Evaluation of a health information exchange system for microcephaly case-finding — New York City, 2013—2015

• Published in: PloS one 2020-08, Vol.15 (8), p.e0237392-e0237392
• Main Authors: Poirot, Eugenie, Mills, Carrie W, Fair, Andrew D, Graham, Krishika A, Martinez, Emily, Schreibstein, Lauren, Talati, Achala, McVeigh, Katharine H
• Format: Article
• Language: English
• Subjects: Biology and Life Sciences; Birth defects; Childbirth & labor; Codes; Computer and Information Sciences; Congenital defects; Data exchange; Diagnosis; Diagnostic systems; Disease control; Disease prevention; Electronic health records; Electronic medical records; Evaluation; Health care; Health care information services; Hospitals; Hygiene; Immunization; Information systems; Medical diagnosis; Medical records; Medicine and Health Sciences; Microcephaly; Microencephaly; Newborn babies; Public health; Sensitivity; Sentinel surveillance; Statistics; Surveillance; Vector-borne diseases; Viruses; Zika virus
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0237392

Birth defects surveillance in the United States is conducted principally by review of routine but lagged reporting to statewide congenital malformations registries of diagnoses by hospitals or other health care providers, a process that is not designed to rapidly detect changes in prevalence. Health information exchange (HIE) systems are well suited for rapid surveillance, but information is limited about their effectiveness at detecting birth defects. We evaluated HIE data to detect microcephaly diagnosed at birth during January 1, 2013-December 31, 2015 before known introduction of Zika virus in North America. Data from an HIE system were queried for microcephaly diagnostic codes on day of birth or during the first two days after birth at three Bronx hospitals for births to New York City resident mothers. Suspected cases identified by HIE data were compared with microcephaly cases that had been identified through direct inquiry of hospital records and confirmed by chart abstraction in a previous study of the same cohort. Of 16,910 live births, 43 suspected microcephaly cases were identified through an HIE system compared to 67 confirmed cases that had been identified as part of the prior study. A total of 39 confirmed cases were found by both studies (sensitivity = 58.21%, 95% CI: 45.52-70.15%; positive predictive value = 90.70%, 95% CI: 77.86-97.41%; negative predictive value = 99.83%, 95% CI: 99.76-99.89% for HIE data). Despite limitations, HIE systems could be used for rapid newborn microcephaly surveillance, especially in the many jurisdictions where more labor-intensive approaches are not feasible. Future work is needed to improve electronic medical record documentation quality to improve sensitivity and reduce misclassification.