As the number of Covid cases in Australia has become impossible to accurately determine using test and case numbers alone, epidemiologists say it is crucial to undertake population-wide surveys to understand both current levels of infection and cumulative past infections.
John Kaldor, a professor of epidemiology at University of New South Wales’s Kirby Institute, and his colleagues at other organisations conducted seroprevalence surveys in 2020 and 2021. They are hoping to conduct repeat surveys this year to better understand the spread of the Omicron outbreak. “Antibodies can fade, but people probably have some detectable antibodies at least six months after infection,” he said.
“We are hoping to conduct some repeat surveys during the course of this year, but there are a range of approvals we need first.”
Serology testing involves taking a sample of a person’s blood to look for antibodies against the Covid virus. The presence of the antibodies indicates a person has been previously infected, vaccinated or both. By taking a large enough number of blood samples from a diverse cross-section of people, this data can be extrapolated to provide a snapshot of the percentage of a population who have been infected and have some immunity.
Kaldor said it would also be important to follow the UK and conduct regular random household surveys to ask people about their recent symptoms, and to undertake home swab testing. This would also detect asymptomatic cases and provide a clearer understanding of current infections.
Fiona Russell, a professor of paediatrics and epidemiology at the University of Melbourne, said seroprevalence studies cannot be done “overnight” and there were particular challenges in collecting a representative sample in Australia.
“There’s practical, logistical and feasibility issues around having a representative sample. For example, making sure you include people who live in regional areas, and then getting consent, collecting their blood … To do all of that you need funding,” she said.
“It takes time, but it can and must be done.”
In the meantime, Russell believes states with widespread Covid cases, including Queensland, Victoria and New South Wales should conduct mini “rapid” seroprevalence surveys.
“It would involve examining blood that has already been taken and stored, for example from people who might have come into hospital for an operation, or had their blood taken for another reason,” Russell said. “You could do it by postcode, or by age, to get a quick look at what is happening in a certain age group or area.”
Seroprevalence surveys can help estimate the level of immunity in a population, but are different to determining the incidence of the virus – that is, the rate of new cases of the virus being detected each day or week.
When Australia had low case numbers and widely available and accessible testing with fast results, it was easier to estimate with higher confidence the number of people with an active case of the virus at any one time.
While cases are always underestimated – even during small outbreaks – the sheer number of people now unable to be tested, asymptomatic people, or tests never being analysed in the laboratory due to lengthy delays from the time of collection means overall test numbers and positive cases from those tests are less accurate in understanding incidence in the population.
Nonetheless, Kaldor said efforts were being made to provide estimates of incidence by using a mixture of data, surveys of testing behaviour, and modelling tools. Key pieces of information will be the proportion of people undertaking rapid antigen tests and actually reporting their results.
“The epidemiological and modelling community is working on methods to make adjustments to correct for levels of testing in different parts of the population, and how to do that is still an area of discussion,” he said.
