Category Health

We’re not doing nearly enough on COVID-19 and time is running out

The Australian Government is moving in the right direction. The Prime Minister announced this morning a ban on non-essential indoor gatherings of 100 people or more, in addition to the ban on outdoor gatherings of 500 or more. The travel ban has been extended to all countries. Visits to aged care facilities are restricted. The importance of spatial isolation has been reiterated.

But results from these sensible measures won’t be evident immediately. In fact, as the Figure below shows, the coronavirus continues to grow rapidly in Australia. We are still on the ‘scary’ part of the curve. As of yesterday, we had 449 confirmed COVID-19 cases. Italy had about the same number of cases on February 26, and now has more than 28,000.

We don’t know how the number of Australian cases will grow in the future. Spain, Iran, and Italy stayed roughly on the path of doubling every two days (shown above). France and the UK doubled every 3 days. Singapore has managed to slow its growth, with cases now doubling every 10 days. In the past week, Australia has doubled its cases every 3-to-4 days.

The actual number of people infected with COVID-19 is likely to be larger than we know now. It takes time for somebody to contract the virus, develop symptoms, get a test, and then get diagnosed. The Figure below shows the number of new cases each day in China between January and early February. The researchers, Wu and McGoogan, used detailed patient records to establish when symptoms first became apparent (in red) and when people were diagnosed (in orange).

Chinese cities were put into lockdown from 24 February. By then, the ‘real’ rate of new infections was five times higher than newly diagnosed cases. The lockdowns slowed the growth in people reporting symptoms almost immediately, but official numbers continued to grow for 7-to-10 days.

The impact of the new measures Australia has introduced will take time to show up in official cases. It’s a step in the right direction, but we are unlikely to see the flow-through to lower case numbers for another week.

Co Authors :

Grattan Institute is sending its staff home. Here’s why other employers should urgently consider doing the same

Grattan Institute’s 30 staff will be working from home from Monday. We thought we’d tell you why, because our reasons for closing our inner-Melbourne office may help other employers – and employees – who are grappling with this very difficult decision.

The less people are physically near each other, the lower the rate of transmission of coronavirus. That’s why ‘social distancing’ – or more accurately, spatial distancing – is a key strategy to slow the spread of the virus. But for the 13 million Australians with jobs, this spatial distancing can affect their ability to work.

We recommend businesses that can feasibly allow staff to work from home should do so as soon as practical. Here’s why.

As of Thursday 19 March, we estimate there is a 10 per cent chance that at least one employee in a company of 500 people has already interacted with a person with COVID-19.

That’s based on the following assumptions. First, in keeping with research by Imperial College London, we assume that two-thirds of people who contract coronavirus will develop symptoms significant enough that they will self-isolate. That would leave one-third of cases in the community, with only mild symptoms. These people are unlikely to get tested. So we assume that for each two confirmed and quarantined cases, there’s one case in the community.

Second, we assume that workers interact with five unique people in the community each day, such as family members, colleagues, and people in supermarkets or cafes. This assumption might be too high if people have been practising spatial distancing, but could be too low if people are carrying on business-as-usual. We assume the population is ‘well-mixed’ – that is, that Australians are free to move about. Given low levels of interstate travel now, the virus may spread slower in states where cases per capita are low.

Finally, we assume that cases in Australia will continue to grow at the same rate as now – doubling every 3-to-4 days. This is slower than in Italy or Iran, but faster than in Singapore or Japan.

Our estimate does not imply there is a 10 per cent chance that someone in a company of 500 people has the virus, but of course the risk of infection increases with the number of times staff are exposed.

The larger the business, the more likely someone has been exposed. The risks are higher for NSW-based companies (because there is more untracked community transmission there), and lower for companies in other states. And the chances increase with each passing day.

Of course, working from home is not possible for everyone. Some companies’ core business is face-to-face human interaction. The easier it is for a business to allow staff to work from home, the lower the risk of infection which should be tolerated.

While Telstra was able to send 20,000 staff home last week, employees at businesses such as cafes, electrical services, and supermarkets need to be physically present to do their job. These employees should follow hygiene advice and practise spatial isolation as best they can in their jobs. This is especially the case for workers providing essential services: grocers, police, and healthcare professionals, for instance.

