Josh Frydenberg says Omicron is 75 per cent less severe than Delta and other COVID variants. Is that correct?

The claim

Following a wave of coronavirus cases driven by the Omicron variant in December and January, Treasurer Josh Frydenberg defended the government’s decision to move towards “normalised settings” for financial support related to the pandemic.

” … [W]e are always watching to see how the pandemic evolves. But what we do know as well is that the Omicron variant is 75 per cent less severe than the Delta variant and previous variants,” Mr Frydenberg told Patricia Karvelas on ABC Radio’s RN Breakfast program.

Is the Omicron variant 75 per cent less severe than Delta and other variants? RMIT ABC Fact Check investigates.

The verdict

Mr Frydenberg’s claim is oversimplified.

It is clear from a variety of academic studies that Omicron causes less severe disease than Delta.

What is less clear is the magnitude of this reduced severity and the contribution of previous infection or vaccination.

Mr Frydenberg’s claim is commensurate with two studies which were conducted in South Africa, but these studies do not unpick the relative reductions in the incidence of severe outcomes due to vaccination, previous infection and reduced virulence. A separate study from the same country estimated the latter to be 25 per cent.

And experts cautioned that the South African setting may not be comparable to the Australian setting due to differing age, vaccination, and previous infection profiles.

Studies in other countries have found a large range of lower reductions in risk to various clinical endpoints.

Another problem is that Mr Frydenberg’s claim implies all variants other than Omicron are similarly severe to Delta.

But research has shown that other variants have different levels of severity. As one expert noted, it’s not possible to compare them all with a single figure.

Experts said a comparison between Omicron and other, non-Delta variants made little sense, when those variants were no longer circulating widely.

Various variants

The existence of the novel coronavirus was first reported in December, 2019. The very first strain is often referred to as the “ancestral”, or Wuhan strain.

Since it first emerged, the virus has been mutating, producing a number of variants with different epidemiological characteristics. The World Health Organisation names a variant with a letter of the Greek alphabet when it has been designated a “variant of interest” or a “variant of concern”.

The WHO has working definitions of a variants of interest and variants of concern of SARS-CoV-2.

Variants of interest are classified as such when genetic changes are predicted or known which affect transmissibility, severity, immune escape, or diagnostic or therapeutic escape, and are causing significant community transmission in multiple countries.

Variants of interest

Variants of concern are a confirmed threat to public health and are classified as such when an increase in virulence or transmissibility or a decrease in the effect of public health measures or diagnostics is observed.

Variants of concern

It’s not the first time Mr Frydenberg has made this claim.

In an earlier interview with Channel 7’s David Koch, he said:

“The Omicron variant, Kochie, as we know, is 75 per cent less severe than previous strains.”

And the same morning, he made the claim in an interview with Channel 9’s Alison Langdon and Karl Stefanovic: “We do know that it’s 75 per cent less severe than in previous variants.”

There has been some discussion in the scientific community about the differences between the terms “variant” and “strain”, but these terms are generally used interchangeably in colloquial discussion.

Measuring ‘severity’

There is a spectrum of clinical outcomes produced by infection with SARS-CoV-2.

WHO provides “living guidance” for the clinical management of COVID-19, with definitions of mild, moderate, severe and critical disease.

The organisation defines these clinical presentations differently in adults and children, using oxygen levels, respiratory rates and ” clinical signs of pneumonia (fever, cough, dyspnoea)”.

The guidelines state that those with moderate disease may or may not require hospitalisation, but the management guidelines for severe disease, which include “proning” and supplemental oxygen therapy imply admission to hospital.

Mr Frydenberg spoke of relative severity, which can be measured by comparing one or more clinical outcomes, the most serious of which include hospitalisation, ICU admission and death.

Importantly, whether being infected with SARS-CoV-2 causes severe or critical clinical outcomes in a patient can be influenced by a number of factors.

Some of these are inherent in a patient, such as their age. Older age brackets, for example, have been associated with an increased risk of death.

Others are less so, such as whether a patient has previously been infected by the SARS-CoV-2 virus, or whether the patient has been vaccinated. Both of these factors have been associated with a decreased risk of severe outcomes.

And then there is the issue of the virulence] of the virus itself. That is, how severe are the clinical outcomes for a given strain, with all other factors being equal?

With a substantial proportion of Australia’s total population still unvaccinated, whether or not the virus is less severe without the effect of vaccination is a pertinent question.

Mr Frydenberg’s claim could convince someone who is vaccine hesitant, for example, to remain unvaccinated or not vaccinate someone in their care, if they interpreted Mr Frydenberg as speaking about the inherent virulence, absent the effect of a vaccine.

