Educational Aims
The reader will gain an improved understanding of:
- •
The global burden of acute respiratory infections in children.
- •
The direct impacts of non-pharmaceutical interventions (NPI) used in the COVID-19 pandemic on the epidemiology of paediatric respiratory infections.
- •
The probable underlying mechanisms for the altered epidemiology of paediatric respiratory infections.
- •
The indirect impacts of the pandemic on the burden of paediatric respiratory infections now and into the future.
- •
Post-pandemic health system priorities for addressing paediatric respiratory infections.
Abstract
Acute respiratory infections (ARI), especially lower respiratory infections (LRI), are a leading cause of childhood morbidity and mortality globally. Non-pharmaceutical interventions (NPI) employed during the COVID-19 pandemic have impacted on the epidemiology and burden of paediatric ARI, although accurately describing the full nature of the impact is challenging. For most ARI pathogens, a reduction was observed in the early phase of the pandemic, correlating with the most stringent NPI. In later phases of the pandemic resurgence of disease was observed as NPI eased. This pattern was most striking for seasonal viruses, such as influenza and respiratory syncytial virus. The impact on ARI-associated bacterial disease varied; marked reductions in invasive Streptococcus pneumoniae and Streptococcus pyogenes were observed, followed by a resurgence that correlated with increases in respiratory viral infections. For Corynebacterium diphtheriae , Bordetella pertussis , and Mycoplasma pneumonia e, a sustained reduction of disease was observed well into 2022 in most regions. Proposed mechanisms for the varied epidemiological disruption amongst ARI pathogens include differential effects of NPI on specific pathogens, population-level immunological effects, and ecological and genetic pathogen adaptations. Additionally, important indirect effects of pandemic restrictions on paediatric respiratory infections have been identified. These occurred as a result of disruptions to routine health services, reductions in vaccination coverage, and disruptions to respiratory infection research and surveillance activities. Impacts have been disproportionately borne by those in low resource settings. We discuss opportunities to leverage pandemic learnings to support improved understanding of the epidemiology of paediatric respiratory infections to inform future prevention and health system strengthening.
Introduction
Acute respiratory infections (ARI) are a leading cause of childhood morbidity and mortality, and in 2019, lower respiratory infections (LRI) were the second most important cause of disability adjusted life years in children younger than 10 years globally . There are a range of viral and bacterial pathogens that contribute to this burden, and these include respiratory syncytial virus (RSV), influenza, Streptococcus pneumoniae , Mycoplasma tuberculosis , and Bordetella pertussis The global burden of LRI associated with some of these pathogens has been characterised however there is considerable uncertainty of the estimates ( Table 1 ).
| Respiratory Pathogen | Annual number of LRI (95% uncertainty range [UR]) | Annual number of LRI hospitalisations (UR) | Annual number of in-hospital LRI deaths (UR) | Annual number of all LRI deaths (UR) | Year of estimate (Source) |
|---|---|---|---|---|---|
| Respiratory Syncytial Virus* | 33.0 million (25.4–44.6) | 3.6 million (2.9–4.6) | 26,300 (15,100–125,200) | 109,600 (97,200–124,900) | 2019 (Li et al. 2021) |
| Influenza* | 10.1 million (6.8–15.1) | 870,000 (543,000–1.4 million) | 15,300 (5800–43800) | 34,800 (13,200–97,200) | 2018 (Wang et al. 2020) |
| Parainfluenza Virus* | 26.1 million (17.8–40.1) | 1.0 million (601,000–1.8 million) | 25,700 (12,000–56,500) | 53,000 (25,300–113,500) | 2018 (Wang et al. 2021) |
| Human metapneumovirus* | 14.2 million (10.2–20.1) | 643,000 (425,000–977,000) | 7,700 (2,600–48,800) | 16,100 (5,700–88,000) | 2018 (Wang et al. 2021) |
| Streptococcus pneumoniae ^ # | 44.7 million (20.9–73.7) | – | – | 341,029 (195,289–493,551) | 2016 (Troeger et al. 2018) |
| Haemophilus influenzae type b^ # | 6.1 million (1.4–13.7) | – | – | 48,011 (13,404–88,744) | 2016 (Troeger et al. 2018) |
Obtaining accurate incidence and disease burden estimates for pathogen-specific paediatric respiratory infections is challenging due to geographic gaps in data, as well as differences in case definitions, diagnostic testing, and healthcare access . Low- and middle-income countries (LMIC) are frequently underrepresented in the data available . Several important respiratory infection syndromes, some with a paediatric focus, are recognised and employed for research and surveillance ( Table 2 ) although heterogeneity of methodologies across studies is a persistent challenge. Bronchiolitis is commonly used as a surrogate for RSV infection, however for this syndrome the use of variable case definitions is problematic . Employing consistent case definitions, such as the modified influenza focused surveillance definition ‘extended Severe Acute Respiratory Infection’ (SARI), can improve sensitivity to capture RSV cases and improve the accuracy of incidence and disease burden estimates .
