Global burden of acute lower respiratory infection (ALRI) associated with influenza virus, human metapneumovirus, and human parainfluenza virus among children under five years
Item statusRestricted Access
Embargo end date27/06/2021
Introduction Acute lower respiratory infection (ALRI) is one of the leading causes of mortality in children under five years. Although childhood ALRI mortality has substantially reduced over the past 15 years, continued progress will in part depend on targeted prevention and treatment against important pathogens in future. Influenza virus (IFV), human metapneumovirus (hMPV), and human parainfluenza virus (hPIV) are three important viruses causing childhood ALRI. However, global burden estimates of hMPV-associated ALRI and hPIVassociated ALRI among young children are unavailable, and there are no licenced vaccines and approved antiviral treatment for the two viruses. Over the past 10 years, several studies have estimated the global burden of influenza virus in young children using different types of data and different models. The estimates varied between these studies, partly reflecting the differences in analytical models between studies. This thesis aimed to estimate the global and regional ALRI morbidity and mortality associated with the hMPV and hPIV, and to update the estimates for global and regional ALRI burden associated with IFV. Methods Systematic reviews were conducted to identify published data on IFVassociated, hMPV-associated, hPIV-associated ALRI burden among children under five years. Relevant data included laboratory-confirmed incidence rates, hospitalisation rates, proportion positives, and in-hospital case-fatality ratios (hCFRs). Additionally, Respiratory Virus Global Epidemiology Network contributed unpublished data by finer age groups from different geographic locations, especially from low- and lower middle-income countries experiencing high childhood ALRI burden. A modified Newcastle-Ottawa Scale was used to assess the risk of bias in included studies. Incidence rates, hospitalisation rates, proportion positives, and hCFRs of virus-associated ALRI were analysed using a generalized linear mixed model. The meta-estimates of incidence rates and hospitalisation rates were applied to United Nation 2018 population estimates to yield the number of cases and hospitalisations of virus-associated ALRI. The point estimates and uncertainty ranges were estimated using Monte Carlo simulation. The hospitalisations and hCFRs for virus-ALRI were combined to yield the estimates for in-hospital mortality. Analyses were stratified by three age groups (0-5 months, 6-11 months, and 12-59 months) and child mortality settings (low and high) where available. Data were also stratified by World Bank income regions and country development status. The overall virus-associated ALRI deaths (including both in-hospital and out-hospital deaths) were estimated using in-hospital mortality estimates and multiple types of data due to the differences in data availability for the three viruses. These data mainly included population-based childhood pneumonia deaths in defined catchment areas, care-seeking for child pneumonia, and the US influenza-associated paediatric in-hospital deaths and out-hospital deaths. Results Globally among children under five years, IFV was associated with 9.1 million (UR 6.4–13.2) ALRI cases, 854,000 (UR 514,000–1,450,000) hospitalisations, 27,400 (UR 10,600–100,000) ALRI deaths, accounting for 7% of ALRI cases, 5– 17% of ALRI hospitalisations, and 3% of ALRI deaths. hMPV was associated with 14.6 million (UR 10.5-21.0) ALRI cases, 643,000 (UR 425,000–977,000) ALRI hospitalisations, 16,100 (UR 5,700–88,000) ALRI deaths, accounting for 11% of ALRI cases, 4–13% of ALRI hospitalisations, 2% of ALRI deaths. hPIV was associated with 29.5 million (UR 19.2–46.7) hPIV–ALRI cases, 1.0 million (UR 0.6–1.8) ALRI hospitalisations, and 53,000 (UR 25,300–113,500) ALRI deaths, accounting for 21% of ALRI cases, 6–20% of ALRI hospitalisations, 7% of ALRI deaths. The three viruses shared several similarities in the burden distribution. For the three viruses, infants had higher hospitalisation rates than older children. About 45–61% of the virus–ALRI hospitalisations occurred among infants under one year old (varying by viruses). hCFRs varied by income regions, and children in low– and lower middle–income countries generally had the highest hCFRs. The differences in hCFR meta–estimates of IFV–ALRI and hPIV–ALRI between age groups was less obvious than hMPV–ALRI. For hMPV–ALRI, the hCFRs were much higher in young infants aged 0–5 months than older children. Conclusion These estimates show that the three viruses are associated with substantial burden in children under five years. Infants under one year old and children in low– and lower–middle income countries were disproportionately affected by severe infections associated with the three viruses. This thesis presented the new IFV–associated ALRI morbidity and mortality estimates in the era with the circulation of influenza A/H1N1pdm09, and the first global burden estimates of hMPV–ALRI and hPIV–ALRI in children under five years, by narrow age groups. These global and regional burden estimates should inform the development of targeted prevention and treatment and guide further health investment priorities and resource allocation. The IFV–associated burden estimates should provide new evidence for maternal and paediatric influenza immunisation and should inform future immunisation policy particularly in low– and middle–income countries as a national influenza immunisation programme has not been adopted in most low– and lower middle–income countries. Large data gaps exist, especially in the mortality of virus–ALRI. Continued efforts are needed to fill and address the data gaps to improve global burden estimates providing evidence for developing future prevention and treatment strategies against childhood ALRI.