How to make sense of COVID numbers as rapid tests rip up the rule book

For our free coronavirus pandemic coverage, learn more here.

COVID-19 case numbers have spiked and plateaued.

Hospitalisations are at record levels in Victoria as Delta’s peaks have been overtaken by Omicron.

Admissions to ICU are also rising but are still well short of state capacities. Death rates have increased and Australia’s infection rate has become among the highest in the world on a per capita basis.

What does all that mean, how can we make the best sense of raw numbers and COVID data, and what do national and international trends suggest we can expect next?

Do daily case numbers still tell a clear story?

For almost two years, daily case numbers are how Australians have primarily made sense of the pandemic, a gauge on whether infections have risen, fallen or remained steady over time.

When the seven-day case average started to plateau during previous outbreaks, this was a sign that daily infections had peaked.

But since the start of 2022, daily case numbers have been far more opaque.

When Omicron started to spread rapidly over the Christmas break some testing sites closed down, while long queues at others sometimes resulted in people being turned away. The accuracy of daily case numbers had always depended on people going out to PCR testing sites and results being processed promptly, but then the state’s testing infrastructure started to buckle.

Recent daily case numbers have a second complication: the overall total now comes from two separate sources.

Until recently, daily case numbers were calculated by tallying up all the positive PCR test results processed over the previous day, but now the results of self-administered (and self-reported) rapid antigen tests are being fed into the daily totals.

In Victoria, there was a sudden spike in reported cases on January 8, which coincided with the day rapid test results were added to daily case reports.

When people started submitting positive results to health authorities they were logging tests that had been carried out as far back as a fortnight prior. The sudden backlog of previously unreported cases led to huge and sudden spikes in the daily numbers.

The new reporting methodology provides a more accurate picture of how much virus is circulating in the community. However, cases can still be missed, including among those who unable to present for PCR testing, those unable to obtain a rapid test (or who don’t self-report a positive result), or those who make an assumption they have the virus based on symptoms and self-isolate without taking a test.

Indeed, Victoria and NSW daily case numbers have started to drop. But given the complexities of accurately interpreting daily case numbers, is the data a true indication that case numbers are on their way down?

University of South Australia biostatistics professor Adrian Esterman’s analysis uses a four-day window to gauge the spread of the virus, based on the typical time it took from exposure to developing symptoms. Based on this analysis, he concluded that daily case numbers have peaked.

“In all states and territories, apart from Western Australia, the peak has been reached and we are on the way down,” he said.

However, he said case numbers would come down slowly, and would not drop off at the same pace as they had spiked.

“There will be bumps. Cases will increase once schools return, but case numbers are unlikely to then reach the levels of this summer outbreak,” Professor Esterman said.

Deakin University epidemiology chair Catherine Bennett said COVID screening programs and increased access to rapid tests are likely to lead to an increased number of cases recorded at schools and workplaces as people return from summer holidays.

That could seem alarming because of an uptick in daily case numbers, but it would be a sign that cases were being detected before infected people could pass on the virus to others.

“February will be a bumpy month,” Professor Bennett said.

Deakin University chair of epidemiology Catherine Bennett.Credit:Jason South

If cases numbers are no longer the best way to make sense of COVID, then what is?

You may have noticed on The Age website we have switched the order of cases and hospitalisations in the little graphic near the top of the homepage. That’s because hospitalisations will likely be the key metric in the weeks ahead.

Last week, when declaring a statewide code brown to ease the pressure on hospitals, the government predicted the number of hospitalisations in Victoria — which usually lag cases by a week or two — could peak next month at 2500 COVID-19 patients, with up to 100 new admissions a day.

Government projections also forecast there would be a rise in intensive care admissions roughly a week after hospitalisations peak if the Omicron wave follows a similar trajectory to previous outbreaks.

In Victoria, there are currently 1089 people in hospital because of COVID-19, more than at the height of 2020’s second wave and last year’s Delta outbreak.

