How To Break the Covid Doom Loop
Yves here. While this proposal sounds more realistic than other schemes on offer, I still wonder if it would work all that much better than the shambolic approaches in countries and communities where Covid has settled in. And on top of that, we have the sobering spectacle of countries in Europe that had gotten infection rates down now seeing troubling upsurges.
In other words, I keep coming back to the notion that the disease is in control and it takes pretty stringent measures to lower contagion rates. This proposal does have the appeal of reducing uncertainty, since the populace would know in advance when restrictions would be in place.
However, I regard the authors as way too optimistic about the effect of short stringent lockdowns (two weeks??). And even if the weight of the law were put behind these two week lockdowns, it seems likely that the effect of “off-on” lockdowns is that people will simply crowd dangerous activities in “on” period, rather than spreading them out more over time. Is that really any sort of gain? I also see the authors as way too optimistic about contact tracing, which seems to work only when you have free, fast, and readily available tests, a citizenry is willing to respond quickly to tracing inquiries, and a pretty low baseline of infections.
And where is the income support?
The big problem, particularly in the Anglosphere, is not enough people are willing to follow rules. Too many people equate putting others at risk and being a scofflaw as an inalienable personal right. The authors acknowledge that “compliance is key” but don’t have new ideas here.
By Anne Williamson, a health economist working with COVID-19 Statistics, Policy Modelling, and Epidemiology Collective, Scott Sheffield, Professor of Mathematics at MIT, specializing in the theory of probability, and Anna York, a Postdoctoral Associate at Yale School of Public Health. Originally published at openDemocracy
What we are doing simply isn’t working. There have been countless individual sacrifices; relatives not seen since Christmas and restaurants shuttered forever. As cases spiral alarmingly again, we are moving back towards economically devastating restrictions or, worse, thousands more patients struggling to breathe in intensive care units.
Can we take hope from the new strategy of ‘circuit breaker’ lockdown? Yes – but only if done correctly.
From our vantage points by the bedside on COVID wards, and as part of the COVID-19 Statistics, Policy Modelling, and Epidemiology Collective, we saw firsthand how we failed before. We argue that a coordinated circuit breaker followed by intermittent and contingent ‘reset’ periods could avoid a longer lockdown, help the economy, keep a handle on COVID, and save lives.
This would mean imminent strict regulations for two or more weeks to reduce the disease reproduction rate, R, below 1. Then in the subsequent months, intermittent repeated resets would allow increased activity overall whilst remaining below this safe threshold.
A Failed Approach
It is pivotal we understand how we arrived back at this precipice. The current ‘moderate distancing’ approach walks a dangerous tightrope, aiming for social and economic activity up to just before the point where disease rates would explode. It then resorts to disordered de-facto lockdowns, like we are currently seeing, but only as a reaction to a situation that is already out of control. It is doomed to fail for multiple reasons.
First, moderate regulations are prone to behavioural slippage. Early on, people largely followed the rules. However, when one or two friends push the boundaries and squeeze seven onto a table, or a high profile figure is compelled to explore Barnard Castle, others relax their internal standards in unravelling policy chaos. If we don’t change course we thus face an interminable loop of moderate restrictions punctuated by kneejerk lockdowns that fail to effectively suppress the disease.
Second, it requires excellent testing to stay on the tightrope, and constant policy tinkering as disease rates evolve. If recent weeks have revealed anything, it is the inability of current Test and Trace to cope with higher rates of disease.
Third, it leads to disparities in infection rates, which make sacrifice inefficient. Many towns and subpopulations have already sacrificed more than enough to eliminate internal spread, but a constant threat of outside reinfection prevents them from relaxing and enjoying their triumph.
Finally, this approach still requires months of major adjustments to everyday life, and rather than minimising cases, it simply spreads them out. Even with all this sacrifice, there is no clear point of ‘success’ or increased freedom until a vaccine is found.
