Year-end Sale % OFF 
Abstract
Background: Early diagnosis of primary immunodeficiency disease leads to reductions in illness and decreased healthcare costs. Analysis of electronic health record data may allow for identification of persons at risk of host-defense impairments from within the general population. Our hypothesis was that coded infection history would inform individual risk of disease and ultimately lead to diagnosis.Methods: In this study we assessed individual risk for primary immunodeficiency by analyzing diagnostic codes and pharmacy records from members (n = 185,892) of a large pediatric health network. Relevant infection-associated diagnostic codes were weighted and enumerated for individual members allowing for risk score calculations (“Risk Vital Sign”). At-risk individuals underwent further assessment by chart review and re-analysis of diagnostic codes 12 months later.Results: Of the original cohort, 2188 (1.2%) individuals were identified as medium-high-risk for having a primary immunodeficiency. This group included 41 subjects who were ultimately diagnosed with primary immunodeficiency. An additional 57 medium-high risk patients had coded diagnoses worthy of referral.Conclusions: Population-wide informatics approaches can facilitate disease detection and improve outcomes. Early identification of the 98 patients with confirmed or suspected primary immunodeficiency described here could represent an annual cost savings of up to $7.7 million US Dollars.
Highlights
MATERIALS AND METHODSBiomedical informatics techniques offer potential for understanding large data sets and harnessing vast networks toward optimizing diagnostic accuracy and patient outcomes [1,2,3]
From this focused medium-high risk (MHR) group that received the targeted intervention, 41 (0.02% of Main Cohort; 3.8% of focused MHR Cohort) were coded as having a Primary Immunodeficiency Diseases (PIs) in the 12 months after our original assessment. Another 57 patients (0.03% of Main Cohort; 5.3% of focused MHR Cohort) had medical conditions coded which were concerning for an underlying PI (Figure 1)
The focused MHR group was similar to the general population except that it had a greater percentage of individuals 5 years of age or less (41% in MHR vs. 18.9% in Main)
Summary
MATERIALS AND METHODSBiomedical informatics techniques offer potential for understanding large data sets and harnessing vast networks toward optimizing diagnostic accuracy and patient outcomes [1,2,3]. The National Library of Medicine and the National Institutes of Health recently released statements calling for novel methods of data-driven research to advance biomedical discovery and optimize healthcare [4, 5]. Tools such as natural language processing, machine learning, and computer-aided diagnostic algorithms can mine large data sets such as electronic health records (EHRs) and data warehouses to facilitate quality and precise care [6,7,8,9,10]. Our hypothesis was that coded infection history would inform individual risk of disease and lead to diagnosis
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