And companies that do implement working-from-home policies need to provide staff with adequate assistance to set-up their home, and then provide ongoing support. Companies should ensure that social interaction opportunities continue, albeit virtually.

Community based psychology services should be re-purposed to support people working from home. And if childcare centres close, having more adults at home reduces the need for nannies, potentially freeing up childminding services for parents in the healthcare sector.

Decision-making is hard in the face of uncertainty. But businesses need to act early to protect their staff – and the community.

Co Authors :

As more Australians get COVID-19, will we have enough hospital beds?

The number of confirmed COVID-19 cases in Australia has been doubling every 3-to-4 days. This is not surprising: the slowly-increasing measures put in place over the past week by the federal and state governments will take time to slow the growth of case numbers. But we should be clear: a shutdown of anything that isn’t truly essential will be needed to avoid overwhelming the healthcare system.

In the coming days and weeks, the number of Australians diagnosed with COVID-19 will rise quickly. This will place great pressure on our health services, staff, and infrastructure, including intensive care units (ICUs).

The likelihood that people who are diagnosed with COVID-19 will be admitted to an ICU depends on their age. About 20 per cent of people over 80 will need to be admitted to an ICU, while the rate for 50-to-59 year-olds is 1.2 per cent. Given Australia’s demographic make-up, the overall ICU rate is estimated at 2.2 per cent of diagnosed cases.

The Figure below shows that with Australia’s current rate of doubling of cases every 3-to-4 days, our ICUs will reach current capacity in mid-April. When we hit a trigger point of 12,000 new cases every day, then we know that we will hit our current ICU capacity soon after if new cases continue to grow.

In the scenario of cases doubling every three days, we would reach current ICU capacity on April 11. If cases double every four days instead, we reach ICU capacity a week later on April 18. Slowing the growth to doubling every five days would buy another week.

States are already purchasing additional ventilators to double ICU capacity, but machines need staff to operate and monitor them. Trained staff are not immediately available and so some relaxation of enterprise agreement conditions about staffing may be required during the peak of the pandemic. There are also other patients who will require ICU beds, reducing the number of available beds. But, looking at the Figure above, the only thing that matters right now is the rate of growth.

The initial plans to ‘flatten the curve’ would still lead to more than 100,000 new cases per day at the peak of the pandemic. While this approach will buy us time, we will still run out of ICU beds in Australia.

This will force us to confront ethical dilemmas as to who gets admitted to the ICU and for how long, and who remains in a hospital bed with less intensive treatment. These ‘tragic choices’ that families and health professionals face are the consequences of broader social and political decisions about the toughness of spatial isolation policies. The quicker we can reduce the rate of infection, the better the health system will be able to cope. Older people are more at risk of ICU admission (and death) and so we should be particularly aiming to reduce infection in the elderly.

Our gloomy ICU forecast is primarily determined by the exponential growth in diagnosed cases. This is what needs to change. The goal should be to bring new cases in Australia down to zero as quickly as possible. All state governments must act decisively and bring in a broad shutdown now.

Co Authors :

Australia’s COVID-19 cases are still growing rapidly. Our hospitals may soon hit capacity.

Cases of COVID-19 are still growing rapidly in Australia. We had 2,630 confirmed cases as of 25 March, up from about 700 a week before.

Pressure on health services is mounting. As we showed in a Grattan Blog post on March 24, if cases continue to grow at this exponential rate, Australia will hit capacity in hospitals Intensive Care Units (ICUs) in mid-April.

Major efforts are being made to boost the number of ICU beds across Australia. But the only thing that matters is the number of new COVID cases. If cases double every three days, doubling ICU capacity will only delay hitting the ICU cap by three days. Cases have been doubling every 3-to-4 days for the past three weeks.

While daily case numbers reported by each state are a bit jumpy, the number of new cases in Australia has flattened in the past couple of days. If this flattening trend continues — if lockdowns and spatial distancing are working and our cases numbers move from growing exponentially to linearly — the situation will be very different. But it is too early to tell if this is a genuine slow-down.

Australians do appear to be changing their behaviour in response to the spatial distancing and quasi-lockdown measures imposed by the state and federal governments over the past weeks. The chart below shows that the number of trips taken by people Melbourne and Sydney has fallen to about 25 per cent of normal.

