Burying The Answer

In the fight against COVID-19, computer modelling is the first casualty. By Nick Cater.

While the campaign to stop the spread of COVID-19 is far from over, some of the more extreme predictions about the collapse of the health service have proved to be ridiculous.

Less than four weeks ago, the Grattan Institute published modelling that forecast that intensive care units would be overflowing by mid-April unless the country was put into severe lockdown. It was illustrated with the obligatory hockey-stick graph running off the chart, predicting that more than 4000 patients would be in intensive care beds. 

Mid-April fell on Wednesday, the 15th day of the month, with 76 COVID-19 patients in ICU wards occupying approximately 3 per cent of the beds that were available at the start of the crisis. The Federal Government has subsequently trebled the number of ICU beds available, meaning that 99 per cent were free.

Exaggerated claims based on computer forecasting are unhelpful at the best of times. In a crisis like this they are positively dangerous, particularly when accompanied by sledgehammer policy advice that doesn’t even try to calculate the economic and social cost of the extreme policy it advocates.

The social distancing and isolation measures adopted by the Australian government fall far short of those the Grattan Institute and others have advocated.

And yet they have demonstrably met their goal of controlling the rate of new infections to maintain the integrity of our health system. 

This goal has been successfully accomplished through a range of actions in the short term to increase supply. The number of acute care hospital beds has been tripled and the availability of specialist equipment, principally ventilators, has been increased. An agreement with private hospitals has further expanded capacity. 

Demand has been addressed through a combination of measures: sealed borders and quarantine for arrivals; increasing testing; social distancing protocols, isolation, contact tracing; protective protocols for vulnerable institutions such as residential aged care and health facilities; the deferment of elective surgery.

The risks to the health service have been considerably reduced. At the start of the week there were 227 COVID-19 patients in hospitals around the nation, 79 of who were in acute care wards. By Friday these numbers had dropped to 192 and 60.

One figure that is under-reported is the number of active cases, those of have tested positive but have neither died nor yet recovered.

On April 4 there were 4935 active cases. On Friday there were 2639, a drop of 54 per cent in just two weeks.

The danger is not yet over. Like previous pandemics, notably that of the Spanish flu of 1918-1919, there is a high risk of subsequent waves of infection in the absence of collective immunity.

Having put these beds in place, however, the policy focus needs to shift to measures that will prevent them being occupied while adjusting social distancing provisions to allow people to return to work.

Fortunately we now have many more tools at our disposal for enable this to occur. The investment in contact tracing and the introduction of a contact tracing app will enable us to contain any future outbreaks far more effectively than we could have just a month ago.

The introduction of faster testing will also allow freer movement. The airline Emirates has begun carrying out COVID-19 blood tests on passengers prior to departure with results available within 10 minutes. The accuracy of the tests has yet to be verified, but it would at least allow a greater degree of confidence that in-bound flights to Australia were virus free.

Social distancing is now entrenched behaviour, which might allow restaurants and cafes to reopen with customers suitably spaced.

The Menzies Research Centre has been researching ways in which the most vulnerable can be given enhanced protection during the period of opening up.

The relative success of isolation procedures at aged-care homes compared to other countries has shown a path forward. People aged over 70 would be strongly advised to stay at home and limit their contact with other people to relatives, who would in turn be conscious of their higher responsibility to avoid the virus.

The dire predictions at the start of the crisis are evidence of the limits of computer modelling, a tool applied far too carelessly in today’s policy world.

No amount of computing power can compensate for the limited knowledge of the human beings who select and input the data. 

In the case of COVID-19, as with climate change, one vital factor was unaccounted for in the computer modelling that will be decisive in bringing the crisis to an end. Call it the X factor if you will: human ingenuity.

COVID-19: Read the MRC’s coverage of the debate in Australia and around the world