BACKGROUND: Advances in molecular laboratory techniques are changing the prospects for the diagnosis of viral infectious diseases. Multiplex polymerase chain reaction assay (multiplex-PCR) can detect dozens of pathogens simultaneously, greatly reducing turnaround time (TAT) and improving detection sensitivity. But as a double-edged sword, due to the high sensitivity of PCR, the type of respiratory specimens is critical to diagnosis. In this work, we performed a head-to-head comparison to evaluate the multiplex-PCR yields between two samples, sputum and flocked oropharyngeal swabs (OPS). METHODS: Eleven common respiratory pathogens were tested in hospitalized children< 13 years of age who met the criteria for lower respiratory tract infection by GeXP-based multiplex-PCR of paired OPS and sputum. RESULTS: From January to June 2018, 440 children with paired OPS and sputum were tested. The positive rate was 84% (369/440) for OPS and 88% (386/440) for sputum (p = .007). The frequency of detection of HRV, RSV, Influenza A virus, HMPV, parainfluenza virus, adenovirus, M. pneumoniae, coronavirus, bocavirus and C. pneumoniae in sputa was higher than that of OPSs (all p < .001). Both types of specimens had similarly very good kappa values for most of pathogens, except for Mycoplasma pneumonia (kappa = 0.61) and Chlamydia pneumoniae (kappa = 0.24). Additionally, 79.3% (349/440) of cases showed consistent results between the two types of samples, and they were significantly younger than patients with inconsistent results (p = .002). CONCLUSIONS: Flocked oropharyngeal swabs and sputum performed similarly for the detection of common respiratory pathogens in hospitalized children by multiplex-PCR, except for Mycoplasma pneumoniae and Chlamydia pneumoniae. Young patients are likely to have consistent results between the two specimens.