But the number of trips has fallen further in other international cities with tougher spatial distancing restrictions. It is hard to believe that these Australian travel patterns reflect compliance with an injunction to stay at home except for essential trips.

Australians must be more compliant with spatial distancing measures. Analysis released this week by Chang et al (2020) shows that compliance rates of 80 or 90 per cent are required for spatial distancing to have the desired effect. The chart below shows that at even at 70 per cent compliance, the number of infections will continue to grow rapidly.

The spread of COVID-19 cases slowed in China about 10 days after strict lockdowns were put in place. Italy fully locked down two weeks ago and has just started to record a decline in new cases.

Whether the measures Australia has put in place will slow the spread here will not be known for about 10 days. By then, if  our COVID-19 case numbers continue to rise exponentially, our ICU beds will have started to fill up. And, by then, any further lockdown measures will take another fortnight to have an effect. It will be too late.

We must continue to clamp down on the spread of COVID-19.

Co Authors :

Why we’ve downloaded the COVIDSafe app

COVID-19 COVIDSafe app on an android phone

About 1.9 million Australians – about 8 per cent of us – downloaded the COVIDSafe app in the first 22 hours of its release.

The speed of the uptake may have exceeded Health Minister Greg Hunt’s expectations, but it is still well short of the 10 million users the app will need to work effectively. And while early adopters are a promising sign, it is no guarantee of mass acceptance of the app.

In Singapore, 1.1 million users had downloaded their TraceTogether app (on which the Australian version is based) in the first month; but half of these downloads came within the first 24 hours. If Australia follows the same path, we will have around 4 million users – 16 per cent of the population – by the end of May.

Whether the Australian Government can convince more of us to download the COVIDSafe app depends on its actions, through legislation, its communication, and its openness.

If you download the COVIDSafe app, which contains a unique encrypted key, it sits silently on your phone. When your phone comes into contact with another phone carrying the app, they exchange keys. If you are subsequently diagnosed with COVID-19, a health professional will give you a PIN which you can choose to enter into the app. If you give your consent, a list of encrypted keys will be uploaded to your state or territory health authority, so they can more easily trace people you have been in contact with over the past few weeks. Nobody, including the health professional or you, can access the contacts through the app on the phone.

Wide use of the app will help the fight against spread of the virus in Australia, because it will help our health authorities to trace the contacts of people with COVID-19 more quickly and comprehensively.

It is noteworthy, and pleasing, that the app has received broad support from the health sector in Australia, and from the Australian Privacy Commissioner.

Technology professionals have decompiled the Android version of COVIDSafe and explored the source code, finding that the app does what it says it does. Minister Hunt announced this morning (Monday 27 April) that the source code for the app will be released in a fortnight. This is a welcome step towards greater openness.

There are two main privacy concerns with the app.

Firstly, the Federal Government has an ‘anything it takes’ attitude to use of personal information in pursuing people who are critical of it or its policies. This includes a track record of leaking personal information, despite there being legislative safeguards against this. Within the past fortnight, there was an apparent leak from the Department of Health against the prominent ABC health journalist Dr Norman Swan, who has criticised the Government’s pandemic response. It is understandable that, despite all the talk of strong safeguards, many Australians will remain concerned about potential misuse of their personal information.

The second concern is not with the app or the legislation as they stand now, but as they might evolve. Fear of ‘function creep’ – where data is used beyond its original purpose – is valid. Indeed, the Federal Department of Health has acknowledged ‘public concerns that information collected by the app will be used for purposes other than contact tracing, including law enforcement’.

The Government will introduce legislation in the next parliamentary sitting week to ‘establish a strict legal framework’ for use of the application’s data. This will go some way to assuage concerns about function creep. But more needs to be done.

Each of us will have to weigh-up these concerns against the undoubted benefits that the app provides. The benefits of the COVIDSafe app will depend on the number of Australians who use it. The app brings traditional public health contact tracing in Australian into the 21st century. Fast, accurate contact tracing, combined with increased testing capability, will help health professionals minimise the spread of COVID-19, and respond quickly to future flare-ups. It may enable people to safely return to school, university, or work sooner.

Despite the risks, we have decided to download the app as part of our support for sensible public health measures to reduce COVID-19 infections in Australia.

