Electronic Health Record Based Algorithm to Identify Patients with Autism Spectrum Disorder

• DOI: 10.1371/journal.pone.0159621
• Publication Date: 2016-07-29T18:12:46Z
• Authors: Todd Lingren; Pei Chen; Joseph Bochenek; Finale Doshi-Velez; Patty Manning-Cou...; Julie Bickel; Leah Wildenger We...; Judy Reinhold; Nicole Bing; Yizhao Ni; William Barbaresi; Frank Mentch; Melissa Basford; Joshua Denny; Lyam Vazquez; Cassandra Perry; Bahram Namjou; Haijun Qiu; John Connolly; Debra Abrams; Ingrid A. Holm; Beth A. Cobb; Nataline Lingren; Imre Solti; Hakon Hakonarson; Isaac S. Kohane; John Harley; Guergana Savova
• Keywords: Electronic health record

Objective Cohort selection is challenging for large-scale electronic health record (EHR) analyses, as International Classification of Diseases 9th edition (ICD-9) diagnostic codes are notoriously unreliable disease predictors. Our objective was to develop, evaluate, and validate an automated algorithm determining Autism Spectrum Disorder (ASD) patient cohort from EHR. We demonstrate its utility via the largest investigation date co-occurrence patterns medical comorbidities in ASD. Methods extracted ICD-9 concepts derived clinical notes. A gold standard set labeled by clinicians at Boston Children's Hospital (BCH) (N = 150) Cincinnati Medical Center (CCHMC) 152). Two algorithms were created: (1) rule-based implementing ASD criteria Diagnostic Statistical Manual Mental 4th edition, (2) predictive classifier. The positive values (PPV) achieved these compared code baseline. clustered patients based on grouped evaluated subgroups. Results produced best PPV: (a) BCH: 0.885 vs. 0.273 (baseline); (b) CCHMC: 0.840 0.645 (c) combined: 0.864 0.460 (baseline). validation Philadelphia yielded 0.848 (PPV). Clustering analyses three-site large 20,658 patients) identified psychiatric, developmental, seizure disorder clusters. Conclusions In a cross-institutional cohort, ASDs provide further hypothetical evidence distinct courses proposed open avenues other EHR studies individualized treatment