Scratch the surface and the Doherty model looks pretty bodgy
So more and more we’re hearing about the ‘Doherty Modelling’. It seems our fate hinges on how the Doherty Institute says we should play the end of lockdowns.
I’d never heard of them so I had to look it up. The Peter Doherty Institute for Infection and Immunity (it’s full name!) is a collaboration between the University of Melbourne and The Royal Melbourne Hospital. It’s named after Peter Doherty, an Aussie champion who won a Nobel Prize in 1996 for discovering how the immune system recognises virus-infected cells.
(There you go.)
And the government asked Doherty to model what would happen if they wound back lockdown restrictions under different vaccination rate scenarios.
(This is important. The government commissioned – and paid for – this report.)
And what did the Doherty Model tell us? Well, if you believed the politicians, it said that we SHOULD open back up once we get to 70-80% vaccination rates.
That’s how it was sold. On the 19th of August, Premier Gladys Berejiklian said in the NSW daily update “as the Doherty Report says once you get to 80% double doses and you have to open up”.
You HAVE to open up.
On the 23rd of August Prime Minister Morrison said in Parliament “The plan as we set out very clearly about…that when we reach 70%, and we reach 80%, we can move through to phase B of that plan, those marks have been set by the Doherty analysis that was undertaken Mr Speaker”.
Only trouble is, the Doherty Report didn’t say that. The Doherty Report didn’t not say that either. All it did was model what would happen under the scenarios the government gave it, using assumptions the government gave it.
Now I haven’t actually read it. I’m a busy man. But I know a guy who has – Matt Barrie, the CEO of Aussie tech darling Freelancer.com.
He gives us a pretty brutal dissection of the report. It’s worth reading. But the key takeaways for me are:
Doherty only models out to 180 days
There doesn’t seem to be a good reason for this, other than perhaps that it takes us right up to the next election, and not beyond.
But as a result of the 180 day cut-off (which the government asked for), some of the charts look pretty ridiculous:
It also means that our totals are, in Barrie’s words, “garbage”, because in many cases they stop counting just as cases and deaths start to rise.
Doherty assumes unlimited ICU capacity
The Report notes that “ICU admissions [assume] unconstrained capacity, even when national thresholds are anticipated to be reached or exceeded”.
To “open up” under the Doherty modelling, even at 70%, Australia would have to double ICU capacity.
But it’s not just a case of rolling in another bed. Each ICU bed is manned by the equivalent of five full-time registered nurses across eight-hour shifts, 24 hours a day, seven days a week. Approximately half of these nurses need to hold a postgraduate qualification in critical care nursing.
Grattan Institute health economist Stephen Duckett has said “When we hit a trigger point of 12,000 new cases every day, then we know that we will hit ICU capacity shortly after if new cases continue to grow”.
The Doherty modelling assumes new cases will reach at least 55,000 per day.
That looks like a problem to me.
Doherty assumes perfect TTI+Q
The “test trace isolate & quarantine”(TTI+Q) assumed in the modelling is based upon the performance of the Victorian public health response at the height of the ‘second wave’ in 2020.
These cases peaked at 700 per day.
In the Doherty modelling the cases peak at over at least 55,000 a day, and could even get up to 120,000 a day.
The assumption that public testing, tracing, isolation & quarantine can perform at the levels Victoria performed at 700 cases a day is fantasy.
Even on the latest outbreak in NSW, tracing seems to tap-out at just 200 cases, with everything above that “unspecified”.
The quarantine program is not with out its problems either. Currently we’ve had one breach for every 67 infected, with only a few thousand travellers to work with.
That doesn’t inspire confidence.
There is no ‘freedom’
Even taking the assumptions at face value, Doherty modelling shows lockdowns occurring 80–273 days a year. That’s 3-9 months of the year!
Is that the freedom we’re aiming for?
Doherty assumes vaccines are incredibly effective.
The Doherty Report assumes a massive (90+%) drop in transmissibility with a double-does of the vaccine.
But there is no international evidence for such a high effectiveness rate.
If this was even remotely close, new daily infections in the US, UK and Israel would be well under control by now. They are not.
Indeed in Israel, case rates have climbed almost as quickly among people with two doses as among the unvaccinated.
So look, like Barrie, I don’t think Doherty is doing anything dodgy here.
They’re just crunching the numbers on the scenarios the government gave them, using the assumptions the government gave them.
That these assumptions are completely ridiculous is not on Doherty.
It’s on the government.
As Barrie says,
“Abdication of leadership to reliance on a modelled scenario is simply a way to blame someone else if it all goes pear shaped, which the Doherty Institute appears all too happy to play along with, either because it is a captured entity or doesn’t see what is coming.”
Yep, in 180 days our politicians will be saying, don’t blame us, blame the models.
Weak as piss.