Covid Modelling Aotearoa is standing by its latest projections of an Omicron outbreak leading to thousands of hospitalisations every day.
That is despite the modelling being completed before the highly transmissible variant was discovered in the community in January.
The modelling was released today, with a scenario of up 800 hospitalisations a day - even with 90 percent of people boosted.
A worst case scenario could see the hospitalisation rate jump to more than 3000 cases a day.
"There's a lot of uncertainty in modelling... we'll always put out a range of scenarios, and of course there's things the public and the government can do to change those scenarios," Professor Shaun Hendy told Checkpoint.
"We're trying to give the range of possibilities that could occur, and then people by their actions will determine what path we take.
"This is modelling that was used to help the government prepare for what was expected to be an Omicron outbreak - we didn't know how long we'd be able to hold it at the border."
The modelling unit, Te Pūnaha Matatini, is hosted by Auckland University and funded by the Department of the Prime Minister and Cabinet.
"We always make a policy of putting out our reports that we've given to government, so that people can see what decisions were being made on," Hendy said.
He said most people understand that modelling shows what could happen depending on a set of underlying influences, but that it's always subject to change as the situation changes and the many different underlying variables change.
"I think it still is quite relevant - we looked at scenarios with an outbreak starting 1 January, we looked at an outbreak starting 1 February - those two scenarios do give us an indication of what we're looking at, at the moment.
"So you can get a sense of what we might see from the scenarios in [our] paper that's been released today. We'll update that of course as those case numbers come in and we get more of an idea of what kind of trajectory we're on - we update that modelling
"But I think that people can still use those numbers to get a sense of what might be coming over the coming weeks and months."
Hendy said the modelling made use of the best programmes developed internationally to predict possible outbreak scenarios, but they aren't perfectly tailored to those countries that have taken a less typical path, such as New Zealand's elimination strategy.
"We're still having control at our borders - those kind of details weren't taken into account in that modelling."
Another point that is useful when looking at the modelling paper is that some terms used within the modelling field may be different to how they are used in everyday language, Hendy said.
"[People] are used to talking about the cases reported on a daily basis, and some of the numbers that were reported [in the modelling] were actually total infections - so also include the infections that we never learn about because people don't get a test."
He doesn't believe most people will see models that show worst case scenarios as 'crying wolf.
"Most people do understand ... we're not actually trying to give a weather forecast, we're trying to give people an idea about the risks they face and how their individual decisions and the government's decisions might change those risks.
"The fact that [people take action based on the risks] actually changes the outcomes - the trajectory we take really depends on the decisions we all make."
Hendy said the modelling released in the paper takes into account the red light Covid-19 framework setting and the booster programme, but it was too early for it to reflect the effect from the government's move to reduce the time between booster shots to three months.
"Over the next few weeks we should begin to see numbers of people who get boosted rise, so that will have a big impact, and will tend to push us toward some of our more optimistic scenarios."