Coronavirus pandemic: sceptical question marks make for better policy than excitable exclamation marksMar 30, 2020
When did the world’s media and politicians become collective versions of Lance Corporal Jones in the British comedy series Dad’s Army, screaming ‘Don’t panic! Don’t panic!’? Colour me contrarian, but since the 2003 Iraq war, my working motto has been: when you come across excitable exclamation marks, substitute sceptical question marks and you’ll be right.
According to World Life Expectancy, Australia’s annual death toll is around 170,000. Heart diseases kill more than 23,000 people. The number of deaths from flu and pneumonia is over 4,000 and from road fatalities around 1,300. Averaging the total, about 40,000 Australians will have died as at 26 March. As of that date, 12 Australians had died from the coronavirus affliction.
Led by the media, panic set in to drive public policy before data was collected to show that, for most people, the Covid-19 virus will be worse than seasonal flu. The crisis escalated with an Imperial College study of 16 March which described Covid-19 as ‘a virus with comparable lethality to H1N1 influenza in 1918’ (the Spanish flu that killed upwards of 50mn people, one-sixth to one-third of them Indians). Without an aggressive suppression strategy of prolonged lockdown, it could cause 0.5mn deaths in the UK and 2.2mn in the US. This caused PM Boris Johnson to switch dramatically from the initial herd immunity strategy – let the virus spread through the community because the mortality rate is well within the parameters of a seasonal flu – to a drastic nationwide lockdown. Other governments were soon infected by the panic virus and the favoured strategy became to prioritise citizen’s health over the nation’s economy. As a total lockdown leads to economic shutdown, the pain is alleviated through industry bailout and social benefit packages.
To kill the virus, must we kill the economy?
In earlier flu epidemics (Asian Flu 1956–58, Hong Kong Flu 1968, SARS 2003, avian flu 2008, swine flu 2009), the numbers infected and killed were sufficient to produce a severe impact on society. But governments didn’t shut down their country, destroy the economy nor jeopardise their way of life. People suffered but endured. This too shall pass.
In responding to an epidemic, there is a trade-off between public health and economic stability. It is the duty of health professionals to focus solely on the former. It is the responsibility of governments to balance the two and intuit the social fulcrum: the sweet spot at the intersection of dangerous complacency, alarmist panic and reasonable precautions. The injunction to first do no harm implies that governments should be wary of prolonged economic lockdowns: the cure might indeed be worse than the disease.
Public policy must be based on a balance of risks and benefits, often amidst uncertainty, incomplete information and unintended and perverse human, health and economic consequences. The health of citizens and the health of the national economy are closely connected and interdependent. A healthy economy requires a healthy workforce. Conversely, most wealthy industrial countries have greatly superior public health systems and outcomes because a strong economy makes it possible to invest in the entire infrastructure of health delivery. Supply chain disruptions can elevate the risks of death from other potentially life-threatening ailments, including heart problems, diabetes, chronic lung disease and overlap between these (co-morbidities).
In the US, people at risk from these illnesses total 70-80mn. At 1% excess fatality in this group caused by shortages resulting from the economic shutdowns, another 750,000 Americans will die. On the other side of this necessarily grim ledger, for those who are not elderly, after infection, the mortality rate is about equal to the risk of dying from all other causes. In addition, layoffs and income losses will cause a spike in long-term mental health problems and suicides. Australia has already begun developing a major mental health package to address heightened anxiety caused by job losses, wage cutbacks, isolation and loneliness. Prolonged house-bound isolation could also see a surge in domestic violence cases.
Meanwhile the rushed returns home from overseas in confined planes to crowded immigration lines, the panic buying in supermarkets to hoard up and the long queues at CentreLink offices have contributed their share to the cross-infection.
Modelling is assumption-driven and data-limited
Was the panic caused by flawed, inflated mortality statistics? The original Imperial College paper used sketchy data and heroic assumptions. More sceptical voices have emerged since then. One team finally got around to testing an entire village of 3,000 people near Florence for the SARS-CoV-2 virus which causes Covid-19. The results show that 50%-75% of those infected were asymptomatic and thus would not be tested in most jurisdictions. This is fully consistent with an Oxford modelling that maps a range of different outcomes using different assumptions.
The limited early observational data suggest that the true infection rate may be thirty (Wuhan) or hundred (Italy) times higher than the number of confirmed cases, which would drop the fatality rate, of the total infected population, to one-thirtieth or one-hundredth. On the basis of the existing albeit inadequate data, Eran Bendavid and Jay Bhattacharya, professors of medicine at Stanford University, conclude that the Covid-19 mortality rate could be as low as 0.01% of all those infected, one-tenth the average flu mortality of 0.1%: ‘Such a low death rate would be cause for optimism’. Hence their conclusion:
A universal quarantine may not be worth the costs it imposes on the economy, community and individual mental and physical health. We should undertake immediate steps to evaluate the empirical basis of the current lockdowns.
