There is a common belief that whatever measures are taken to reduce the spread of coronavirus, they are unlikely to reduce the ultimate death toll. But well-designed protective policies could save many lives, as well as buying time for pharmaceutical innovations to counter Covid-19.
In a recent opinion piece in The Lancet, the controversial former chief epidemiologist of Sweden Johan Giesecke boldly predicted that: ‘There is very little we can do to prevent this spread: a lockdown might delay severe cases for a while, but once restrictions are eased, cases will reappear. I expect that when we count the number of deaths from Covid-19 in each country in one year from now, the figures will be similar, regardless of measures taken.’
In Shakespeare’s The Merchant of Venice, the merchant Antonio owes Shylock a pound of flesh, a terrible and unfair, but inescapable debt that cannot be absolved. Is Giesecke correct and does humanity owe a pound of flesh to coronavirus?
There is a widespread belief – if not always clearly articulated – that when infection is suppressed, what you lose on the swings, you gain on the roundabouts. It is often taken for granted that the cost of suppressing infections today is an increase in future infections and that the overall number of infected (and deceased) people across the entire epidemic is an immutable constant, a ghoulish debt owed to nature.
According to this view, controlling the spread of the disease is a fool’s errand, as those who will die will die – and there is therefore no reason to disrupt society and the economy more than necessary.
Many strategies have been proposed to contain the spread of the disease (see, for example, Moll, 2020, for a number of different options). The different policies adopted in practice by different countries have been hotly debated, with two main opposing standpoints emerging.
One strategy, called the laissez-faire strategy or the herd immunity strategy (although this is a misnomer, as explained here), contends that the disease burden of Covid-19 is somehow inevitable and that all that lockdowns can do is to postpone the pain. This is a position espoused not only by leading politicians, but also by leading epidemiologists.
The other strategy – the managed epidemic strategy – instead employs different measures of disease control to shape the path of the epidemic. Most famously, one aim of a managed epidemic is to ‘flatten the curve’ – that is, to reduce peak prevalence in order to ensure that health services such as the UK’s NHS are not overwhelmed.
In a recent interview, prominent epidemiologist Sunetra Gupta stated: ‘I think the lockdown simply delays things, it doesn’t solve the problem at all, it just delays it’. Similarly, Martin Kulldorff, another of the signatories of the so-called Great Barrington Declaration, said in an interview that: ‘trying to suppress the disease with contact tracing, testing and isolation, together with severe lockdowns, is not going to solve the problem. It will just push things into the future.’
Why do these beliefs matter? They matter because such ideas have taken hold with very powerful people who make decisions about life and death. The president of Brazil, Jair Bolsonaro, has quite explicitly based his hands-off approach to disease control on the notion that the effort is futile. After himself catching Covid-19, he stated: ‘I knew I was going to catch it someday, as I think unfortunately nearly everyone here is going to catch it eventually. What are you afraid of? Face up to it.’
In other words, if one believes that a policy cannot be effective, then it will be discarded out of hand. This is doubly so with very economically costly measures such as lockdowns.
The pound-of-flesh fallacy is only a statement about the narrow technical effects of social distancing – that is, about whether suppressing the disease is likely to decrease the overall burden and mortality from the disease. It is not a statement about the ultimate desirability of such measures.
In principle, it is perfectly possible to find that lockdown measures do in fact reduce the burden of infection, but that they are not worth pursuing. This would be the case, for example, if the effects of the disease were minor or if the economic consequences of such measures greatly outweighed the benefits. But this type of trade-off merits an entirely different conversation.
To illustrate the effects that suppression of the epidemic can have on the overall number of infected and deceased individuals over the course of the epidemic, Figure 1 shows the simulated trajectory of infected people over a six-year period in a simple variant of the classic SIR epidemic model, calibrated to the US population, which currently stands at approximately 330 million people (see Giannitsarou et al, 2020).
The biological details of the model – for example, the infectiousness of the disease, the infection fatality rate and the recovery rate – match the best available evidence to date (see the paper for detailed description and references).