Co Authors :

How children get and transmit COVID-19 is still a mystery

Small child in window of house with facemask COVID-19

Despite the unequivocal assurances provided by the Deputy Chief Medical Officer and the federal Education Minister, much about COVID-19 remains a mystery. There’s still a lot we don’t know about how the virus affects children. That means the stakes are high when deciding whether kids go back to school now, either full or part-time, or remain at home for now and return at the start of Term 3.

Children can get COVID-19, but we don’t know whether they are less likely to become infected

One thing we do know is that children of all ages can get COVID-19. There have been 271 confirmed cases of children with COVID-19 in Australia so far. Some studies, discussed below, suggested they were less likely to catch the virus than adults, but good recent evidence suggests children may be just as vulnerable.

Under symptom-based testing, such as has been the policy in Australia until recently, asymptomatic people with COVID-19 are unlikely to be tested and diagnosed. So symptom-based testing tells us how many symptomatic people tested positive for COVID-19, rather than how many people have COVID-19. Tests are mostly done on people who appear sick, and asymptomatic people, by definition, do not appear sick. Evidence so far suggests many children with COVID-19 are symptom-free.

To discover the COVID-19 rates for children, people without symptoms need to be tested. This happens in two scenarios: when all close contacts of a confirmed case are tested, and when a random sample of people is tested.

In Iceland, 10,800 asymptomatic people were tested for COVID-19 in mid-to-late March. About 0.8 per cent were positive. Of the 848 children under 10, none had the virus; of the 1200 children aged 10-19, 0.4 per cent tested positive – half the rate of the adults.

In Germany, a COVID-19 lab processed 60,000 tests. Of the 2,200 children under 11, 2 per cent tested positive; of the 1,900 people aged 11-20, 4 per cent had the virus. About 22,000 adults aged 20-40 were tested over the same period, suggesting they were more likely to have symptoms. But only 5 per cent of them were positive for COVID-19, not much higher than the rate for the children.

In a well-designed study, the Shenzhen Center for Disease Control and Prevention identified 391 COVID-19 cases and 1,286 close contacts. It looked at people in households with a confirmed COVID-19 case. The authors found that children were ‘just as likely’ to contract the virus as adults under 50.

That finding runs counter to analysis published a few days later, in which researchers examined the contract tracing information from the CDC in Hunan, China. Contacts to COVID-19 positive patients were placed under medical observation for 14 days. Analysis of people’s susceptibility to the virus concluded that children did have a lower risk of infection.

Children can spread COVID-19, but we don’t know whether they spread it to fewer people

Another thing we know is that children can pass on COVID-19 to adults and other children. But observational evidence so far has shown that they are less likely to spread the virus than adults.

At the beginning of the COVID-19 epidemic in Australia, the National Centre for Immunisation Research and Surveillance (NCIRS) studied 9 adult and 9 child cases of COVID-19 in 15 NSW schools. The study identified 832 ‘close contacts’ – 735 of them children. One-third of this group were interviewed, tested with nasal swabs 5-to-10 days after contact, and had a blood sample examined for antibodies one month later.

One child in primary school tested positive on both the nasal swabs and for antibodies; and one child in high school tested positive for antibodies, but not on the initial nasal swab.

These cases happened in early March, before government recommendations for spatial distancing and lockdowns. Back then, Australia had done very little to reduce the spread of COVID-19. That two children out of 288 tested positive indicates that child-to-child transmission is possible, but suggests the rate of transmission is low.

The NSW study is in line with other observational studies. In a case study of an outbreak in the French Alps, a symptomatic child visited and had ‘close interactions’ in three schools without passing on the virus to anyone. The authors said this suggested ‘different transmission dynamics in children’.

A multinational study of 33 household clusters found that a child under 18 was the initiating contact (‘index case’) for three. The authors of this study note that this is well below otherwise similar infections such as the H5N1 influenza virus, in which children are the index case about half the time.

In the Netherlands, the Ministry of Health studied 54 households with COVID-19 infections and found that while children did become infected, they were never the source of the spread.

Whether or not schools were open at the time of infection is important for studies that examine household index cases. A detailed study of 36 paediatric cases in Zhejiang, China, found that almost all children got the disease from family members rather than the community or other children. But schools were closed for the spring festival holiday during this period, so child-to-child contact was drastically reduced.

One thing we don’t yet know is why an infected, symptomatic child would spread the disease less than an infected, symptomatic adult.

