New recoveries have topped new cases of the disease for a week, meaning the effective reproduction rate of the virus has fallen below the critical level of one.

The number of new COVID-19 cases rose by only 34, or 0.5 per cent, to 6444, as of 5.30pm AEST. Modest increases were recorded in NSW, up 16 to 2886; Victoria, up eight to 1299; and Queensland, up five to 999.
This data shows more evidence that the rate of growth of the infection continues to flatten.
Professor McBryde said that when recoveries exceed new cases, it indicates the reproductive or R number is under one, and that one infected person is infecting less than one other person, on average.
But there are two R numbers in infectious diseases: R nought and R effective, known as the “Reff”.
Deputy Chief Medical Officer Nick Coatsworth. Alex Ellinghausen
The nought describes a virus infectivity with no intervention and no herd immunity. The effective rate describes it with intervention and herd immunity.
Professor McBryde expected the R effective threshold of under one to hold, and allow the epidemic to decline, but said at least some interventions would have to remain in place and this would require experimentation.
Deputy Chief Medical Officer Nick Coatsworth said work on Australia’s basic effective reproduction rate was still being completed but was likely below one.
He said the R nought that was used in initial modelling was about 2.6 or 2.7.
While several models have missed the mark on COVID-19 in Australia, one from Sydney University appears to be on track.
Professor Mikhail Prokopenko, of Sydney University. Supplied
Led by Professor Mikhail Prokopenko, a leading COVID-19 modelling academic, it correctly predicted prevalence (number of active cases) would peak in early April.
While the rise of this disease was rapid, the decline will be slow and steady. It could take up to three months to trickle down,” he said.
His model predicted it would reach a total of 8000 to 10,000 cases, on the assumption that 80 to 90 per cent of people practise social distancing.
If we stay at 90 per cent, by July we should get local transmissions to almost zero new cases, which means the risk of resurgence will be smaller,” said Professor Prokopenko, who heads the universitys Inter-faculty Centre for Complex Systems.
But if Australia lifts suppression before cases hit zero, without improving testing and tracing, cases will rise again. If it is improved, we could get there sooner.
Professor Sally Cripps, professor of statistics at Sydney University, said at the maximum, the percentage of recovered cases can only ever be 100 per cent minus the death rate.
On current figures this would appear to be about 98.3 per cent, she said.