Indirect effects of the covid-19 pandemic on childhood infection in England: population based observational study

Principal findings

Since the onset of the covid-19 pandemic, there have been major reductions in hospital admissions for respiratory, severe invasive, and vaccine preventable infections in children in England. These reductions were seen in all demographic subgroups, including in children with pre-existing comorbidities who are more likely to be admitted to hospital with severe disease, require prolonged intensive care support, and be at greatest risk from dying of infection.

The findings indicate the extent to which measures related to the control of covid-19 can also impact on the transmission of other specific infectious diseases in childhood. Before the pandemic, bronchiolitis accounted for 12% of all admissions to paediatric intensive care units and had a case fatality rate of 1.75% in children aged <12 months and 4.4% in children aged ≥12 months.22 Exacerbation of asthma caused by acute respiratory infections was also one of the commonest reasons for hospital admission in children. In this study, after 1 March 2020, we report substantial reductions in admissions for bronchiolitis (80%), viral wheeze (56%), and predominantly bacterial causes of childhood respiratory infection: 60% reduction in the overall number of children admitted with pneumonia, 74% reduction for otitis media, and 66% reduction for tonsillitis. Children with haematological malignancies and cystic fibrosis who are most susceptible to major bacterial infections also had lower incidence of admission for respiratory infections compared with previous years (supplementary fig 2). Children with respiratory infections account for high healthcare costs, substantial morbidity, and in rare instances, mortality.

Contributing causes and implications for the future

While the pattern of hospital recorded covid-19 is probably almost entirely driven by universal testing for the infection at the point of admission for any medical or surgical condition, the dramatic reduction in transmission rates of non-covid-19 infectious pathogens among children is probably because of changes in patterns of social contact (social distancing, school closures, and minimal recreational activities) and shielding measures (at risk groups). Vastly restricted domestic and international travel (reduced air pollution) might have also contributed to the reduction in childhood respiratory infections.1623 However, these measures are generally unsustainable outside of the pandemic because of their inherent negative financial and societal impact.

The immediate and substantial reduction in respiratory and severe invasive childhood infections is also probably caused by the widespread introduction of non-pharmacological interventions. A range of behavioural measures, including face masks, better and regular hand hygiene, improved respiratory etiquette, and environmental cleaning have been adopted across homes, schools, and workplaces in England, which helped in reducing transmission of SARS-CoV-2.24 Our data indicate that these interventions might have also minimised childhood transmission of non-covid-19 infections. In particular, this study has highlighted that children with severe underlying comorbidities, including extremely preterm infants, those with congenital cardiac disease, and those with asthma have been protected from severe and potentially life threatening infection.

School closures might have contributed to some of the reductions seen in this study, but it was not possible to disentangle the effects of school closures and other policies around physical distancing. School closures during influenza pandemics have been shown to reduce epidemic peaks, but conflicting evidence exists about optimal timing and duration. The success of school closures is largely dependent on the characteristics of the circulating influenza serotype and when transmission is higher in children than adults.2425 National surveillance data in 11 966 children from 131 schools in England showed extremely low rates of SARS-CoV-2 infection and transmission in children aged up to 11 years at the point of school closure and also at reopening.26 During the covid-19 pandemic, good hygiene practices have been implemented (hand sanitisers in classrooms, regular hand washing, education on respiratory etiquette) along with measures to minimise spread of infection (avoiding overcrowding and good ventilation in classrooms). Furthermore, children with respiratory symptoms were excluded until confirmed to have a negative covid-19 test. Ensuring children with symptoms are off school, and encouraging employers to support parents to be at home with their unwell child, could help reduce the burden of childhood infections transmitted at school. However, a systematic review that included 72 observational studies assessed the impact of school closures on the mental health of children during covid-19 and showed increased anxiety, depressive and emotional behaviours.27 Preliminary observational data from France suggest there had been no increase in the rates of common respiratory infections among children immediately after the reopening of schools after national lockdown measures were lifted15; this could be because of societal behaviours persisting immediately after schools reopened and unwell children being kept home. Further analysis is required to assess which measures might be maintained in future to reduce transmission of infections more broadly while avoiding the negative impact of closing schools. Ongoing clinical and microbiological surveillance is required to evaluate any epidemiological changes (eg, substantial increase in rates, delayed peaks, disease in older infants) in respiratory infections that might have occurred because of a lack of population based immunity during the pandemic. The unusually high rates of croup and upper respiratory tract infections in the last months of this study (in May and June 2021) further underline this need.

The reduction in musculoskeletal infections could be because of decreased exposure to Staphylococcus aureus and Kingella kingae, which are the commonest causative pathogens. The larger reductions seen in hospital admission due to cellulitis, compared with septic arthritis or osteomyelitis, might be because these children were treated with oral antibiotics and could be more easily managed at home compared with those with bone and joint infection. The epidemiology of bone and joint infections in children is poorly understood because of a lack of clinical or microbiological surveillance. These data could encourage future studies to evaluate the cause (eg, contribution of preceding respiratory infections, trauma, or underlying comorbidities) to better inform management strategies.