The virology data to date suggest children are as infectious as adults. A study of 3,700 COVID-19 patients in Germany found there was no difference in the viral load – a measure of infectivity – between people in different age groups, including children. The virologists concluded that their findings, combined with the evidence of children’s vulnerability to infection,suggested that the ‘transmission potential in schools and kindergartens should be evaluated using the same assumptions of infectivity as for adults’.

Children with COVID-19 are less likely to become severely ill, but we don’t know whether they suffer long-term effects

We know that children with COVID-19 can become severely ill. But the available evidence strongly suggests they become severely ill at lower rates than adults.

In a comprehensive study of 2,135 paediatric cases in China, more than half had mild (flu-like) symptoms at worst. About 40 per cent had moderate symptoms, such as pneumonia, frequent fever, and dry cough. The remaining 5 per cent were classified as severe or critical, compared to 19 per cent of adults in China at the same time.

While the severe-illness rates for children with COVID-19 are low, the medium- and long-term effects are still unknown. This week the UK Health Secretariat warned of a serious emerging syndrome affecting children, potentially related to COVID-19. There have been similar reports in Italy. Only time and regularly updated research will tell us how serious this is.

We do know that children can and do die from COVID-19. There have been deaths of children in China, the United States, the United Kingdom, France, and other countries with substantial outbreaks. Death rates of children with COVID-19 are very low. But there are 4 million school children in Australia, meaning that in an outbreak, even a low death rate could translate into the deaths of many children.

In the face of this uncertain evidence, in deciding whether to open schools, policy makers have to weigh up the evidence – what is the likelihood of infections, of passing that infection along, and what are the potential health and economic consequences. Because we have seen no child deaths from COVID-19 in Australia, decision makers may be inappropriately ignoring that possibility. We all are prone to ‘optimism bias’ – erring on the positive on all the issues which should be taken into account in the difficult decision.

Although children are much less likely to get seriously ill and die, it is possible this will occur. It is probably the case that they are less likely to transmit the virus than adults, but nevertheless they can be the primary source of transmission. It is wrong for decision makers to pretend that the evidence is clear when it is not. Opening schools is not a risk-free choice and should not be portrayed as such.

There is uncertainty around COVID-19 and its effects on and transmission through children. We are safer if we make decisions while fully aware of that uncertainty, rather than with an unfounded surety. We need to know how firm the ground is under the science that guides our decisions. Only then can we properly assess the risks, measure the trade-offs, and make the tough decisions that need to be made about our schools – and about protecting our children.

Co Authors :

Victorians must all share the lockdown load

Victorians in 10 postcodes across Melbourne’s north and west have been ordered back into lockdown. But regaining control of COVID-19 remains the responsibility of all Victorians.

The virus is spreading in other parts of Melbourne, beyond the 10 ‘hotspots’. These new cases are coming from transmission in the community rather than from overseas arrivals. Victoria has more cases of community transmission now than it did in the initial peak of the virus in March.

Following the early success in flattening the curve, Victorians have mostly – but not entirely – returned to normal patterns of life, as the chart below shows. They’re going back to their workplaces and out to the shops. Far fewer of them, however, are catching the train or hopping on a tram: public transport use remains at about half of what it was in January.

But for the 300,000 people living in the 10 hotspots, life has gone back on hold. Movement in and out of the lockdown zones will be partially controlled by police checkpoints. But suburbs are porous. And people in those suburbs are still allowed to leave the house, for essential shopping, care, exercise and – importantly – work.

So people will continue to make trips to their workplaces outside the lockdown areas. The chart below shows the job locations of the 130,000 workers who live in hotspot areas. Tens of thousands work in neighbouring suburbs. More than 10,000 work in the CBD, and thousands more work in Southbank, Docklands, and Parkville.

Strict social distancing practices in workplaces is therefore crucial if Victoria is to again drive cases down. All Victorians should work from home if they can. If they can’t, they should spend as little time as possible at their workplace.

Employers must do their bit. Businesses must ensure staff can avoid shared spaces and high-risk areas. Handshakes should be a thing of the past. Workplaces should be thoroughly cleaned regularly.

Locking down suburbs that are the source of rapid COVID-19 transmission is sensible. But these lockdowns don’t absolve other Victorians from their social distancing responsibilities. The virus still lurks. There are hundreds of active cases in the state, many of them outside the hotspots.