According to the RealClearPolitics running tally, as of 29 March, the total deaths were 30,627 from a total confirmed cases of 659,435. Italy had the highest number of fatalities at 10,023. Among countries with a recorded number of deaths over 100, the confirmed cases per million people varied from 4.3 in Indonesia, 18.6 in Brazil and 58.4 in China, to a high of over 1,500 each in Italy and Spain. The fatality rate compared to confirmed cases similarly varied widely, from 0.75% in Germany, 1.67% in the US and 4.05% in China, to 8.83% in Indonesia and 10.84% in Italy. The extreme variability of all these merely confirm the widely disparate standards of health-related surveillance and data collection around the world. Reliable data are simply not available of the total numbers of those actually infected and those dying from Covid-19.
According to the Centers for Disease Control and Prevention(CDC), as of 28 March, the total US Covid-19 cases were 103,321, of whom 1,668 people had died. By that date, according to WHO, the total global deaths from Covid-19, the first case of which was reported to WHO on 31 Dec 2019, was 26,654. Meanwhile the CDC data show that in the first ten weeks of 2020, 40,149 Americans had died from influenza and pneumonia. On 25 March Neil Ferguson, lead author of the fateful Imperial College report, gave evidence to a parliamentary select committee inquiring into Britain’s response to Covid-19: ‘UK deaths from the disease are now unlikely to exceed 20,000… and could be much lower’. Moreover, between 50%-67% of them would likely have died within one year anyway owing to old age and co-morbidities.
Between them, these examples show firstly the perils and pitfalls of modelling, and secondly the problems and risks of basing public policy on models before sufficient data is available to test their assumptions and conclusions.
In the meantime we have imposed devastating social and economic costs. The sensible strategy would have been to isolate the elderly and vulnerable and let everyone else get on with their lives. Dr. David L. Katz of Yale University, an expert in public health and preventive medicine, calls this a ‘vertical interdiction’ strategy to distinguish it from the ‘horizontal interdiction’ strategy of mass social distancing. The end result of such a strategy, to which we can still pivot, would be to achieve herd immunity for the population at large, but without the social and economic costs that will be with us for some time, with significant long term physical and mental health damage as well.
Influenza and pneumonia cause 3.2mn deaths worldwide annually. If each individual case around the world was tallied daily on the front page of all newspapers, a similar panic would arise. Road accidents kill another 1.3mn annually that could be saved by banning all automobiles. Sensibly, we hold the resulting disruption to everyday life to be too high a price to pay. Similarly, life and death is part of an eternal cycle. Take sensible precautions, carry on calmly and put in place emergency measures only after the facts are in. But retain built-in flexibility to adjust policy interventions continually in line with the threat’s evolution, spread and gravity; the collection and analysis of facts and evidence; and the availability of diagnostic tools and preventive and therapeutic interventions.
To be clear, we may yet get to the stage where truly draconian measures are necessary to control a Spanish flu-like epidemic. But as of now, the limited data we do have don’t seem to support such extreme measures. Where most Western governments proved negligent, as Ian Johnson has argued, was in failing to use the time after the outbreak in China to build up their response capacity across the board in order to be able to scale up emergency measures as required, if and when an epidemic hits.
Balancing risks, costs and benefits
The first balance a government must assess is the risk of creating mass hysteria and panic with premature reporting and the risk of losing control by delaying public announcements of the true scale, gravity and urgency of a nascent epidemiological emergency. They can then settle on the optimal balance between sufficiently slowing the disease, preventing an economic meltdown and maintaining a functioning society while the threat and responses evolve and the virus spreads.
In retrospect, China got it wrong but it’s not clear most governments would have acted sooner. With the benefit both of hindsight and of the experience of scrambling, shambolic responses by so many Western governments to play catch-up with the rapidly transmitting and evolving threat, China deserves credit for having mobilised quickly, efficiently and effectively to confine, contain and defeat the epidemic. Its greater risk now is importing the virus!
Japan, South Korea and Taiwan have shown how democracies can manage epidemiological crises without sacrificing the economy: wear protective masks and avoid physical touching and proximity outside the home: the Indian ‘Namaste’ is a simple, elegant, cordial yet respectful alternative to the ubiquitous handshake; wash or sanitise hands frequently; check temperatures at airports, seaports and on entry into crowded areas like train stations, office complexes and malls; test those with elevated temperatures; hospitalise those testing positive; quarantine those with symptoms at home with random inspections to ensure compliance; and trace and isolated those with whom they have been in contact.
I write as someone in the high risk category of over 70 with cardiovascular co-morbidity. I have no wish to live for an extra year or two at the cost of crushing my children’s careers, plans and dreams.
Ramesh Thakur is emeritus professor, Crawford School of Public Policy, Australian National University