New Zealand's Covid-19 vaccination coverage rates are about to drop – yet that's something being welcomed by experts who've long called for finally-announced data improvements.
The Ministry of Health has confirmed it's changing the way it reports on vaccination rates, following a new Stats NZ review into a population dataset widely criticised for under-counting coverage among Māori and Pacific people.
That dataset was called Health Service Utilisation 2020, or HSU 2020, which captured all people who'd used the health system in 2020 with an age cut-off at June 30 that year.
Because it had been known to undercount Māori in particular – a group that used health services less frequently, and sometimes misclassified under other identities – experts have questioned whether ministry reporting accurately reflected actual coverage rates among Māori.
That was a major worry - given Māori and Pasifika people were more than twice as likely to be hospitalised with Covid-19 – and one singled out in a critical December finding by the Waitangi Tribunal.
In May, the Herald's data team revealed similar issues with the HSU data, finding the ministry's total of "fully vaccinated" people over the age of 12 – 95.2 per cent – to be higher than the figure produced using Stats NZ's December 2021 population estimates, which came to 91.8 per cent.
Today, Public Health Agency's deputy director-general Dr Andrew Old announced the ministry was making several improvements to its dataset – which itself was to be replaced next week with an updated set for 2021.
This would add 233,000 people to the pool, resulting in the two-dose coverage of the eligible population aged over 12 falling from 95 to 90 per cent.
Among Māori, coverage rates will drop from 88 per cent to 83 per cent, and for Pacific peoples, from 97 per cent to 89 per cent.
But Old pointed out this didn't mean the total number of vaccinations had dropped.
"Put simply, it's the same number of people vaccinated, but we're now dividing that by a bigger number."
Along with being refreshed every six months, there'd be two other key changes to the ministry's vaccination dataset.
Someone who'd had two doses would now no longer be considered "fully vaccinated", but had received their primary course – while the reporting would also reflect peoples' current ages, rather than that at the time of their last vaccination, which would give more information about booster eligibility.
With one exception, Old said the ministry had either implemented the improvements suggested in Stats NZ's review, or was in the process of doing so.
That report, which investigated the methodology behind the ministry's wider HSU dataset, confirmed issues with ethnicity differences in the HSU.
"Careful examination" was needed to see whether the lower representation of Māori was down to systemic issues or "genuine reporting choices" by people using services.
It also flagged issues with age-based counting in the 2020 HSU, although noted these would be addressed in the updated set.
While the review acknowledged HSU came with limitations, it also noted its advantages.
One was a "consistency of reporting" of Māori and other ethnic groups in the numerator compared with the population denominator – including in vaccination rates.
"This use is consistent with Stats NZ's advice that customers should validate the consistency of the numerator with the denominator when undertaking calculations by ethnicity."
Māori health researcher Dr Rawiri Taonui said the review and the ministry's new changes didn't address all of his concerns – namely with how Māori deaths and hospitalisations were being presented – but he was nonetheless happy to see progress.
"It's good to see this update because, at the end of the day, the HSU is two years out of date – and it should have been updated annually."
Covid-19 modeller Dr Dion O'Neale similarly welcomed today's announcements – particularly that there'd be six-monthly updates.
But he questioned why this hadn't been done earlier – and why those 233,000 extra people hadn't been gradually added to the vaccination coverage dataset as they were discovered, rather than all at once.
"There were some easy wins that could have been made along the way that didn't require revising the entire HSU, which would have meant our numbers were up to date at the time."