Laboratory epidemiology of respiratory viruses in a large children's hospital: A STROBE-compliant article

Ye, Sheng; Wang, Tianlin

Several limitations must be acknowledged in the present study. First, the study was based on clinical samples and some relevant clinical characteristics were not included. For example, the severity of illness of the children, the diagnosis (upper respiratory tract infection vs pneumonia), and clinical outcomes (outpatient visit, hospital or ICU admission) can provide further insights into infection caused by different pathogens. Second, the study was retrospective in design, with inherent limitations such as selection bias and data missing. [26, 27] However, the study was based on electronic healthcare records (EHR), which benefits from large number of samples and could be considered as a kind of big data. [28] [29] [30] With the development of computer technology, the big data analytics have found its way into all walks of life. [31, 32] In our study, the utilization of EHR can provide more insights into the epidemiology of viral respiratory tract infection in children. Third, the indications for the test of respiratory viruses in sputum and/or swab samples were not clearly defined in the study. Because the study was not prospectively designed and the ordering of sample test was largely determined by the treating physician. However, in our clinical practice, there were protocols for ordering tests for respiratory viruses and included patients were suspected to have ARTI. Finally, there are other minor respiratory viruses such as human bocavirus and rhinovirus that were not included in this analysis. Since they are also important pathogens for causing ARTI in children, we will plan future studies by incorporating these viruses.