Modelling shows we need a total lockdown now — but it only needs to last a month 

Australia still has a chance of flattening the curve of coronavirus cases, and we only have 10 days to act. 

A draft report by adjunct professor Michael Mike Georgeff, a mathematician and leading Australian expert on artificial intelligence and health technology, shows we need to go hard now, placing 20 million Australians into lockdown by April 7 to keep coronavirus cases under 200,000. 

The report’s predictions come from the Susceptible, Infected, Recovered epidemiological model, mapping worldwide trends and responses to Australian cases. Georgeff and his team fed various assumptions into the model, looking at data gathered from China, the US, Italy, and Imperial College London.  

“We wanted to find out what was happening because we were not getting enough information from the government,” he told Crikey. 

What they found could mean life or death for the country.  

If cases climb above 100,000: Australia’s hospitals will be overwhelmed, and the mortality rate will jump from 1% to 10%. 

Australia has around 2,500 ICU beds, and 1,500 with ventilators (though this number is increasing). About 4.4% of coronavirus patients require hospitalization, and 30% of these cases require ICU beds for an average of 10 days. 

If peak cases stay below 100,000 at any given time, the mortality rate will stay at 1%. As is the case in Italy, if hospitals are swamped the number will jump to ten times that amount.

If we just rely on social distancing: If we keep cafes and restaurants closed and shut schools, the rate of infection would be reduced by 20%, though the timing of the peak wouldn’t change.

If implemented by the first of April, peak cases would climb to just under 4 million. 

The model assumes each infected person infects another every 2.6 days; Australia’s growth rate is currently doubling every three days. 

If testing and isolation can happen in under two days: Australia’s infectious period is about six and a half days — that is, from the time someone is infected with coronavirus, to the time they self-isolate. 

If as soon as someone suspected they may have coronavirus, they were able to get tested, receive results and isolate within two and half days — bringing the average time to four and a half days, counting people who are asymptomatic or don’t get tested — by April first, cases would climb to over two million. 

“If we could get tests immediately then you might not need isolation, but that’s not possible,” Georgeff said.

If we tested fast and implemented social distancing: Combining the two above strategies would limit peak infection loads to less than one million. 

If we go into lockdown by mid April: If we went into a month-long lockdown from April 19 to May 16, we’d see a reduced peak caseload to around three million people.

Around four million would never be infected and the second wave of the virus, once people are reintroduced into the community, would be smaller than the first. 

If we implemented fast testing, social distancing and mid-April lockdown: A combination of all three measures implemented before April 10 would see the peak number of infected people drop to under 700,000, with ten million people not being infected.

Importantly, action after April 10 would add millions to the peak infection load. 

If we go hard and fast, now: The report recommends a “hammer and dance” approach: extreme measures now, and then tweaking with re-introduction measures. 

“You hit it really hard then play with numbers as you slowly release it back,” Georgeff said. For this to work, people over 70 would need to go into lockdown immediately; 20 million people (most of Australia’s population) would need to be quarantined by the April 7. 

In line with other models, 80% of Australians would have to be in lockdown for it to work.

Lockdown wouldn’t last very long either — between one to two months with the appropriate testing measures in place. Case numbers would still be higher than our health system can cope at 200,000 people (reduced by putting more people in quarantine earlier), with four million Australians affected. The death toll would likely be less than 40,000. 

Professor Tony Blakley, an epidemiologist at the University of Melbourne told Crikey his modelling had produced similar results. 

“Keeping infections beneath 200,000 a day seems plausible. In my very basic model I made it 125,000 minimum,” he said. 

Whatever happens, don’t go out between April 16 to May 13 — this is when the probability of infection is the highest.  

Action is needed now, Georgeff says. “It’s all over in a month or two if we act immediately,” he said. “We have to start giving a consistent message to the public.”