There is a 95 per cent probability that Covid-19 has been eliminated in New Zealand, new modelling by researchers has found.
"It's not the absolute nail in the coffin for Covid," said Professor Shaun Hendy, director of centre of research excellence Te Pūnaha Matatini.
"But it should give people a bit more confidence," he said. That includes members of Cabinet, who are deciding whether New Zealand moves to alert level 1 on Monday.
Hendy said the 95 per cent probability was a pessimistic analysis of New Zealand's situation, which assumed relatively low rates of detection, reporting and surveillance.
"Over the next week, if we don't see cases, even the pessimistic scenario is going to climb to 100 per cent," he said.
Under a more optimistic scenario, with high detection and reporting rates of Covid cases and moderately effective restrictions, the 95 per cent probability of elimination would have been reached after 10 days of no reported cases.
The Ministry of Health reported this afternoon that New Zealand was now up to 15 consecutive days of no new cases.
Modelling only provides one part of the picture, and it can change quickly if a new case is confirmed - especially if the transmission source is not known.
Hendy noted new Covid-19 cases were potentially being introduced at New Zealand's border.
About 1000 people were now arriving a week in this country, and based on infection rates overseas it could be assumed that one person per week could have the virus.
"The important thing is that we don't let our guard down," Hendy said. "Because there is still the possibility of the disease coming in through the border.
"So we will certainly be celebrating once we are 100 per cent sure in the simulations that we've eliminated, but there's still that risk while it's out in the rest of the world."
— Te Pūnaha Matatini (@PunahaMatatini) June 5, 2020
The elimination estimate is based on computer simulations of how Covid-19 could spread in New Zealand.
"We run it thousands of times, because there is an element of random chance in the simulation. And then we look at how many of those times we've seen case numbers like we see in New Zealand, the runs of zeroes and the odd case popping up here and there.
"And then from those we look at how many eventually lead to elimination. So it's about running a simulation to see how many actually match what we've seen so far in New Zealand, then how many of those go on to have no more cases."
The modelling was useful for measuring potential transmission by infected people with subtle or no symptoms.
"They are very hard to detect, so you've got to really look at simulations to understand how likely it is that those people are out there.