| Term | Definition(s) |
|---|---|
| Acute respiratory infection (ARI) | Acute (onset withing past 10 days) AND respiratory infection (at least one of cough, sore throat, shortness of breath or runny nose) |
| Influenza-like illness (ILI) | Acute (onset withing past 10 days) AND respiratory infection AND fever |
| Extended ILI | As for ILI but removal of requirement for fever increases sensitivity to capture RSV infections |
| Severe acute respiratory infection (SARI) | Severe (overnight hospitalisation) AND acute (onset withing past 10 days) AND respiratory infection (cough or shortness of breath) AND fever |
| Extended SARI | As for SARI but removal of requirement for the presence of fever increases sensitivity to capture RSV infections. Additionally, for infants < 6 months of age, apnoea and sepsis is accepted in place of respiratory infection |
| Lower respiratory infection (LRI) | Broad definition that encompasses a range of clinical syndromes and infections, including pneumonia, bronchiolitis, respiratory syncytial virus, and ILI |
| Bronchiolitis | Clinical syndrome of LRI with respiratory distress in young children. Diagnostic criteria vary by region, including with respect to age which may be < 12 months or < 24 months |
| Pneumonia | Clinical syndrome of LRI with various case definitions recognised which may include the presence of specific clinical signs and symptoms, sometimes with defined time periods e.g., acute onset including 14 days or less) |
In late 2019, SARS-CoV-2 emerged as novel cause of SARI . Its recognition as a public health threat resulted in the introduction of unprecedented, population-wide non-pharmaceutical interventions (NPI) to curb the impact of the coronavirus disease (COVID-19) . To date COVID-19 has been associated with 6.95 million deaths globally (as of 12th July 2023) however,. COVID-19 is associated with a low risk of severe disease and mortality in children, and consequently there is considerable controversy regarding the benefits of shielding children from infection that may not be proportionate to the risks . Pandemic restrictions have had significant impacts on the epidemiology and burden of other paediatric respiratory infections and the future implications of this impact are not yet fully understood.
In this paper, we will discuss the direct impacts of the COVID-19 pandemic on the epidemiology of non-SARS-CoV-2 paediatric respiratory infections. We will then explore the possible mechanisms for the observed altered epidemiology. Lastly, we consider some of the indirect impacts of the COVID-19 pandemic on ARI, specifically arising from the impact on essential healthcare services, vaccination coverage, and other critical health system activities. Throughout this paper, we aim to identify gaps and health system priorities for addressing the ongoing burden of all paediatric respiratory infections in a post-pandemic world.
The impact of the COVID-19 pandemic on paediatric respiratory infection epidemiology
There was considerable variation in the stringency of NPI implementation during the COVID-19 pandemic. Interventions ranged from facemask use and enhanced hygiene, to border closures and quarantining . The strictest use of NPI consisted of stay-at-home orders, or ‘lockdowns’, which were utilised by as many as a quarter of all countries at any one time in from March 2020 to March 2021, the first year – early phase – of the pandemic . Population wide NPI have proven unsustainable for extended periods, largely because of population acceptability, and even the strictest NPI failed to completely prevent transmission of SARS-CoV-2, especially as more transmissible variants (e.g. Omicron) arose . Relatively early on following NPI implementation, observations of the considerable impact of NPI on other respiratory pathogens were published ( Table 3 ). Substantial heterogeneity across published studies is evident. Variable case definitions, pathogens, and time periods, coupled with differing NPI use, changes to healthcare-seeking behaviour and testing practices complicate interpretation. LMIC countries are notably underrepresented in the published literature.
