12 January 2022

Modelling offers hope that Omicron's worst may be over sooner than expected

| Ian Bushnell
Join the conversation
Dave peffer

CHS CEO Dave Peffer: scenarios “give us a level of confidence we aren’t too far from seeing a peak”. Photo: Michelle Kroll.

There is cautious optimism that the ACT’s Omicron-driven COVID-19 surge may be close to peaking, with new internal Canberra Health Services modelling showing the current outbreak may subside sooner than expected.

In an email sent to hospital staff yesterday (11 January) and sighted by Region Media, CHS CEO Dave Peffer outlines three scenarios. The most likely one has new cases peaking on 17 January and the number of COVID patients in hospital peaking three days later on 20 January.

Health officials had believed that the ACT was a few weeks behind NSW, where the Omicron outbreak is tipped to peak in mid-January, although some believe it may have already done so.

But now the modelling, used to plan for hospital capacity, points to the current outbreak giving way before the end of the month, and the number of people in hospital peaking at just under 100, four times the current load but manageable without resorting to activating the Garran Surge Centre, according to hospital sources.

READ ALSO ACT records 1078 new COVID cases and hospitalisations fall, but testing down

Today there are 23 patients in hospital with COVID, down from 28 yesterday, but the figures are expected to jump around.

The best-case scenario has new cases peaking yesterday (11 January) and hospitalisations three days later, while the worst-case has cases peaking on 21 January and hospitalisations on 27 January.

Both of these are considered possible but unlikely.

Mr Peffer warns there are unknown factors such as the level of underreporting that may upset the modelling, but the scenarios suggest there is light at the end of the tunnel for stretched hospital staff, struggling businesses and the Canberra community.

“They’re based on data we have at this point, the experiences we’re seeing locally and overseas, and give us a level of confidence we aren’t too far from seeing a peak in cases and hospitalisations,” he said.

According to sources, CHS modelling has been quite accurate so far.

Mr Peffer told staff that managing the workload will be tough going, but the recent changes to patient cohorting will help.

“Our planning for this potential demand won’t stop – we’ll continue to make changes within our health services on a daily (and sometimes hourly) basis, to sustainably respond to the patients we see presenting,” he said.

A Rapid Antigen Test. Results can now be submitted online as the ACT transitions to this form of testing. Photo: File.

The issue of underreporting should be resolved with the announcement today that people will be able to notify their Rapid Antigen Test results online as part of a transition to that form of testing.

Chief Minister Andrew Barr said a Check In CBR push notification would be sent today to users to let them know a reporting form on the ACT COVID-19 website is now live.

He said the change would provide more accurate data, better connect people with care and ease the pressure on testing centre queues as RATs become more available.

“We are working to secure a greater supply of tests for the ACT, including the purchase of one million test kits in partnership with NSW and our own orders,” Mr Barr said.

Acting Health Minister Chris Steel said the ACT would not be following NSW’s move to hit people with $1000 fines if they fail to report their RAT results.

He said those who have already had taken a rapid test should submit their results to ACT Health.

READ ALSO How to care for someone with COVID-19 at home

The government was still working on how to distribute RATs, but Mr Steel said some would be handed out at testing centres.

A sign of the increasing pressures on the health system was today’s announcement that the Central Health Intake (CHI) phone line used to book some services at community health centres and outpatient clinics will be closed temporarily from tomorrow (13 January).

CHS says the change is temporary so staff can be redeployed to support the team caring for people recovering at home with the virus.

Nurses will only be available to process urgent referrals from GPs, and people are being advised to speak to their GP if an urgent referral is needed, or cancel an appointment online.

Like many sectors, the health system is managing the challenge of staff falling ill or needing to isolate.

Join the conversation

All Comments
  • All Comments
  • Website Comments

The temptation to form premature theories based on insufficient data and faulty modelling and to then proffer poor advice is the bane of the medical profession.

There’s actually plenty of data available and the modelling isn’t “faulty”.

The main problem is that when modelling scenarios are presented in the media (usually as clickbait or sensationalism), most people don’t know how modelling actually works and what it’s used for.

Far too many people think these numbers are forecasts or predictions, when they are not. The modellers are typically giving a range of scenarios based on different underlying variables. These scenarios can then be used for planning purposes to achieve better overall outcomes.

The modelling on Covid has typically been very good but people think the models weren’t “accurate”, almost exclusively because governments actually responded to the scenarios (through health controls, restrictions etc), which change the underlying assumptions in the models.

I disagree. GIGO. Garbage in, garbage out. Modelling is proven faulty when outcomes predicted by the model fail to match reality. A model can produce a range of scenarios depending on data chosen, which can be omitted or manipulated depending on the desired outcome. For example, if the desire is to show high (or low) interest rates, or a high (or low) probability of an astroid hit, or high (or low) deaths from covid, one changes the data and assumptions used in the model. Elements within the media sensationalise, seek out dire theories and predictions and present worst case scenarios to spread alarm. So too do those with their own self-serving agendas. Politicans are no less guilty of persuing their own interests. There are always models to support worst case scenarios and always people who can be found to promote them. Some early models predicted wildly inaccurate death rates, which failed to eventuate not because action was taken, but because the predictions were simply wrong. The result of using faulty models are premature theories, poor advice, bad decisions, over-reactions, panic,s lock-downs, masks, sign-ins, travel restrictions, job loses, economic disruption, mandatory impositions… The misinterpretation of data and misuse of models achieves worse outcomes.
“Men, it has been well said, think in herds; it will be seen that they go mad in herds, while they only recover their senses slowly, and one by one.” From ‘Extraordinary Popular Delusions and the Madness of Crowds’ by Charles Mackay (1841), about the regular outbreaks of mass hysteria, manias and panics.
Behaviour is based on models of particular form, whether generated by computer or the mind. All models should be tempered by rational thought and common sence.

“All models should be tempered by rational thought and common sence [sic].”

So by tautology, you demonstrate that you have no relevant knowledge about modelling.

The rest of the rant we can safely discard.

“Modelling is proven faulty when outcomes predicted by the model fail to match reality”

This is literally the type of thinking that I pointed out was wrong and explained why.

The modelled scenarios are not forecasts or predictions and they don’t exist in isolation of any actions that individuals or the government might take to shift the outcomes.

“A model can produce a range of scenarios depending on data chosen, which can be omitted or manipulated depending on the desired outcome”

Which hasn’t remotely happened here with the modelling the government has been relying on, although I won’t say that certain lobby groups haven’t tried this on to push an agenda.

“Some early models predicted wildly inaccurate death rates, which failed to eventuate not because action was taken, but because the predictions were simply wrong”

And some early models were very accurate and even the ones that initially had problems were useful and were updated over time to better reflect reality. Which is exactly what they should do.

Based on your previous comments and the one above, it seems you are less worried about the usefulness and accuracy of the models and more around the fact that they didn’t fit your predetermined ideological position on what you “wanted” them to say.

Which is mighty ironic.

Daily Digest

Want the best Canberra news delivered daily? Every day we package the most popular Riotact stories and send them straight to your inbox. Sign-up now for trusted local news that will never be behind a paywall.

By submitting your email address you are agreeing to Region Group's terms and conditions and privacy policy.