You can see how the peak in hospitalisations tends to come after case numbers peak in these graphs from previous outbreaks. During Victoria’s second wave of infections in 2020, the seven-day average reached a maximum of just over 500 cases per day on August 5, but the total number of people in hospital at a given time peaked at 675 exactly two weeks later, on August 19.

And during last year’s Delta outbreak, hospitalisations in NSW peaked at 1268 just over a week after the seven-day average peaked at 1366.

You can see from this graph that the Delta outbreak in Victoria resulted in higher case numbers than NSW, but that hospitalisation numbers in Victoria remained below NSW. That’s partly because by the time Victoria experienced its outbreak, the vaccine rollout was further along, so a higher proportion of its adult population had received at least one jab.

Associate Professor James Trauer, from Monash University’s School of Public Health and Preventative Medicine, said he suspects the surge in hospitalisations could fall short of government projections, potentially peaking at 2000 people in hospital because of COVID-19 in February.

He theorised testing numbers had stabilised in the past week, and case numbers were consistently falling, mirroring the downward slope of outbreaks in countries such as South Africa.

But importantly, he said, Omicron appeared to have a shorter time frame from infection to symptom onset, which he suspected meant those needing hospitalisation would be admitted sooner than previous waves. (With previous variants, such as Delta and Alpha, the time between infection and hospitalisation was roughly one to two weeks.)

“You have to be cautious, but I suspect that we won’t hit 2500 as we once thought,” the respiratory physician said.

Those infected with Omicron were also staying in hospital for shorter periods, often being discharged within a few days, and not requiring intensive care, so the turnover of patients would be quicker, he said.

Professor Bennett had a different view; she said it was likely that hospitalisations had already peaked. She said the state had not broken from the pattern of hospitalisations lagging case numbers, but that delays in people receiving PCR test results and the addition of belated rapid tests to the daily totals meant that the case data itself was playing catch-up.

Here is how hospitalisations have been tracking against case numbers in Victoria since the start of the Omicron wave:

The sheer increase in case numbers over the past couple of months outstrips that of hospitalisations. Even taking into account that increases in hospitalisations tend to lag increases in case numbers, a lower percentage of current infections are ending up in hospital compared with previous waves.

A NSW Health report found that 10 per cent of cases confirmed during last year’s Delta outbreak led to hospitalisation, compared with 1 per cent of cases confirmed in the six weeks to January 8 of this year.

Studies in the UK have found that Omicron is less likely to result in hospitalisation than previous variants. However, it also spreads faster than previous variants. So while it is less likely to put someone in hospital, the sheer number of cases being confirmed is causing record hospitalisation numbers in Australia.

But Professor Esterman said high vaccination rates made it more difficult to say whether Omicron was milder than previous variants, or if it’s simply perceived as milder because most of those who contract the variant are already vaccinated.

New hospital data obtained by The Age reveals Victorians hospitalised during the current COVID-19 outbreak have a median age about two decades older than those hospitalised at times during the Delta wave.

According to the latest health data, the median age of COVID-19 patients in Melbourne hospitals in January was 62, and 67 for regional hospitals.

The data also reveals Victorians who have received their third dose are being admitted to hospital at extremely low rates, with older people who are yet to receive a third vaccine dose being hit the hardest.

It shows just 45 people who have received three doses of a COVID-19 vaccine have needed a hospital bed – less than 5 per cent of the 1029 coronavirus patients in hospitals on Saturday. The majority were aged 70 or over.

The news adds to scientific studies concluding a third dose provides the highest level of protection against Omicron illness and the observations of local doctors and officials that booster doses will be critical to attempts to control the highly infectious variant.

When hospitalisations rise, so too does the demand for intensive care beds.

While COVID-19 hospitalisations are at record highs in Victoria and NSW, neither state is at the stage where there are more intensive care beds occupied than during the height of last year’s Delta outbreak.

