Discovering novel disease comorbidities using electronic medical records

• DOI: 10.1371/journal.pone.0225495
• Publication Date: 2019-11-27T13:39:55Z
• Authors: Shikha Chaganti; Valerie F. Welty; Warren Taylor; Kimberly Albert; Michelle D. Failla; Carissa Cascio; Seth Smith; Louise Mawn; Susan M. Resnick; Lori L. Beason-Held; Francesca Bagnato; Thomas Lasko; Jeffrey D. Blume; Bennett A. Landman
• Keywords: Leverage (statistics); Medical record; Diagnosis code; Rochester Epidemiology Project

Increasing reliance on electronic medical records at large centers provides unique opportunities to perform population level analyses exploring disease progression and etiology. The massive accumulation of diagnostic, procedure, laboratory codes in one place has enabled the exploration co-occurring conditions, their risk factors, potential prognostic factors. While most readily identifiable associations are (now) well known scientific community, there is no doubt many more relationships still be uncovered EMR data. In this paper, we introduce a novel finding index help with that task. This new uses data mined from real-time PubMed abstracts indicate extent which empirically discovered already (i.e., present literature). Our methods leverage second-generation p-values, better identify truly clinically meaningful. We illustrate our method three examples: Autism Spectrum Disorder, Alzheimer's Disease, Optic Neuritis. results demonstrate wide utility for identifying have highest priority among complex web correlations causalities. Data scientists clinicians can work together effectively discover both reliable understudied.