Our data also show important reductions in respiratory infections among children with pre-existing lung disease. To obtain a holistic overview of protective mechanisms that have reduced hospital admissions for respiratory infections in these patients, there is a need for more quantitative data (changes in patterns to elective admissions, exposure to respiratory pathogens, colonisation with resistant pathogens) and qualitative data (from children and their caregivers).

The absence of any reduction in admissions for pyelonephritis might be because non-pharmacological interventions and social restrictions have no impact on this condition; the slight percentage increase (4%, 95% confidence interval −8% to 18%), though not statistically significant, could be caused by delays in presentation for children who had uncomplicated urinary tract infections that might have progressed to ascending infection by the time of hospital attendance.

Reductions in bacterial pathogens during the covid-19 pandemic, which have been reported in laboratory confirmed data from 26 countries worldwide,28 might be caused by fewer social interactions limiting the transmission of vaccine preventable infections.2930 However, these reductions might also be partly caused by reduced circulating respiratory viral infections (in particular, respiratory syncytial virus and influenza), which can increase susceptibility to pneumococcal and meningococcal disease. In regions with interrupted vaccination programmes, active enhanced surveillance mechanisms and catch up vaccination programmes should be prioritised to identify any resurgence in vaccine preventable infections and ensure herd immunity is maintained.

Reduced hospital attendance could be partly due to reluctance to attend the emergency department during a pandemic and potentially being exposed to SARS-CoV-2. Reassuringly, this study indicates that the pandemic has not had an adverse effect on absolute numbers of fatalities because deaths within 60 days have also reduced alongside hospital admissions (investigation of death rates from infections among children who did not have a related hospital admission was outside the scope of this study). However, for pneumonia, although the absolute number of 60 day fatalities decreased, the proportion admitted who died within 60 days increased. This increase could be because hospital admissions for pneumonia after 1 March 2020 were generally at the more severe end of the spectrum if people with less severe disease were being more frequently managed outside hospital or because of delays in seeking or difficulty in accessing appropriate healthcare during the pandemic. At the Royal Children’s Hospital in Melbourne, the Hospital in The Home service observed that a dedicated outpatient team can safely manage children with meningitis, severe cellulitis, and complex urinary tract infections in their home.313233 This service has been shown to deliver cost savings to the healthcare provider and improves quality of life for the child and caregiver. Future work should include caregiver perceptions in seeking healthcare during the pandemic to minimise anxiety, ensure clear public health messaging that encourages caregivers to attend emergency departments when necessary, and develop outpatient services to manage children with infection in the community if appropriate.

Strengths and limitations

The strengths of this study include the national coverage of data available because they capture all relevant hospital admissions for the entire child population of England over several years; determination of underlying comorbidities based on patients’ entire previous NHS hospital admission history; and the inclusion of a range of different infections providing the means to compare them. This study has some limitations. The reliability of HES data is dependent on information collected from clinical notes and clinical coding practice, which is important when comparing data from one year to the next because the quality might vary. Beyond the clinical diagnoses (ICD-10 codes), no other clinical or microbiological data were available. Therefore, disease severity (such as admission to paediatric intensive care unit) could not be evaluated and ICD codes could not be corroborated with laboratory results. The HES data represent hospital admissions for infection and do not necessarily provide a complete picture of the true rates of actual infection in the population as they do not cover patients who did not attend an NHS hospital (eg, owing to patient or caregiver reluctance or possible shortage of healthcare provision) or those who only attended the emergency department or outpatient settings (that is, they did not require inpatient admission). For comorbidity data, although we were able to look back historically at each patient’s entire NHS hospital admission history, comorbidity profiles might be incomplete because the data did not cover emergency department or outpatient settings (only inpatient admissions and day cases), primary care, or non-NHS care.

Comparison with other studies

This large epidemiological study evaluates the impact of covid-19 on hospital admission rates for childhood infections. Similar findings have been reported in regional studies in Australia, Denmark, France, and the Netherlands, with reported reductions of more than 50% in paediatric attendances with common infectious diseases.343536 However, it would be valuable to establish similar continuous analyses in other locations where centralised electronic health record data are available (eg, China, Taiwan, or Sweden, or in health maintenance organisations in the United States) to monitor such trends until the challenges associated with the pandemic diminish.

Conclusions

This study has shown dramatic overall reductions in hospital admissions for respiratory, severe invasive, and vaccine preventable infections in children during the covid-19 pandemic in England. Children with potentially life threatening comorbidities were also substantially protected. Further evaluation of non-pharmacological interventions that could be sustained beyond the pandemic is required to inform policy makers about potential strategies, especially during winter months, to minimise the burden on health systems and protect vulnerable children. Continued monitoring of hospital admissions for these infections is required as social restrictions evolve.