Figure 1: Infections and optimal restrictions compared with an uncontrolled epidemic
In the upper panel, the dotted line shows the path of disease prevalence in the baseline epidemiological model without spontaneous behaviour or suppression. The solid line shows the path of disease prevalence under a policy that chooses suppression to trade-off costs and benefits, including both economic costs and the disease burden from infection. In the lower panel, the path of the optimal level of restrictions is plotted. These can range from zero restrictions (a laissez-faire policy) to full restrictions (a complete lockdown of society).
As is clear from the plots, optimal suppression has two distinct effects on the path of infection:
- First, it lowers peak prevalence substantially, an effect popularly referred to as ‘flattening the curve’.
- Second, it shifts the entire disease path into the future, thereby postponing the peak of the epidemic. As can be seen from the plot, this may in fact slightly increase the peak of the second wave relative to the do-nothing scenario and so be interpreted as causing the epidemic to be somewhat prolonged.
But crucially, these two effects do not cancel each other out.
At the end of the six-year period of this simulation, the disease settles on an endemic steady state, whether it is suppressed or not. Under the laissez-faire strategy, a total of 9.6 million people would have died from the disease. In contrast, under optimal suppression, a total of 8.2 million people would have died. In other words, a correctly designed suppression policy would save 1.4 million lives in six years from the start of the epidemic.
Although these numbers may seem high, Swanson and Cossman, 2020, show that reported numbers may have to be scaled by a factor of 14 to get an accurate picture of the actual disease prevalence in the population. In addition, these simulations assume that immunity to re-infection with coronavirus wanes, as recent evidence suggests is very likely to be true.
We note that this policy is not formulated with the aim to reduce infection or mortality specifically, but rather to trade-off costs and benefits of suppression. If policy-makers want to reduce infection and mortality further, that is certainly also possible and would only strengthen the result.
These findings are robust to a number of enhancements of the basic framework, as can be seen from the work of the Imperial College report (Ferguson et al, 2020), which formed the basis for much of the UK’s policy for managing the epidemic.
But we should not forget that there are potentially other reasons why it may be good policy to suppress disease spread aggressively in the early stages of the epidemic, which fall under the category of buying time. These reasons are distinct in nature from the one outlined here, but they are nonetheless very important in their own right.
If the epidemic is one of a new strain of infectious disease, as Covid-19 is, much about it will be unknown initially. At the start of the epidemic and with limited data, it is very difficult to determine the infectiousness of the disease, the long-term health consequences of infection, which individuals are most at risk, what the infection fatality rate is and whether people can become re-infected once they recover. Suppressing the spread of infections allows policy-makers to develop a better understanding of all these features of the disease to design and to fine-tune control measures.
Moreover, postponing infection gives time for pharmaceutical innovations, such as antiviral therapies and vaccines, to become available. Again, suppressing infection at the early stages may allow more people eventually to benefit from such innovations – and to survive (a prospect that is explored further in Makris and Toxvaerd, 2020).
Where can I find out more?
- Herd immunity – crucial yet irrelevant: Flavio Toxvaerd and Robert Rowthorn argue that, if feasible, herd immunity will be indispensable for overcoming the epidemic. But when formulating optimal health interventions, society should adopt a holistic approach and carefully trade-off all the different aspects that affect social wellbeing, including both health and economic outcomes.
- Waning immunity and the second wave: Some projections for SARS-CoV-2: Study by Chryssi Giannitsarou, Stephen Kissler and Flavio Toxvaerd.
- Great expectations: social distancing in anticipation of pharmaceutical innovations: Study by Miltiadis Makris and Flavio Toxvaerd.
Some non-technical, general interest introductions to the economics and policy of infectious diseases (some written pre-Covid-19) written by Flavio Toxvaerd for the Bennett Institute for Public Policy are available here:
- From epidemiology to economic policy (on Covid-19 policy).
- Thought experiments – can economic models help disease control?
- COVID-19 policy must take all impacts into account: Charles Manski argues that human health is obviously crucial, but epidemiological models should not ignore economic and ethical considerations.
Who are experts on this question?
- Flavio Toxvaerd, University of Cambridge
- Chryssi Giannitsarou, University of Cambridge
- Ben Moll, London School of Economics