Among the many unknowns that have featured in the steep learning curve of our covid-19 year, the question of risks to (and from) children has been an ongoing area of uncertainty. At first there was the conundrum that the young seemed so little affected by the infection that at least one world leader pronounced that they were “somehow immune.”
A year into the pandemic we have a much clearer perspective, especially from random testing surveys such as REACT, seroprevalence studies, and contact-tracing studies. The overwhelming observation has been that, across the pandemic, children and adolescents are considerably less likely to suffer severe disease or to be hospitalised. In transmission-tracing studies, children are roughly half as likely as adults to become infected.  While this is still the case during the latest wave in the context of current variants, the age distribution of cases has shifted downwards to some degree.
In general, child and adolescent cases account for no more than a few per cent of reported, symptomatic cases and few severe or fatal cases. The difference may relate partly to the lower expression of ACE2 in younger children. However, seroprevalence is similar between adolescents and adults suggesting a high prevalence of asymptomatic spread in the former group. [1,2] Since access to testing has for the most part been predicated on presence of case-defining symptoms (and these have been defined in adults where they are likely to be more explicit), the likelihood is that the caseload in children has been under-estimated. This seems to be confirmed by comparison between unbiased population sampling (such as the REACT study) and symptom-based surveys. 
The clinical presentation and immune profile in those children who are hospitalised seems similar to the disease in adults. An exception to this is the relatively small subset who develop multisystem inflammatory syndrome in children (MIS-C)—a disease of vascular involvement, shock, and strong T cell activation.
Where does this leave us in terms of evaluating risks—both to and from children—around the return to school? The first point to consider is whether the high seroprevalence among adolescents offers potential hubs of super-spreaders for the community? Certainly, in those children who become sufficiently unwell for hospitalisation, viral load reaches much higher levels than seen in adults, though it is unclear whether these high levels are also seen in asymptomatic disease. While some reported contact-tracing studies in school settings identified relatively little spread, other studies indicate that teachers engaged in face-to-face teaching in school have roughly twice the infection risk of those doing online teaching. [3-5] Also, the data indicate that jumps in the R value have followed term date restarts in countries where children have been at school during the pandemic.  This scenario is somewhat reminiscent of the role played by children in transmission of winter flu: themselves relatively unlikely to suffer severe disease, but playing a significant role of spread into the community and to older relatives.
Arguably the biggest unknown is the evaluation of risk of developing long covid during asymptomatic spread in a school setting. Estimating the prevalence of long covid in adults remains a considerable ongoing challenge, and for children and adolescents, considerably more so. Until more detailed studies are completed, evidence of the total long covid case-load appears to lie somewhere between the ONS estimate at around 10%, and some research cohort studies that tend to pitch it considerably higher. The ONS data also indicates that around 79,000 of UK sufferers are under the age of 19. Most of the symptoms described by sufferers at https://www.longcovidkids.org/ are reminiscent of those reported by adults: fatigue, shortness of breath, joint pain, rashes, headaches.
Given the strong imperative to get children back into full-time, face to face teaching after the disruptions of the past year, the key challenge is to maximally offset the risks of community transmission and paediatric cases of covid-19 and long covid. Analysis has shown that the protective impact of mitigation measures are cumulative as one adds in mask-wearing, ventilation, regular testing, small class sizes and spaced classrooms. The next few weeks will pose some enormous logistical challenges for heads and teachers. Luckily this is happening as we head into Spring, with great potential to make use of outdoor teaching and dining. After that, it will be time to start planning for extension of vaccination programmes into schools.
Daniel M Altmann is professor of Immunology at the Department of Immunology and Inflammation, Hammersmith Hospital, Imperial College, London.
Rosemary J Boyton is professor of Immunology and Respiratory Medicine, Consultant Respiratory Physician, Department of Infectious Disease, Faculty of Medicine, Imperial College London, UK and Lung Division, Royal Brompton & Harefield NHS Foundation Trust
The authors declare no competing interests
- Viner RM, Mytton, OT, Bonell C et al., Susceptibility to SARS-CoV-2 infection among children and adolescents compared with adults. A systematic review and meta-analysis. JAMA Pediatr (2021), 175, 143-156
- ONS: Coronavirus (COVID-19) Infection Survey pilot: England and Wales, 11 September 2020. https://www.ons.gov.uk/releases/coronaviruscovid19infectionsurveypilotenglandandwales11september2020
- Yonker LM, Neilan, AM, Bartsch, Y et al., Pediatric severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2): clinical presentation infectivity and immune responses. J. Pediatr 2020. 227, 45-52
- Yung CF, Kam KQ, Nadua KD et al., Novel coronavirus 2019 transmission risk in educational settings. Clin Inf Dis 2020 ciaa794, https://doi.org/10.1093/cid/ciaa794
- Vlachos J, Hertegård E, B Svaleryd H. The effects of school closures on SARS-CoV-2 among parents and teachers. Proc Natl Acad Sci U S A. 2021. 118(9):e2020834118.
- Li Y, Campbell H, Kulkarni D et al., The temporal association of introducing and lifting non-pharmaceutical interventions with the time-varying reproduction number (R) of SARS-CoV-2: a modelling study across 131 countries. Lancet Inf Dis 2020. 21, 193-212…