However, as you can see from this graph of the total number of people in intensive care because of the virus in Victoria over time, the number of people in intensive care on a given day has been rising.

There are currently 113 people in intensive care in Victoria because of COVID, and this figure has stabilised in recent days. During last year’s Delta outbreak, there was a peak of 163 people in intensive care statewide.

A recent government analysis of hospitalised cases in Victoria found that unvaccinated COVID-19 patients made up 22.2 per cent of cases and 37.2 per cent of cases in ICU – despite unvaccinated adults accounting for less than 6 per cent of the entire Victorian community.

OK, so what about the numbers of deaths we’re seeing?

You have most likely noticed a pattern in the data by now – when case numbers rise, so do hospitalisations, followed by intensive care admissions. The result is that deaths also start to rise.

On Wednesday, Victoria recorded 35 deaths, the highest single-day total since the second wave of infections in 2020.

But the graph above is more complex than it may appear at first glance. During the state’s second wave of infections in 2020, 41 deaths were confirmed on August 31 and 59 on September 4, but these single-day totals included delayed reporting of deaths from previous weeks.

Professor Esterman said daily death numbers tended to start decreasing about two weeks after daily case numbers started to drop. That would mean about one week from now.

However, he warned that daily death numbers could fluctuate, making it harder to discern a pattern.

Doctors describe an older group with moderate to severe COVID-19 infection, usually not requiring ICU treatment. The patient may have had one or two vaccine doses but not their booster, and underlying lung disease, diabetes and other conditions that make them more susceptible to falling very ill.

University of Melbourne vaccine expert and epidemiologist Fiona Russell believes “our saving grace” in Victoria has been the high double dose vaccination, which is hovering at 92.4 per cent of the population aged 12 and over.

Professor Fiona Russell of the University of Melbourne.

Professor Russell said that without such coverage, the toll of Omicron would undoubtedly have been far worse and the sheer volume of cases would have led to mass hospitalisations.

Evidence suggests three doses are about 90 per cent effective in preventing severe disease with Omicron, and Professor Russell said we must now focus on speeding up our booster program, particularly for the over 60s, and those at heightened risk of severe disease from the virus.

There is a growing body of research that suggests immunocompromised people need a fourth dose as their protection against the virus wanes quickly, even after a third dose.

Professor Russell said crucial questions remained about vaccinated people being admitted to hospital after being infected with coronavirus.

“It’s really important to understand who those people are: is it the elderly who have not yet been able to get their booster? The people with underlying problems who may need four doses to protect them against severe disease? All this information is needed to understand what is going on.”

How does Australia compare with other countries?

Other countries are also dealing with Omicron outbreaks, with case numbers surging in the United States and Britain over the Christmas period. However, on a per capita basis, Australia has been recording a higher rate of infections:

Figures from Our World in Data, which tracks global COVID-19 numbers, show that Australia’s reported seven-day case average peaked at about 4000 cases per million residents, tracking above both the US and Britain.

Britain’s Omicron outbreak kicked off a short time before cases started to surge in Australia, reaching its zenith in the first week of January, and its seven-day average has since stabilised at about 1500 cases per million residents.

The hospitalisation rate in Britain then peaked during middle of January and has also started to dip over the past week.

Both the US and Britain, despite recording lower case rates so far this year, are both tracking above Australia when it comes to the hospitalisation rate from COVID.

The hospitalisation rate for Britain is currently about 250 per million residents, while in the US it is about 430 per million people. In Australia, the rate is about 200 people in hospital for every million residents.

Australia is also recording a lower rate of new deaths than the US and Britain, however this graph also reflects how the peak in deaths typically follows the peak of cases and hospitalisations.

In addition, the US and UK both had a higher new daily case rate than Australia until early January, which means their death rate has also been higher over the period shown in the graph.

Stay across the most crucial developments related to the pandemic with the Coronavirus Update. Sign up for the weekly newsletter.

Most Viewed in National

From our partners

Source: Read Full Article