COVID-19 : tracking clusters to keep ahead of fresh surges

With many countries around the world beginning to see or anticipating a resurgence of COVID-19, attention has been turning to how to avoid an economically punishing second round of lockdowns while keeping transmission down. One method is by examining the types of hotspots where clusters of cases emerge.

During the virus, clusters have already been reported in widely diverging locations ranging from a choir practice in a church to meat-packing plants, ski resorts, restaurants and migrant worker dormitories.

The thinking behind investigating these further is that just 10 to 20 per cent of COVID-19 cases have been estimated to result in 80 % of transmission – so-called “superspreading”.

If the identities and characteristics of places where this commonly happens can therefore be pinpointed and the right actions then taken in those locations to prevent spreading, it might help contain the virus while numbers of cases are low and stave off the reinstatement of full-blown lockdowns.

In the first cycle of COVID outbreaks early this year, some researchers and countries started tracking this – with Japan commonly cited as adopting a cluster-based approach. Now, as cases begin to rise again in some countries, we will begin to learn more about the promise of such approaches and the potential they have to pay off.

Database initiative

Among researchers tracking clusters is a group based at the London School of Hygiene & Tropical Medicine (LSHTM), which began developing an open-access online database in the first wave through a systematic search for cluster events in scientific literature and media reports.

When the researchers published their initial findings in May and in an updated version in June, they detailed how they had found reports on clusters in more than 20 mainly indoor settings. A limited number were from schools, while many were from households, and a rising number from hospitals and elderly care settings. At present, the database goes up to the start of July, containing 265 recorded events.

The LSHTM team is now seeking to combine the database with two others to create one containing more than 1,500 cluster events.

The researchers pointed out that there are challenges to collecting the data, such as that it turned out there was a very limited amount of open-access systematic contact-tracing data available. There can also be a range of biases in the data, such as recall bias towards big events and media bias.

But despite the challenges, this is all helping to build up the evidence base – and the hope is that some countries also start opening up their data more.

Dr Gwenan Knight, assistant professor at the LSHTM: “If we think back to where we were at the start of this, we really had no idea where this was transmitting. For example, the lack of symptomatic clusters linked to schools was very different to what we would have expected. This database is another piece of evidence on understanding this.” – Read Gwenan Knight’s full interview

As things progress, one of the key things she is interested in analysing is how the settings for clusters change over time as the pandemic enters different phases. If we can see how these change pre and post lockdown or in light of different interventions, it could then shed more light on the effectiveness of these strategies.

Dr Knight said there is some evidence that more clusters have been linked to schools as they opened before summer breaks, but the data since early July needs further analysis to assess trends more closely.

Lessons from Japan?

One country widely cited as using a cluster-based approach is Japan. There, health experts sought to minimise the spread of disease by keeping a close eye on these potential hotspots and carrying out retrospective tracing to try and track patients’ movements from before they became infected.

In a recent study, researchers identified the country’s cluster approach as a key element in mitigating the initial impact of COVID-19 – while noting that it failed to stop the disease spreading in the slightly longer term.

Prof. Rajib Shaw, Graduate School of Media and Governance at Keio University, Japan: “These clusters were highly publicised in all different types of media, which also helps bring awareness among people so that they avoid these types of cluster areas.”– Read Rajib Shaw’s full interview

In the first wave, the curve flattened with deaths well below 1,000 – much less than many other countries – while the amount of testing was also kept lower than in many other locations. Prof. Shaw thinks this type of approach offers a pathway to helping keep the economy running while reducing transmission, and helps give other countries ideas on how such strategies can be successful.

And although some of this may have been helped by Japan having more of a mask culture than some countries such as those in Europe and the US, he thinks that improving this is possible with the right risk communication. “I think every country has its own responsible citizens, so it’s possibly that the way of risk communication is different from country to country,” he says. “[It matters] how we communicate the same message with a different group, ethnicity, background or age group.”

On top of that, he points out that Japan carried out all its contact tracing manually at the peak time rather than using apps, but an app introduced by Japan called COCOA in June could make this even more effective.

And Prof. Shaw still maintains that the initial strategy has paid off despite alarm bells being rung by some in the past few weeks due to reports of a significant resurgence of COVID-19 in Japan, with daily cases topping 1,000 multiple times and suggestions that the country may have moved too fast to normalise. Globally, contact tracing apps have also struggled to gain real traction.

“I do not think the surge nullifies the initial success,” says Prof. Shaw, pointing out that some resurgence was expected and that it’s important to look beyond the raw numbers of cases. Although the number of reported cases has risen recently, he says, those requiring intensive care are still in the low double digits and deaths are numbering only around three to five per day as of early August.

Prof. Rajib Shaw: “The key point is that detailed contact tracing and cluster identification at the earlier stage helped Japan to flatten its initial curve, and gave enough time to public health and medical facilities to prepare themselves. So there is no collapse in the healthcare system, unlike in several countries.”

Meanwhile, even if countries that use a cluster-based strategy do see setbacks, it should still provide key information for others on the paths they might follow. And if a larger amount of systematic, in-depth data on susceptible settings can be made available, it could be highly valuable in the battle against the virus.