Assistant Professor Texas Children's Hospital, Baylor College of Medicine Baylor College of Medicine Houston, Texas, United States
Background: Congenital Cytomegalovirus (cCMV) remains the leading cause of non-hereditary hearing loss worldwide, yet many questions remain as to the disease course in the era of antivirals.
Objective: The widespread adoption of Electronic Medical Records (EMRs) at large academic centers suggests that high-throughput informatics approaches would be a reasonable means to extract relevant clinical information to this end. In this study we used a series of database queries to extract records directly from our EMR to build a database of outcomes for patients with cCMV. We developed a model for clinical progression of disease in cCMV using semantic terms to guide disease state classification.
Design/Methods: SlicerDicerTM, a reporting tool in our Epic® EMR, was used to identify patients seen in our cCMV Clinic at Texas Children’s Hospital in Houston, Texas, with a diagnosis of cytomegalovirus (CMV) infection. The ICD-10 diagnoses, encounters, imaging/diagnostic studies, hospital stays, medications, labs, and surgical procedures were extracted from our EMR with a series of generalizable Structured Query Language (SQL) queries. Using SAS V9.4 (Cary, N.C.), we parsed this information into relational databases and categorized ICD-10 codes by semantic terms known to be associated with cCMV using SNOMED-CT®.
Results: Data acquisition took less than a minute to complete and included 55,558 unique encounters, 48,582 billing diagnoses, 3,594 problem list records, 56,725 lab/imaging studies, 47,087 medication administration records, and 864 procedure records. Of 190 patients seen in our clinic with an ICD-10 diagnosis of CMV infection, 152 had cCMV, and 112 of these were born after 1/1/2008 (the inception date of our EMR). Between 2008-2018, these 112 patients accounted for a total of 11,653 provider encounters. Of these patients, 93 (83%) had hearing deficits, 17 (15%) had cochlear implants, 89 (79%) had developmental abnormalities, and 41 (37%) had neurologic abnormalities noted in the EMR (Figure 1). Conclusion(s): The spectrum of disease of cCMV is broad and over time these patients contribute significantly to healthcare utilization. The EMR gives us the potential to further study this disease in finer detail and identify rates of disease progression by mining the ICD-10 codes associated with these patients throughout time. These results should prove invaluable for anticipatory guidance and in generating cost-models for the economic impact of cCMV.
Authors/Institutions: Ryan Rochat, Baylor College of Medicine, Houston, Texas, United States; Gail J. Demmler-Harrison, Baylor College of Medicine, Houston, Texas, United States
