Diagnostic uncertainty of herpangina and hand-foot-and-mouth disease and its impact on national enterovirus syndromic monitoring

• Published in: Epidemiology and infection Vol. 144; no. 7; pp. 1512 - 1519
• Main Authors: YANG, T. O., HUANG, W.-T., CHEN, M.-H., CHEN, P.-C.
• Format: Journal Article
• Language: English
• Published: Cambridge, UK Cambridge University Press 01.05.2016
• Subjects: Acute Disease; Age; Ambulatory care; Birthdays; Case-Control Studies; Child; Child, Preschool; Children & youth; Cohort Studies; Datasets; Enterovirus; Enterovirus - physiology; Female; Hand, Foot and Mouth Disease - diagnosis; Hand, Foot and Mouth Disease - epidemiology; Hand, Foot and Mouth Disease - virology; Hand, Foot, and Mouth Disease; Health care; Herpangina - diagnosis; Herpangina - epidemiology; Herpangina - virology; Humans; Incidence; Infant; Infant, Newborn; Infections; Male; Monitoring systems; National health insurance; Original Papers; Population Surveillance; Respiratory Tract Infections - diagnosis; Respiratory Tract Infections - virology; Syndrome; Taiwan - epidemiology; Uncertainty; United States > US; Taiwan; ISEW, Taiwan; Enterovirus; hand-foot-and-mouth disease; syndromic monitoring; herpangina
• ISSN: 0950-2688; 1469-4409
• DOI: 10.1017/S0950268815002988

The community burden of enterovirus is often monitored through syndromic monitoring systems based on reported cases of enterovirus-related infection (EVI) diagnoses. The extent to which this is affected by under- and over-diagnosis has not been reported. In Taiwan, children often make more than one healthcare visit during an episode of infection. We used change of diagnosis within an episode of infection as a guide of diagnostic uncertainty in a nationally representative cohort of Taiwanese children (n = 13 284) followed from birth to the 9th birthday through electronic health records. We conducted a nested case-control analysis and estimated cross-diagnosis ratios (CDRs) as the observed proportion of acute respiratory infection (ARI) diagnoses following an EVI diagnosis in excess of background ARI burdens. With 19 357 EVI diagnoses in this cohort, the CDR within 7 days was 1·51 (95% confidence interval 1·45–1·57), confirming a significant excess of ARI diagnoses within the week following an EVI diagnosis. We used age-specific CDRs to calibrate the weekly EVI burden in children aged 3–5 years in 2008, and the difference between observed and calibrated weekly EVI burdens was small. Therefore, there was evidence suggesting a small uncertainty in EVI diagnosis, but the observed EVI burdens through syndromic monitoring were not substantially affected by the small uncertainty.