When rapid antigen tests were handed out at her daughter's school in southern Sydney, Alyce Odams assumed they would be reliable.
The saliva-based RATs — dubbed 'lollipop tests' — were given out in primary schools across New South Wales, Victoria, and Western Australia last term.
Ms Odams initially liked the child-friendly prospect of a test without a painful nasal swab.
But when her family contracted COVID, the mother-of-two felt let down.
"The entire time that we were positive, we were testing negative on a lollipop," Ms Odams said.
"We have never actually gotten a positive lollipop RAT test."
Experts say Ms Odams is not alone in feeling frustrated about false negative results.
When approving RATs for use in Australia, the Therapeutic Goods Administration (TGA) relies on data provided by the same companies that sell the tests.
University of Sydney clinical epidemiologist Katy Bell said this puts the data "at high risk of bias".
"It's well known that if you use the case-control study design, which is what the RAT manufacturers tend to use, you're going to overestimate the true diagnostic accuracy of the test," Dr Bell said.
On the TGA website, all RATs available in Australia are reported to have a minimum sensitivity of 80 per cent.
For lollipop tests, this reported sensitivity is 95 per cent.
The Australian Medical Association's (AMA) Chris Moy said research from overseas suggests RATs perform nowhere near this well in real life.
"Some studies in the United States show that the sensitivity is down to 60 per cent if we're lucky," Dr Moy said.
"That means the miss rate is very high."
Dr Moy said receiving false negative results could stop COVID patients from accessing life-saving medication.
"One of the key management aspects of the COVID outbreak at the moment is the use of antiviral treatments — particularly for those at higher risk.
"But these medications have to be taken within the first five days (of infection).
"If you've got a rapid antigen test ... that will only prove you're positive on day four or five, it'll be too late to be able to implement one of these medications."
The TGA told the ABC it was planning to commence independent testing on RATs to verify manufacturers' claims about their sensitivity.
Results from the independent testing — run by the Doherty Institute — will be released on the TGA website "as they become available".
But parents like Ms Odams said they were disappointed that unreliable tests were given out in schools.
"There were only a handful of kids that didn't end up getting COVID at that time," she said
"Everyone was using the lollipop tests, so they were getting the negative results, and the kids were going to school.
"It's really concerning."
In order to present more accurate data on the sensitivity of RATs, Dr Bell said the TGA should seek evaluations that use better research methods, including population based studies.
For now, Dr Bell said parents with children who have COVID symptoms should keep them home from school and not rely on negative RAT results.
