - septembre 2017 Retour au numéro

Article précédent

Real-Time Polymerase Chain Reaction–Based Detection of Bordetella pertussis in Mexican Infants and Their Contacts: A 3-Year Multicenter Study
Alejandra Aquino-Andrade, Gabriel Martínez-Leyva, Jocelin Mérida-Vieyra, Patricia Saltigeral, Antonino Lara, Wendy Domínguez, Silvestre García de la Puente, Agustín De Colsa

| Article suivant

Twenty-Five Year Survival of Children with Intellectual Disability in Western Australia
Jenny Bourke, Wendy N. Nembhard, Kingsley Wong, Helen Leonard

Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.

Déjà abonné à cette revue ?

connectez-vous ou créez un compte

A Computable Phenotype Improves Cohort Ascertainment in a Pediatric Pulmonary Hypertension Registry - 24/08/17

Abstract

Objectives

Study design

Results

Conclusions

Trial registration

Plan

Export citations

Fichier

Contenu

Accès rapides

Mon compte

Aide & support

Plateformes Elsevier Masson

Déclaration CNIL