Fact Check takes Mr Frydenberg to be talking about this inherent virulence, rather than the effect of vaccination or previous infection on the severe outcomes of the virus.

The source of the claim

Fact Check contacted Mr Frydenberg’s office to ask for the source of his claim.

A spokeswoman pointed to an opinion piece written by Doherty Institute director Sharon Lewin in the Sydney Morning Herald on January 2.

Professor Lewin compared the risk of hospitalisation between Omicron and Delta.

” … data from South Africa shows the risk of hospitalisation with Omicron compared with Delta is reduced by 80 per cent and, once in hospital, the risk of severe diseases with Omicron is reduced by 70 per cent.” Professor Lewin wrote.

Mr Frydenberg’s spokeswoman also directed Fact Check towards comments made by Chief Medical Officer Paul Kelly on December 31:

“Omicron, though, is less severe. I think that is becoming very clear now … There was a preprint paper that I read from South Africa. South Africa’s a different country to in various ways, but they have had the longest and most extensive experience of this particular form of the COVID-19 virus, the Omicron variant. They are seeing a 73 per cent decrease in severity.”

So which studies are Professor Lewin and Professor Kelly referring to?

South African studies

On December 21, 2021, a few weeks before Professor Lewin’s opinion piece, researchers from South Africa’s National Institute for Communicable Diseases published a pre-print, which was later published in peer-reviewed journal The Lancet on January 29, 2022.

The study used the lack of a certain gene, the S-gene, on the spike protein as a proxy to distinguish the Omicron infections. It referred to these as S-gene target failure, or SGTF, cases. At the time Omicron emerged in South Africa, the prevailing “non-SGTF” variant was Delta.

Loading

The study assessed comparative severity in two time periods. First, it looked at contemporaneous infections between October 1 and November 30, 2021. By the end of that period, the Omicron-proxy infections accounted for 98 per cent of infections.

Second, it compared SGTF infections in that period with genomically-sequenced Delta infections between April 1 and November 9, 2021.

The first analysis found an 80 per cent reduction in the risk of hospitalisation of SGTF infections relative to non-SGTF infections.

This analysis adjusted for a range of factors known to affect clinical outcomes for COVID-19 in South Africa, including age, sex, province, and health-care sector (private vs public) and known previous SARS-CoV-2 infection. But it could not adjust for the effects of vaccination status or comorbidities due to limited available data.

The second analysis found a 70 per cent reduction in severe disease with the Omicron variant relative to Delta in already hospitalised patients. Severe disease was defined as “a hospitalised patient meeting at least one of the following criteria: admitted to an intensive care unit; received oxygen treatment; was ventilated; received extracorporeal membrane oxygenation; had acute respiratory distress syndrome; or had died”.

On top of other factors already mentioned, the analysis additionally adjusted for the effects of vaccination status and comorbidities.

The study’s corresponding author, Nicole Wolter, of the Centre for Respiratory Diseases and Meningitis at the NICD, told Fact Check that reinfection status is generally under-reported and vaccination status was self-reported and only available for a subset of hospitalised patients in the study.

This, she said, meant the study could not fully adjust for these factors.

Furthermore, she said via email that the results of this study could be used to make comparisons between Omicron and Delta only.

“Our study found a reduced risk of hospitalisation/severity among Omicron compared to non-SGTF/Delta infections. This is likely due to a combination of high immunity in our population (from high numbers of previous infection and some vaccination), and potentially also reduced virulence of the Omicron variant,” she said.

“From our study we are not able to disentangle the relative contribution of each of these factors, and cannot assign an overall magnitude as to how much less severe Omicron is relative to previous variants. We also did not compare Omicron to the ancestral strain, Alpha and Beta variants.”

Which study was Professor Kelly referring to?

A spokeswoman for the Department of Health confirmed to Fact Check that Professor Kelly was relying on a different South African pre-print study.

This study analysed the beginning of three COVID-19 waves in Gauteng province which were dominated by different strains — Beta, Delta and Omicron, and is again from the NICD.

It found a reduction of 73 per cent in severe disease in the recent Omicron wave in hospitalised patients when compared to the Delta wave in the province, which is commensurate with the remarks made by Professor Kelly.

Severe disease was defined as “one or more of acute respiratory distress, supplemental oxygen, mechanical ventilation, high/intensive care or death”.

The Beta wave was also analysed relative to the Delta wave, with only a 5 per cent reduction in the risk of severe disease, indicating similar severity profiles between Beta and Delta.

However, the study notes that its source, a national database called DATCOV, contains incomplete data on previous infection and vaccination, and therefore: “severely limits exploration of their potential roles in lower disease severity observed”.

“This has meant that the role of vaccination could not readily be studied in relation to clinical severity of COVID-19,” the authors wrote.