Breaking the Doom Loop
In recent weeks, discussion of ‘circuit breakers’ lockdowns have made the headlines, with everyone from Keir Starmer to SAGEcalling for a snap policy to regain control. Yet there is little discussion about what a good circuit breaker would look like and even less about what must happen afterwards.
In a recent paper, we considered exactly this question. Suppose a town wishes to allow 1,000 small but risky activities like essential workplace meetings, haircuts, and small social gatherings over the course of a month. Is it better to put all the activities in the same fortnight or space them out evenly? The mathematical answer is clear: we should take the former strategy. Temporally concentrated sacrifice beats temporally spread out sacrifice.
The exact optimal way to cluster these activities in time then becomes a policy question, not just a mathematical one. In practice, business and social activities could be permitted during the second fortnight of each month from November to January, but curtailed during the first fortnight. This would immediately lead to better disease control and more economic activity than the present approach. The resets might mimic what we had in early June: multiple households only meeting outdoors, with financial support for businesses that suspend trading. We emphasise that some economic activity could continue even during these periods: thanks to contactless supply chains, PPE, and other innovations, many businesses could safely remain open. But safety would be paramount. The goal is to drive the as low as possible over two strict weeks; the lower the disease rate, the less costly and more feasible a Track and Trace system becomes.
After this circuit breaker, relaxed periods could resemble current rules where masks and sensible caution coexist with increased social freedom. The advantage of intermittency is that it is flexible but effective: at first resets would be scheduled in advance, but as the situation improves the resets may become less strict. A highly effective circuit breaker could even eliminate the need for subsequent resets (as demonstrated in New Zealand), though this would most likely require a larger initial sacrifice than the UK is prepared to make.
So, would an intermittent strategy work? The maths shows better outcomes, but more importantly, it makes sense. Disease spread comes from total cases multiplied by the growth rate. A circuit breaker means many infected people will recover, suppressing total cases, so we can then enjoy several weeks of careful socialising and work activity. Intermittency also matters, because just when transmission starts to rise, we would have a preorganised reset to reduce disease prevalence before it gets out of hand again.
Over several months, rates could be driven towards zero without a continuous and interminable moderate lockdown. After cutting oxygen off from a fire to reduce it to embers, restoring airflow intermittently doesn’t allow large flames. Compare this to a steady continuous oxygen supply, and the blaze quickly rages out of control.
Compliance Is Key
Of course, there are caveats. Not all activities can be shifted in time, so inflexible activities like urgent medical care must continue. Since a month’s worth of flexible activity will be compressed into two weeks, businesses may have to extend opening hours to maintain appropriate social distancing. Still, the benefits of policy fluctuation remain even after the model is tweaked to allow for this.
However, one tweak causes particular trouble. If the model involves multiple subpopulations, and one of these refuses to follow the initial circuit breaker or subsequent resets, it will export infection and make it impossible for its neighbours to cut off the oxygen from the fire. This could occur between geographic areas, or between groups in society – precisely the concern about university students returning to their hometowns. The worst case is a lockdown in which groups with low disease prevalence must take even more precautions, while those unwilling or unable to lock down continue as before. It is critical that the highest prevalence populations have both the rules and resources to drive down and nearly eliminate the disease. Once we’ve decided upon circuit breakers, coordinated sacrifice beats uncoordinated sacrifice.
Thus the final key question is a behavioural one. Will the public comply? We would have to be crystal clear, with effective early enforcement, to avoid people carrying the freedom of the relaxed weeks into the reset periods. Nonetheless, it isn’t hard to explain: if you’re born in November, you can have your small birthday event – but only after the 14th. This may actually increase compliance, because of the psychological appeal of regular social “rewards” for committing to strict measures. It also allows affected businesses to plan for these intermittent cycles, coordinating staffing and stock far more effectively than the current unpredictable parade of rule changes.
The hope is that a vaccine will emerge soon, and we can return to the safety and freedom that we value more now than we ever realised. In the meantime, with proper compliance and coordination, we can at least reduce the amount of sacrifice required. That will be a win for everyone.