| Respiratory pathogen and associated clinical syndromes # | Incidence of infections during the early phase* of the pandemic | Incidence of infections during the later phase^of the pandemic | Location: Study Date Range (Citation) |
|---|---|---|---|
| Respiratory syncytial virus Bronchiolitis | Reduced rates of RSV infection and disease during NPI use Near-complete suppression of usual seasonal activity in many regions | Increased rates of RSV infection and disease following easing of NPI Many studies described early or inter-seasonal epidemics Some evidence of larger epidemics compared to pre-pandemic years Evidence of reduced genetic diversity in Australia | Australia: Jan-16 to Dec-22 (NSW Ministry of Health 2023)§ Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ UK: Dec-14 to Mar-22 (Bardsley et al. 2023)§ China, Shenzhen: Jul-18 to Jan-22 (Wang et al. 2022)‡ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ New Zealand: Jan-15 to Jul-21 (Hatter et al 2021)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ Australia: Jan-17 to Mar-21 (Eden et al. 2022)§ France: Jan-07 to Mar-21 (Rybak et al. 2022)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Influenza Influenza-like-illness | Reduced Influenza infection during NPI use Near-complete suppression of usual seasonal activity in many regions | Reduced intensity of influenza epidemics with considerable regional variation Inter-seasonal epidemics observed in some countries Reduced incidence of influenza B detections in some regions, in particular complete absence of influenza B-Yamagata strain | Global: Jan-95 to Jul-23 (WHO 2023) Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ China, Shenzhen: Jul-18 to Jan-22 (Wang et al. 2022)‡ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ Asia: Jan-15 to Dec-21 (Davis, Mott & Olsen. 2022) Australia, VIC: Jan-15 to Dec-21 (Bhatt et al. 2022)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ France: Jan-07 to Mar-21 (Rybak et al. 2022)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Parainfluenza virus | Continued detection throughout periods of NPI use observed in Germany, Israel, and Shenzhen, China Reduced detections observed in Canada, USA, and China | Increased detections following easing of NPI Some evidence of larger epidemics compared to pre-pandemic years Some studies observed inter-seasonal increases in detections | Australia: Jan-16 to Dec-22 (NSW Ministry of Health 2023)§ Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ China, Shenzhen: Jul-18 to Jan-22 (Wang et al. 2022)‡ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Human metapneumovirus | Reduced detections during NPI use Near-complete suppression of usual seasonal activity in many regions Continued detection throughout periods of NPI use observed in Germany | Increased detections following easing of NPI Most studies demonstrate return to expected pre-pandemic levels Increased detection and evidence of reduced genetic diversity seen in Australia | Australia: Jan-16 to Dec-22 (NSW Ministry of Health 2023)§ Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ Australia, WA: Jan-17 to Dec-21 (Foley et al. 2022)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Endemic Coronaviruses | Reduced detections during NPI use Near-complete suppression observed in Germany, Canada, and USA Reduction of detections was observed in China however less pronounced compared to other regions | Increased detections following easing of NPI however considerable variability across regions Some studies observed inter-seasonal increases in detections (Germany and USA) Increased detections were observed in Canada but were below pre-pandemic levels Increased detections above pre-pandemic levels were observed in China | Germany, Munich: Jan-19 to Nov-22 (Maison et al.2023)§ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al.2023)§ USA: Oct-20 to May-21 (Olsen et al.; CDC2021)§ China: Jan-12 to Jan-21 (Li et al.2022)‡ |
| Adenovirus | Short period of reduced or continued detection throughout periods of NPI use | Most studies observed reduced detections compared to pre-pandemic levels Increased detections above pre-pandemic levels observed in Australia and Germany | Australia: Jan-16 to Dec-22 (NSW Ministry of Health 2023)§ Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ China, Shenzhen: Jul-18 to Jan-22 (Wang et al. 2022)‡ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Rhinovirus # | Short period of reduced or continued detection throughout periods of NPI use | Increased detections above pre-pandemic levels | Australia: Jan-16 to Dec-22 (NSW Ministry of Health 2023)§ Germany, Munich: Jan-19 to Nov-22 (Maison et al. 2023)§ USA: Oct-20 to May-21 (Olsen et al.; CDC 2021)§ USA: Dec-16 to Feb-21 (Rankin et al. 2023)§ China: Jan-12 to Jan-21 (Li et al. 2022)‡ |
| Measles | Reduced notifications during NPI use across all WHO regions in 2020 | Sustained reduction of notifications in many regions Except for WHO EMRO region where notifications increased in 2021 | Australia, VIC: Jan-15 to Dec-21 (Bhatt et al. 2022)§ Global: Jan-00 to Dec-21 (WHO 2023) Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ |
| S. pneumoniae Pneumococcus Invasive pneumococcal disease (IPD) | Reduced IPD during NPI use | Increased IPD observed in many studies and often ecologically associated with increased viral infections | Australia, NSW: Aug-21 to Jul-22 (Williams et al. 2023)§ Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ Canada, Quebec: Jan-13 to Jan-22 (Ouldali et al. 2023)§ Australia, VIC: Jan-15 to Dec-21 (Bhatt et al. 2022)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ France: Jan-07 to Mar-21 (Rybak et al. 2022)§ Global: Jan-18 to May-20 (Brueggemann et al. 2021) |
