The COVID-19 pandemic as an opportunity for unravelling the causative association between respiratory viruses and pneumococcus-associated disease in young children: a prospective study

BACKGROUND: In young children, rates of lower respiratory infections (LRI) and invasive pneumococcal disease (IPD) have been associated with respiratory syncytial virus (RSV), human metapneumovirus (hMPV), influenza (flu), and parainfluenza (PIV) (collectively termed here as pneumonia and pneumococcal disease-associated viruses [PDA-viruses]). However, their contribution to the pathogenesis of these disease endpoints has not yet been elucidated. The COVID-19 pandemic provided a unique opportunity to examine the question. METHODS: This prospective study comprised all children <5 years, living in southern Israel, during 2016 through 2021. The data were previously collected in multiple ongoing prospective surveillance programs and include: hospital visits for community-acquired alveolar pneumonia (CAAP), non-CAAP LRI; nasopharyngeal pneumococcal carriage (<3 years of age); respiratory virus activity; and nationwide, all-ages COVID-19 episodes and IPD in children <5 years. A hierarchical statistical model was developed to estimate the proportion of the different clinical endpoints attributable to each virus from monthly time series data, stratified by age and ethnicity. A separate model was fit for each endpoint, with covariates that included a linear time trend, 12-month harmonic variables to capture unexplained seasonal variations, and the proportion of tests positive for each virus in that month. FINDINGS: During 2016 through 2021, 3,204, 26,695, 257, and 619 episodes of CAAP, non-CAAP LRI, pneumococcal bacteremic pneumonia and non-pneumonia IPD, respectively, were reported. Compared to 2016–2019, broad declines in the disease endpoints were observed shortly after the pandemic surge, coincident with a complete disappearance of all PDA-viruses and continued circulation of rhinovirus (RhV) and adenovirus (AdV). From April 2021, off-season and abrupt surges of all disease endpoints occurred, associated with similar dynamics among the PDA-viruses, which re-emerged sequentially. Using our model fit to the entire 2016–2021 period, 82% (95% CI, 75–88%) of CAAP episodes in 2021 were attributable to the common respiratory viruses, as were 22%–31% of the other disease endpoints. Virus-specific contributions to CAAP were: RSV, 49% (95% CI, 43–55%); hMPV, 13% (10–17%); PIV, 11% (7–15%); flu, 7% (1–13%). RhV and AdV did not contribute. RSV was the main contributor in all endpoints, especially in infants. Pneumococcal carriage prevalence remained largely stable throughout the study. INTERPRETATION: RSV and hMPV play a critical role in the burden of CAAP and pneumococcal disease in children. Interventions targeting these viruses could have a secondary effect on the burden of disease typically attributed to bacteria. FUNDING: There was no funding for this study.