Call for antibody testing to understand silent spread

Antibody testing is ‘‘overdue’’, to understand how much disease might lurk undetected, some health experts say.

Professor Des Gorman, from the University of Auckland’s school of medicine, said New Zealand could only be confident it had controlled Covid-19 if it had ‘‘deep knowledge’’ of how the virus spread in the community.

While the Ministry of Health’s surveillance testing includes people who show no signs of being unwell, diagnostic swabs can only tell you whether a person is infected at that moment.

So far, that testing has revealed only one positive case.

Gorman said officials should also do antibody testing, to identify people who had been exposed to the virus without knowing.

‘‘It’s overdue . . . It’s not instead of, it’s as well as. The more testing you do, the more confidence you can have that prevalence really does approach zero.’’

Philip Hill, a University of Otago McAuley professor of international health, and Chris Bullen, a University of Auckland professor of health innovation, presented a proposal to the ministry several weeks ago to test Aucklanders.

They planned to progressively test representative samples of 2500 residents, plus 2000 Ma¯ori and Pasifika, until they uncovered 10 positive cases.

Hill said antibody surveys would help researchers understand how far Covid-19 had penetrated.

They could also be useful to check for hot spots in places such as Queenstown Lakes before opening the region to Kiwi or Aussie skiers.

‘‘It might be a good idea to get a feel for whether it was just sporadic cases from travellers . . . or whether there’s evidence that it might have impacted more of the population than people realise.’’

Scientists have been wary of antibody tests because some were wildly inaccurate. An analysis of 14 tests by 50 scientists found only three were consistently reliable.

Drug maker Roche this week released a test it claimed would produce no false negative results and was 99.8 per cent accurate at identifying people with no antibodies, meaning it would produce two false positives in 1000 tests.

Hill said accuracy problems were amplified in places such as New Zealand, where the disease was not thought to be widespread.

A test would need to be about 99 per cent accurate to prevent false positives skewing the results. The ‘‘point of care’’ tests resembling pregnancy tests were not fit for use here, he said.

The ministry said it was talking with drug makers Abbott and Roche about their antibody tests, and assessing point-of-care tests. It was considering how antibody testing could be useful, but would need to be confident in the tests’ reliability.

Professor Michael Baker, an Otago University epidemiologist, believed antibody testing would be a waste of time and resources.

In places such as New Zealand where a ‘‘vanishingly small’’ proportion of the population had been exposed, false positives would probably exceed real positive cases, he said.