Enlarge /. Testing and contact tracking can be essential for ending pandemic closures.
Sean Gallup / Getty Images
View more stories
When the scale and threat of the COVID-19 pandemic became clear, researchers who track the spread of disease were pretty much in agreement: to give us time to develop a therapy or vaccine, countries had to put in place strict restrictions to limit the opportunities for the pandemic virus to spread. Experts painted terrifying pictures of huge spikes of infection that would overwhelm local hospital systems if locks were not set, resulting in many unnecessary deaths. For countries like Italy and Spain, which were already in an uncontrolled spread, reality confirmed these predictions. The peaks rose sharply before the restrictions, but fell almost as much after their introduction.
However, the same models also predicted that ending the restrictions a few months later would put countries at risk of the virus returning, which would force governments to re-struggle between strict restrictions or an out of control in the next step of a cycle Choosing a pandemic would repeat until a vaccine or therapy became available. These countries now have a slightly different question: Are there ways to control the virus without resorting to a cycle of locks on and off? For countries like the United States, which have introduced restrictions that are short, unpredictable, and half-hearted so that the peaks are not separated by a large low, the same question becomes relevant if we ever get the virus under control.
A new study by a large international team is using epidemiological models to look at how things can be kept in check while the majority of the population can resume a semi-normal life. There are ways to deal with easing restrictions, but they require a combination of an effective contact tracking system, extensive testing, and the willingness of households to quarantine together.
From the real world to the model
The study was based on an epidemiological model – in this case, the basis was a fairly standard version that divides the model population into pools of people who are vulnerable, infected, or recovered. However, the model used here was much more complex because we now understand the behavior of SARS-CoV-2 much better.
As usual, people in the vulnerable pool are infected by virtual contacting someone in the infected pool. But instead of getting directly into the infected pool, they experience a latency period in which they are not yet infectious. Some people leave latency with an asymptomatic infection that continues to allow them to transmit the virus until they reach the restored pool. Others follow a different route through a pre-symptomatic but infectious phase, through a symptomatic-infectious phase, the severity of which is modeled and may include the time spent in hospital and / or in an intensive care unit. In this way, researchers can track the type of stress that a certain level of infection would mean for the health system.
To model the type of interactions that could allow the virus to spread, the researchers used an important source of real data: anonymized cell phone data. In this case, they had data on the location of mobile phones prior to the Boston region pandemic that they used in conjunction with census data for six months. In this way, they were able to create a three-tier interaction model and track infection opportunities at the level of people who share an apartment, who interact at work or in society, and who go to school.
To provide a basis for their model, the researchers ran it without restrictions. As with other, less sophisticated models, this resulted in an enormous number of infections that far exceeded the capacity of hospitals to treat patients. At its peak, the typical infected person passed the virus on to four people in the vulnerable pool. By the time things slow down, 75 percent of the population was infected.
Limitations and how to end them
The researchers then focused on what happens after the restrictions are relaxed. They ran the model in a scenario where, after an eight-week order at home, jobs were reopened for four weeks before most restrictions, including those for restaurants and theaters, were lifted. Schools and universities remain closed. The model here successfully matched what happened in New York City in terms of the number of contacts people normally have under the various restrictions each day.
But it also has bad news for New York City and elsewhere: a very large number of infections that occur as the restrictions are completely removed. This has not happened in most places, also because many jobs have not been reopened and many people have taken additional measures, such as using face masks that limit the spread of the pandemic. However, it points out the risks of full reopening, especially in places where the other measures cannot be guaranteed. While the increase in cases is only about half that of the model, we would see if we didn't do anything, it is still easy enough to overwhelm health systems.
However, the researchers' main focus was on a way to handle the reopening that they claim could significantly reduce the risk. It requires test capacity large enough that about half of COVID-19 symptomatic cases can be identified within two days of the onset of symptoms. Once identified, they are quarantined at home for two weeks. It is crucial that anyone who normally lives with this person must be quarantined at home for two weeks. Essentially, households are treated as single units for quarantine purposes. Another key to the researchers' plan is that contact tracking is efficient enough to track down a reasonable percentage of people who are close to those who get a positive test (they try values between 20 and 40 percent of the contacts tracked) . and that said people are quarantined.
Overall, this seems pretty reasonable. It does not require that we identify asymptomatic cases or people with symptoms with complete efficiency. A high level of successful contact tracking is also not expected, which would likely require the majority of the population to install contact tracking applications on their smartphones. However, it must also be recognized that the US testing capabilities are far from good enough to make this a reality today.
The results in this model are quite dramatic. Even with successful 20 percent contact tracking, the "second wave" peak of new infections drops to about 20 percent the size it would otherwise have, and remains within the capacity of hospital systems. When contact tracking reaches 40 percent, the second wave is essentially eliminated with the background increasing only slightly and gradually. And although the quarantine restrictions appear to be strict, the researchers note that at the height of the wave, only nine percent of the population has to be in quarantine.
This system also allows a degree of flexibility. If we can better identify infected people, we could get away with a fraction of the quarantined households. The test capacity for regularly checking the status of people in the same household as a positive case could enable more flexible quarantine. The research team also modeled the opening of schools and universities and found that we can take this step and keep the pandemic at bay by increasing contact tracking efficiency to 50 percent.
The authors also show that important additional safety measures such as the use of face masks improve the effectiveness of this exit strategy and allow us to get away with other steps with less efficiency.
It is also clear that the approach can actually be tried in the real world. There are a large number of countries where test capacity is high enough to ensure that cases are identified quickly, and few where contact tracking is likely to be effective. For countries like the United States where this is not the case, there are guidelines for how states and cities can set goals before restrictions can be relaxed and what to do.
In general, the focus on how blocking can be effectively completed marks a crucial transition in the way health professionals think about the pandemic. The first models were only carried out to understand the possible consequences of an uncontrolled pandemic, and they indicated that we had to face an increase in infections for a year, followed by bans – an apparently depressing future. However, over time, research has shifted to finding out how to end the locks in a way that reduces the risk of future surges. We are now at the point where models are used to identify the approaches that least constrain society while keeping the pandemic at bay.
The key question remains whether we can get society to trust public health experts by showing the safest ways out of the current chaos.
Nature Human Behaivor, 2020. DOI: 10.1038 / s41562-020-0931-9 (About DOIs).