The Covid protection framework is no more so what does that mean for Kiwis and what should we expect from the pandemic now? Video / NZ Herald
This Covid-19 wave might not peak until the end of this year, or it could come much sooner – and higher. Why can’t modelling tell us for sure? Science reporter Jamie Morton explains.
What are we looking at right now?
Not so much the steep crest of a clearly towering wave, but the foggy foot of a hill we don't yet know the height of.
We do know that cases began to track upwards from about the start of October – climbing from fewer than 1000 to around 3000 now – and that these represent just a fraction of what's spreading about our communities.
We also know that the make-up of Omicron subvariants helping drive transmission is changing, with detections of the emergent BA2.75 - nicknamed Centaurus – creeping up alongside a still-dominant BA.5, which fuelled our winter wave.
"Cases were decreasing from the BA.5 peak until the middle of September; since then, they have been increasing slowly, at a lower rate than the start of the previous two waves, with cases concentrated in older age groups," said Dr Samik Datta, of Covid-19 Modelling Aotearoa.
"In the coming weeks, we may continue to see a rise in cases due to BA2.75, which is fast-spreading and has been detected in a number of countries including India and the UK."
Right now, modellers are placing New Zealand's effective reproduction number – or how many others one infected person might pass the virus onto – at above one.
"We do not know whether the increasing cases are due to relaxation of Covid protection measures and changes in behaviour, or because of increased presence of variants that are better at evading our existing immunity, which means we do not know whether this increase will continue to ramp up or will start to drop off," modeller Dr Emily Harvey said.
Unlike in past waves, there isn't a single big driver to give us a clear idea of what's in front of us.
Cases are doubling every two weeks in some age groups, but halving in others – and while some former DHB regions like MidCentral and Southern are reporting higher rates, geographic trends aren't revealing much.
"Without an infection prevalence survey, we are unable to be sure whether this is driven by increased reporting or increased infections," Harvey said.
We did have another surveillance tool that confirmed the case rise was indeed real: wastewater sampling.
ESR's latest analysis, covering the two weeks up to October 16, showed virus levels had increased at 56 per cent of monitored sites, but fallen at 23 per cent.
"Wastewater data provides a useful measure of infections in the community at a national, regional, and catchment scale," modeller Dr Leighton Watson said.
"Unlike cases, the wastewater data is independent of changes in testing, which means it can be a more reliable indicator of the true number of infections."
In July, for example, this data revealed a larger spike than the case data, suggesting the true number of infections during the winter wave was indeed higher than reported, due to fewer people testing.
"This is currently being seen in Wellington, where the wastewater data is increasing at a faster rate than the cases," Watson said.
Wastewater surveillance was also proving incredibly useful for detecting new variants.
BQ1.1 - responsible for a wave of infections in Singapore and forecast to become the dominant variant in Europe over the next months – was picked up recently in Wellington, Auckland and Rotorua.
Why are subvariants so important?
The modelling group's co-lead, Dr Dion O'Neale, said his team was particularly interested in knowing the relative abundance of different variants among us.
"This can give us information about how much of the growth we are seeing is due to possible immune evasion of new variants," he explained.
"The rate at which new variants start to outcompete existing variants, such as BA.5, can help to tell us about how steep we might expect the current rise in cases to be.
"We can also use information about the variants circulating in Aotearoa to compare with what has been observed in other countries."
"This is because the impact of each variant depends on the immunity profile of the population it is spreading in," he said.
"At the moment there have been a number of different waves of different combinations of variants in different countries.
"This makes it tricky to work out the effect of immunity evasion of a new variant for different immunity profiles."
What about reinfections?
If we were seeing a big leap in reported reinfections, that might tell us this wave is being powered by waned immunity or tricky new subvariants on the loose.
Yet reinfections are only accounting for about 13 per cent of current reported cases.
"This is certainly an undercount as to be recorded as a reinfection requires both the initial and subsequent infection to have been recorded," O'Neale said.
The fraction of cases recorded as reinfections has been slowly growing throughout the year, partly because the protection people received from vaccination and exposure has faded over time.
"For many people, their first infection was back at the start of 2022 and the immunity from it will have waned significantly," he said.
"This, combined with relaxing of Covid protections and changes of behaviour means that people who were infected in the first Omicron wave are at more risk of a reinfection now, than they were earlier in the year."
And hospitalisations?
Previous waves have tended to be accompanied with a big spike in hospitalisations that's lagged about two weeks behind case rises, before we see jumps in deaths trail higher hospital cases.
But so far, Datta said, we didn't have enough information to draw any concrete conclusions about what toll this wave is likely to have.