| Haemophilus influenzae | Reduced detections during NPI use | Increased detections returning to pre-pandemic levels by August 2020 in Israel Sustained reduction in 2020–21 in Japan compared to pre-pandemic years | Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ Global: Jan-18 to May-20 (Brueggemann et al. 2021) |
| Streptococcus pyogenes Invasive Group A Streptococcal disease (iGAS) | Reduced iGAS notifications during NPI use | Increased iGAS notifications following easing of NPI Increase in proportion of iGAS presenting as empyema | Spain: Jan-19 to Dec-22 (Cobo-Vázquez et al. 2023)§ USA, MN & CO: Jan-16 to Dec-22 (Barnes et a.l; CDC 2023)§ France, Reims: Jan-08 to Dec-22 (Lassoued et al. 2023)§ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ |
| Mycoplasma pneumoniae | Reduced detections during NPI use | Sustained reduction in detection and associated pneumonia presentations | Israel, Jerusalem: Jan-17 to Jun-22 (Oster et al. 2023)§ China, Shenzhen: Jul-18 to Jan-22 (Wang et al. 2022)‡ Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ |
| Bordetella pertussis Whooping cough | Reduced notifications during NPI use across all WHO regions in 2020 | Sustained reduction of notifications in many regions Except for WHO EMRO region, where notifications increased in 2021 | Australia, VIC: Jan-15 to Dec-21 (Bhatt et al. 2022)§ Global: Jan-00 to Dec-21 (WHO 2023) Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ |
| Corynebacterium diphtheriae Diphtheria | Reduced diphtheria notifications during NPI use across all WHO regions in 2020 | Sustained reduction of notifications Except for WHO EMRO and EURO regions where notifications increased in 2021 | Global: Jan-00 to Dec-21 (WHO 2023) |
| M. tuberculosis Tuberculosis | Reduced notifications during NPI use across all WHO regions in 2020 | Sustained reduction of notifications in WHO EURO and WPRO regions Increased notifications in other regions, in particular WHO AFRO and EMRO regions | Japan: Jan-15 to Sep-21 (Hirae et al. 2023)§ Global: Jan-15 to Dec-21 (WHO 2022) |
# Studies used laboratory/surveillance data in which testing for combined rhinovirus/enterovirus was used rather than an independent rhinovirus test; †Low-middle income country ‡Upper-middle income country §High income country (country income level sourced from World Bank 2021 development indicators) .
Impact on RSV infections
The first published reports of reduced paediatric respiratory infections during the COVID-19 pandemic came from Australia where the introduction of strict NPI, stay-at-home orders and border closures, in March 2020 coincided with the usual onset of the RSV season . One study examined several RSV-associated indicators and found that from April to June 2020, a usual peak RSV period, the observed mean frequency of RSV detections was 94.3% (standard error [SE] 22.8) lower than predicted compared to 2015–2019, and bronchiolitis admissions were 89.1% lower (SE 32.7) . The near complete suppression of the RSV season was soon reflected in reports from other regions in the southern hemisphere including in Brazil , New Zealand , and South Africa and subsequently the northern hemisphere for the 2020–2021 RSV season .
As NPI eased, a sharp resurgence of RSV disease indicators was observed in most regions with unseasonal (Spring/Summer) resurgence in many locations. Many studies described a shift in the average age of infection to older children , and impacts on the viral diversity have been described with reduced genetic diversity noted in one Australian study . Changes in the molecular and clinical phenotype of RSV during the pandemic highlight the importance of establishing robust RSV surveillance as these may have implications for future public health activities, such as vaccination.
Impact on influenza infections
Seasonal influenza epidemics typically follow the RSV season in temperate regions, and like RSV, a clear suppression of the influenza season occurred from 2020 in the southern hemisphere coinciding with the introduction of NPI ( Table 3 ) . Although most countries saw a slow return to pre-pandemic seasonality of influenza, several countries experienced inter-seasonal outbreaks upon relaxing of NPI . The genetic evolution of influenza viruses has important public health implications and robust molecular surveillance of influenza viruses occurs through the Global Influenza Surveillance and Response System. In 2022, when influenza re-emergence occurred, H3N2 was the dominant strain in most regions until it was replaced by drifted 2009-pandemic-strain-H1N1 in early 2023 . Influenza B has been notably low or absent in many regions with speculation around the absence of the B-Yamagata strain, however, large outbreaks of influenza B in eastern Europe in mid-2023 remain largely untyped .
Impact on other respiratory viral infections
There was significant heterogeneity in the observations of how NPI impacted the epidemiology of other non-influenza, non-RSV respiratory viruses during the COVID-19 pandemic ( Fig. 1 , Table 3 ). Human metapneumovirus and endemic coronaviruses appeared to behave similarly to RSV and influenza with clear suppression of detections early in the pandemic . Reports examining parainfluenza virus detections demonstrated variable suppression . Easing of NPI, resulted in resurgence of human metapneumovirus and parainfluenza virus, but had varying impact on endemic coronaviruses . Non-enveloped viruses, rhinovirus and adenovirus appeared less affected by NPI use and when decreases were observed, they rebounded more rapidly than seen with other viruses . Rhinovirus detections were observed to increase above historic levels upon NPI easing . A sustained reduction in adenovirus-associated pharyngoconjunctival fever was noted in Japan .
