Unknowns of Severe Acute Respiratory Infection (SARI) in Australian Children: Enhancing understanding of epidemiology to inform disease prevention

Abstract

My doctoral research aimed to enhance understanding of the virology and clinical epidemiology of severe acute respiratory infection (SARI) in Australian children to inform disease prevention strategies. I first studied respiratory syncytial virus (RSV), a leading cause of acute lower respiratory infection in young children. Specifically, I aimed to estimate the incidence of respiratory syncytial virus (RSV) hospitalisations in the Australian population, examine the circulation and associated disease severity of RSV subtype infections, determine whether RSV infection contributes to deaths in Australian children, and determine if RSV infection can cause severe acute neurological disease. RSV served as an exemplar for other understudied viral SARI and I subsequently established a project designed to enhance understanding of SARI epidemiology in Australian children more broadly. To estimate the incidence of hospitalisation in the Australian population, I led a retrospective epidemiological review of National Hospital Morbidity Database data for all RSV-associated hospitalisations in Australia from 2006 to 2015. In this study, 63,814 RSV-associated hospitalisations were identified in the Australian population. Of these, 94.9% were in children aged less than five years and the highest hospitalisation rate was for infants aged up to two months (2778 per 100,000 population). Hospitalisation rates were higher for adults aged 65 years and over than for people aged 5–64 years (incidence rate ratio [IRR] 6.6; 95% confidence interval [CI] 6.2 to 7.1) and were also higher for Indigenous Australians than for other Australians (IRR 3.3; 95% CI 3.2 to 3.5). RSV subtype circulation and disease severity were examined in a single-centre, retrospective, observational study of children aged less than 16 years from 2014 to 2018. This study described subtype circulation of all RSV-A and RSV-B infections in children presenting to the study hospital. A random sample of children with RSV-A and RSV-B infections was selected and compared with respect to demographics, clinical features and clinical severity. We identified 3591 RSV subtype infections and found consistent co-circulation of subtypes with alternating predominance of either subtype. Demographic and clinical characteristics were similar between children presenting with either subtype. This study found no clear difference in disease severity between subtype infections. The features and frequency of RSV-attributable deaths were assessed in a single-centre, retrospective, observational study in children aged less than 16 years over a 21 year period (1998–2018). All RSV-associated deaths were identified, reviewed and classified according to RSV contribution to death. A total of 20 RSV-attributable deaths were identified. The case fatality rate among hospitalised cases was 0.2% (20/9779), and the population RSV mortality rate estimated at 1.2 (95% CI 0.5 to 2.7) per million children aged less than 16 years in New South Wales. All children had at least one medical comorbidity, and over half the deaths occurred in children aged two years and over (11/20, 55%). Healthcare-associated RSV infection was common (11/20, 55%). RSV-associated severe acute neurological disease was examined through a systematic review and aggregated case series conducted in accordance with the PRISMA framework and registered on PROSPERO (CRD42019125722). A total of 87 unique studies from 26 countries were identified and described a spectrum of RSV-associated severe acute neurological syndromes including proven encephalitis, acute encephalopathy, complex seizures, hyponatraemic seizures and immune-mediated disorders. Aggregation of data from 155 individual cases with RSV-associated severe acute neurological disease highlighted that the median age of cases was 11.0 months (interquartile range 2.0–21.5) and the majority were previously healthy (71/104, 68%). Most children recovered (81/122, 66%); however, some reports described partial recovery (33/122, 27%) and death (8/122, 7%). Twelve cases had RSV detected in a central nervous system sample providing strong evidence of causality. The epidemiology of SARI in Australian children will be explored using two, large, routinely collected datasets; NSW respiratory syndromic data (1,237,072 individual presentations) and sentinel site laboratory data including 16 viral respiratory pathogens (29,890 virus specific detections). The virology and clinical epidemiology of syndromes and pathogens will be described. The associations between syndromes and pathogens will be examined. The utility of these data to be aggregated for developing a comprehensive, timely and robust SARI surveillance system will be assessed. Ethics and governance approvals for the project have been obtained, and the data has been released. Unpublished descriptive summary statistics are presented in this thesis. My research on severe RSV disease has made a substantial and novel contribution to understanding RSV epidemiology and will inform the design of future disease prevention and surveillance. My established project examining SARI in Australian children will enhance understanding of the epidemiology of a variety of childhood respiratory syndromes and viruses. This will provide novel data to inform future research relating to disease prevention and surveillance of SARI more broadly.