2 days ago
Counting the true toll of the Covid-19 pandemic in New Zealand
As the Royal Commission of Inquiry into New Zealand's Covid-19 response continues, modelling expert Michael Plank presents a new study on the country's excess mortality rate.
How many people died because of the Covid-19 pandemic in New Zealand? It sounds like a simple question, but the answer depends on more than just counting reported Covid-19 deaths.
My colleagues and I recently published a study in the International Journal of Epidemiology looking at excess mortality – the number of deaths above what we would have expected if there hadn't been a pandemic.
Excess mortality helps us to measure the overall impact of the pandemic, not just from Covid-19 itself but also from things like delayed medical care or the side effects of lockdowns. And because it's based on the total number of deaths from any cause, it doesn't depend on how the cause of death was determined, nor whether the death was recorded as being due to Covid-19 or not.
Many people will have seen the Our World in Data Covid-19 dashboard, which allows users to compare excess mortality between countries. This dashboard shows that New Zealand's total excess mortality up to the end of 2023 was less than 1%. In other words, the number of deaths during the pandemic was less than 1% higher than expected.
But not everyone agrees with this conclusion.
A 2024 study by John Gibson argued that the excess mortality in New Zealand was actually much higher than this. Our World in Data's method missed a crucial factor: New Zealand's population growth ground to a halt in 2020 due to pandemic travel restrictions. With fewer people in the country, Gibson claimed, we should have expected fewer deaths; so the excess mortality was actually higher.
We wanted to know if this was really true. Could the Our World in Data dashboard be inadvertently hiding a swathe of excess deaths in New Zealand?
To answer this question, we built a statistical model that estimated trends in the death rate in different age groups over time. We then used this model to calculate how many deaths would have been expected if the pandemic had never happened and pre-pandemic trends had simply continued.
Our model accounts for changes in population size and ageing to ensure a fair comparison. We looked at excess mortality up to the end of 2023 because we wanted to include the period after New Zealand's elimination strategy ended and the virus became widespread.
Was New Zealand's pandemic death toll higher than reported?
The answer from our work is a resounding 'no'.
We estimated the total number of deaths between 2020 and 2023 was somewhere between 2% higher than expected and 0.8% lower. In other words, we can't be confident that more people died during the pandemic than would have died anyway. We can be confident that the number of deaths was no more than 2% higher than expected.
In 2020, the number of deaths was unusually low, mainly because border closures and lockdowns inadvertently wiped out influenza as well as Covid-19.
In 2022 and 2023, deaths increased as Covid-19 became widespread. The timing of excess deaths matched very closely with reported Covid-19 deaths, with the highest death rates occurring during the biggest Covid-19 waves of 2022. This suggests that the virus itself was the main driver, rather than indirect factors.
Overall, New Zealand's estimated excess mortality of less than 2% is far lower than that in countries like the United Kingdom (10%) or United States (11%) over the same period.
And although our study only covered the period up to the end of 2023, the overall age-standardised mortality rate in 2024 was below pre-pandemic levels.
This shows that there is no mystery factor causing large numbers of unexplained deaths. Reported Covid-19 deaths appear be an accurate measure of the pandemic's death toll.
So why the controversy?
Gibson was right that New Zealand's population growth stalled during the pandemic. But that's only part of the story.
Most deaths happen in older people, and this part of the population continued to grow during the pandemic. So, even though total population growth slowed, the number of elderly people – the group at highest risk of dying – still increased as expected.
In other words, New Zealand's ageing population was a more important driver of the expected number of deaths than the number of immigrants, who tend to be relatively young.
Why does this matter?
The next pandemic is a question of when, not if. If we are to respond better to future pandemics, it's essential that we understand the full impact of our response to Covid-19.
Some critics argue that New Zealand's elimination strategy just delayed the inevitable. Deaths that were prevented in 2020 and 2021 – the argument goes – were simply delayed until 2022 or 2023, when the virus became widespread.
But the data tell a different story. Our response bought time for people to get vaccinated before they were exposed to the virus. And that massively reduced the fatality risk.
New Zealand's response was far from perfect, and there were undoubtedly harms as a result of lockdowns and other measures that are not reflected in mortality statistics. But there can be no doubt that the response saved thousands of lives compared with the alternatives.
Michael Plank led a group of researchers who were commissioned by the New Zealand government to provide modelling in support of the response to Covid-19 between 2020 and 2023.
