How Victoria, Australia gained control of its second COVID-19 wave

  

View ORCID ProfileJames M Trauer, Michael J Lydeamore, Gregory W Dalton, David Pilcher, Michael T Meehan, Emma S McBryde, Allen C Cheng, Brett Sutton,  View ORCID ProfileRomain Ragonnet

doi: https://doi.org/10.1101/2021.04.03.21254866

This article is a preprint and has not been peer-reviewed [what does this mean?]. It reports new medical research that has yet to be evaluated and so should not be used to guide clinical practice.

PDF: https://www.medrxiv.org/content/10.1101/2021.04.03.21254866v1.full.pdf

Abstract

Victoria has been Australia’s hardest hit state by the COVID-19 pandemic, but was successful in reversing its second wave of infections through aggressive policy interventions. The clear reversal in the epidemic trajectory combined with information on the timing and geographical scope of policy interventions offers the opportunity to estimate the relative contribution of each change. We developed a compartmental model of the COVID-19 epidemic in Victoria that incorporated age and geographical structure, and calibrated it to data on case notifications, deaths and health service needs according to the administrative divisions of Victoria’s healthcare, termed clusters. We achieved a good fit to epidemiological indicators, at both the state level and for individual clusters, through a combination of time-varying processes that included changes to case detection rates, population mobility, school closures, seasonal forcing, physical distancing and use of face coverings. Estimates of the risk of hospitalisation and death among persons with disease that were needed to achieve this close fit were markedly higher than international estimates, likely reflecting the concentration of the epidemic in groups at particular risk of adverse outcomes, such as residential facilities. Otherwise, most fitted parameters were consistent with the existing literature on COVID-19 epidemiology and outcomes. We estimated a significant effect for each of the calibrated time-varying processes on reducing the risk of transmission per contact, with broad estimates of the reduction in transmission risk attributable to seasonal forcing (27.8%, 95% credible interval [95%CI] 9.26-44.7% for mid-summer compared to mid-winter), but narrower estimates for the individual-level effect of physical distancing of 12.5% (95%CI 5.69-27.9%) and of face coverings of 39.1% (95%CI 31.3-45.8%). That the multi-factorial public health interventions and mobility restrictions led to the dramatic reversal in the epidemic trajectory is supported by our model results, with the mandatory face coverings likely to have been particularly important.

Competing Interest Statement

BS and ACC wish to emphasise their important statutory roles during Victoria's pandemic response in 2020, as Chief Health Office and acting Chief Health Officer respectively. MJL and GWD were also employed by DHHS during 2020. JMT provided regular advice to DHHS during this time as an independent advisor.

Funding Statement

The Epidemiological Modelling Unit of Monash University provided the health system cluster-level projections for notifications, admissions and deaths under contract to the Victorian Department of Health and Human Services in 2020. JMT is a recipient of an Early Career Fellowship from the Australian National Health and Medical Research Council (APP1142638).

https://www.medrxiv.org/content/10.1101/2021.04.03.21254866v1