Grattan Institute modelling has shown that it is people’s behaviour when they are out in the community that will determine Victoria’s and Australia’s COVID-19 future. If people drop their guard while the virus is still around, it will spread with rapid and devastating effect.

Anyone who could come into contact with the virus bares responsibility for containing it. Right now, that means all Victorians.

Co Authors :

Tracking Victoria’s COVID response

The daily announcement of new COVID cases has become part of the rhythm of life for Victorians. It can be an unsettling announcement, especially over the past fortnight as a ‘stubborn tail’ casts doubt over the state’s ability to ease restrictions as soon as hoped.

The headline number attracts a lot of attention, as do the 14-day moving average and the 14-day ‘mystery case’ total, since these are the metrics used in the State Government’s roadmap to a ‘COVID normal’. But they are not the only useful numbers to keep an eye on. When the total caseload is so small, it is also useful to distinguish between cases from a known source, cases which are successfully traced, and cases which remain a mystery. These provide a perspective on contact tracing efforts.

This information is provided in daily media releases and in statements and tweets by the Victorian Chief Health Officer. The chart above presents it in a simple visual form.

The chart makes a few things apparent. First, there have been 10 or fewer cases requiring investigation each day in the past fortnight. Second, most of those investigations have been successful: about 90 per cent of cases have been traced. Third, there is still a small but significant number of mystery cases. These cases indicate gaps in our knowledge about where the virus has spread, and those gaps are riskier the fewer restrictions are in place.

No single metric should be taken as the perfect measure of progress. Monitoring the types of cases, not just the total number, offers a more vivid picture of the trends that will shape the coming weeks and months.

Co Authors :

The vaccine rollout – going well according to which plan?

The increasingly strident and wide-spread criticism of the COVID-19 vaccine rollout is a problem of the Commonwealth Government’s own making.

During 2020, Australians came to believe that the public health strategies – lockdowns and restrictions, testing, tracing, and isolation – were interim until the vaccine arrived. They came to believe that because the Government encouraged them to do so.

From August, the Government has celebrated every minor milestone in the vaccination journey, contributing to the expectation that vaccines would be the game changer. Unlike other countries, the coronavirus is not circulating in the Australian community so it was appropriate to defer the vaccination rollout to help ensure it went smoothly, but this was a hard sell to the community

The first vaccinations were given with much hype and nationalistic symbols, and the public expectations were heightened. And then the wheels fell off the bus.

To be sure, some of the problems were due to the nasty Europeans reneging on some vaccine delivery promises – a risk that could and should have been foreseen by Australian officials. But others failures in Australia’s vaccine rollout were due to failures in the local supply chain.

The target for the number of Australians to have received their first vaccination by the end of March was revised down, from 4 million to 2 million, to take account of the international supply chain problems. But only about a third of that was actually delivered.

Back in mid-February the Government said 500,000 aged care residents and staff would be vaccinated ‘in the coming weeks’, with so far only one fifth of that target met.

Australians were told that CSL would produce 1 million doses a week in Australia, but so far it is falling well short of that, although how far short seems to be a state secret.

Yet an increasingly sceptical public keeps being assured everything is going according to plan.

The truth is that there is currently a huge gap between the Government’s vaccination promises and its vaccination delivery. Both elements of that need to be addressed. Delivery needs to be improved – including through mass vaccination centres – but community expectations also need to be re-calibrated.

If the rollout is going according to plan, then 600,000 doses by the end of March must have been the secret target, not 2 million. Perhaps the Government could tell us. And the issue now is what should be the target for end of May, end of June, and so on. What proportion of the so-called Phase 1 population – quarantine and border workers, healthcare workers, aged care and disability care staff and residents, people aged 70 or older, Aboriginal and Torres Strait Islanders aged 55 or older, adults with an underlying medical condition or significant disability – will be vaccinated by when?

The Government needs to be clear about what its plan is, taking into account both the local and international supply chains, and the capacity of local delivery channels.

The Government should publish a realistic vaccine rollout plan to which it is prepared to be held to account, and to publish progress against that plan – including doses produced, doses acquired, and people vaccinated. And it should publish that information each week.

It is time for the Commonwealth Government to level with Australians, so we can better understand what the plan is and whether everything really is going to plan.

Co Authors :