, “There are a lot more people with COVID-19 than there are confirmed cases,” said Douglas Isles, a member of the Actuaries Institute’s COVID-19 working group. Actuaries are mathematicians and statisticians who often work in insurance, forecasting customer risks.
“The reported data is an order of magnitude less than the actual data.”
The reported data is an order of magnitude less than the actual data.
Douglas Isles, Actuaries InstituteThe official government modelling, led by the Melbourne-based Doherty Institute, said Australia’s testing system is picking up 93 per cent of all symptomatic cases of COVID-19.
This is despite Australia imposing tight criteria on who can be tested, restricting it to those who have been overseas or have had known contact with someone with COVID-19. These restrictions have recently been relaxed.
The calculation behind the “case detection rate” is based on another number called the “case fatality rate” – the number of people who are infected for every person who dies.
The modelling assumes Australia will have the same case fatality rate as China: on average 138 people will die for every 10,000 who are infected.
“This is a brave assumption, and unlikely to be true,” said Professor John Mathews, an epidemiologist at the University of Melbourne and a former senior adviser to the federal government.
That figure is not adjusted for Australia, where people may be less likely to die from the disease. Australia’s healthcare system was not overwhelmed as China’s was, and our population demographics are different.
Professor Mathews and Professor Catherine Bennett, chair of epidemiology at Deakin University, said our true case fatality rate was likely far lower.
That would mean the number of people infected is far higher than 6623. It would also mean our testing is capturing far fewer than 93 per cent of those who are ill.
“We may underestimate how much virus is still circulating, and how much natural immunity there may be, and so government may not get the timing right for the relaxation of restrictions,” Professor Bennett said.
We risked “relaxing too quickly if there is more virus circulating than we can currently see,” she said.
The University of Melbourne’s Professor James McCaw, a member of the Doherty modelling team, agreed there was a reasonable chance Australia’s case fatality rate was different from China’s, which would affect the estimates.
“Particularly in uncertain times such as now, we must interpret the numbers cautiously, and not simply assume they must be ‘right’.”
However, he stood behind the robustness of the modelling.
“National testing guidelines have always allowed for clinician discretion, so testing beyond the case definition has been occurring throughout the epidemic.
“Australia’s extensive contract tracing system combined with the natural advantages from having a clearly defined border have evidently performed well.”
Sign up to our Coronavirus Update newsletterGet our Coronavirus Update newsletter for the day’s crucial developments at a glance, the numbers you need to know and what our readers are saying. Sign up to The Sydney Morning Herald’s newsletter here and The Age’s here.
Most Viewed in NationalLoading
, http://www.smh.com.au/national/official-covid19-figures-underestimate-spread-by-order-of-magnitude-20200421-p54lnm.html, The Sydney Morning Herald