Fortunately, there wasn't yet any evidence the new subvariants were proving any more severe than their Omicron predecessors.
"Other than avoiding infection completely, which is always the ideal situation, vaccination remains the best protection against severe disease with Covid-19," O'Neale said.
"Although protection from vaccination does wane over time, the waning of protection against severe disease is much slower than for protection against infection.
"This means that any increase in vaccination and booster levels will help to reduce hospitalisation numbers."
Yet uptake levels for second boosters remained well down compared with first boosters – and there were many eligible Kiwis now almost a year past their last booster.
"In contrast during the earlier March-April wave these groups, had been boosted only a month or two before," Harvey said.
Severe disease and mortality remained much more likely in older age groups - in which most some 2000 deaths reported this year have been concentrated – but a recent Ministry of Health analysis also suggested receiving two or more doses of vaccine slashed the risk of death by about 62 per cent.
What are some other factors in the mix?
Along with the main ones, modellers have plenty of other wildcard drivers to work into their maths – most of them human ones.
"It's common to see a pattern where reported Covid-like or influenza-like symptoms decrease during school holidays, especially during the winter, and then rise again following the return to school," O'Neale said.
"We can even see this pattern in different age groups with symptoms first reported in school-aged children and then a couple of weeks later in the age groups that would correspond to parents.
"This highlights the importance of measures to reduce transmission in schools."
Then there's the ongoing flow of people coming through our borders.
Modellers have been watching a gradual, national uptick in Covid-19 cases. Image / NIAID
With most international travel, we might expect infection rates among travellers to mirror those of the places they've come here from.
But as we've seen last week, visiting cruise ships pose a unique risk, with prevalence rates among them estimated at about 3 per cent – or four to five times higher than what we have nationally.
"This means that it is especially important for cruise ships to be implementing processes to prevent infectious passengers from going ashore where they might risk infecting others - often in smaller regional centres that may not have health facilities to deal with a sudden surge in cases," O'Neale said.
Covid-19 Modelling Aotearoa's Datta also pointed to the contribution of an increasing number of large, social events as we neared summer.
"It is difficult to determine the relative contribution of each of these factors to the uptick in cases."
Why aren't we regularly surveying the population?
Modellers estimate that up to two thirds of us have now had the virus: yet, without proper infection prevalence and sero-prevalence surveys in place, they can't give a precise figure.
They've been calling for a such a survey – as the UK has long been running – for most of the year, yet the Ministry of Health has only told the Herald a launch will come some time before 2023.
"The timing will depend on how quickly consultation can be completed, the necessary digital tools can be built, funding arrangements can be finalised, and ethics approval can be granted," a spokesperson said.
Harvey said one of the biggest concerns was a variant with increased severity arriving here: a scenario her team have been modelling for the Government.
"Ideally, we'd be able to see an early warning sign of this by keeping an eye on the proportion of cases that become hospitalised to see if there is any step change in this — as was seen with the Delta variant," she said.
"Unfortunately, we can't easily use this metric because case ascertainment rates are changing so much through time.
"With an infection prevalence survey, we'd have a much better estimate of the true prevalence of infection – and we'd be able to use that to compare against hospitalisation numbers."
As it stood, the team could only draw on data on reported cases and on hospitalisations throughout this year.
"From a modelling perspective, this is very problematic because we can construct models that 'match' this data throughout the year by 'simply' varying our parameters for things like hospitalisation rates, immunity waning rates, and reporting rates for different groups," Harvey said.
"This means that there can be models that fit the data that produce different estimates of past infections and consequently different future projections for infections and hospitalisations."
Having detailed survey data, she said, would help pin down specific variables at points in time, and would allow for much more accurate estimates of past and future infection levels.
"In short, an infection prevalence survey helps to ensure that models are getting the right answer for the right reason."
So, what can we actually say about how large this wave might get?
Amid this stew of variables, O'Neale said the general pattern his team expected to see was a peak either roughly comparable than our last – which topped out at just 10,000 daily cases - or smaller.
"This is largely due to the high number of people who have some immunity from past infection and vaccination."
The "when", meanwhile, depended on drivers.
"If case number continue to grow relatively slowly, compared with the previous two Omicron waves in Aotearoa, then we would expect a lower, slower peak with cases possibly not peaking until around the end of the year," he said.
"However, if case numbers grow rapidly, like they did in June and July, then it may be only a matter of weeks before we see numbers peak.
"The unfortunate side of this is that it is also likely to be associated with a higher peak height."
Given it took time for infections to spread to people still susceptible, growth powered by behavioural factors tended to play out more slowly.
"In contrast, growth that's driven by increased transmission due to immunity evasion of a new variant could be associated with faster rates," O'Neale said.
