Categories
Science

To truly understand the pandemic, you need to think about this number (and it’s not R0)

https://www.zmescience.com/other/pieces/dispersion-factor-k-pandemic-20112020/

Since the pandemic first started (and even before it was officially a pandemic), there was a lot of talk about the basic reproductive number (R0) — and rightfully so. This number, R0, basically tells you the average number of people infected by one infectious individual; it’s a sort of measure of average contagiousness.

The first estimates published for SARS-CoV-2 were around 2.5, which would indicate that every infected person will pass the virus to 2.5 people on average. Since then, there have been various estimates for this R0, some higher or lower, generally within the 1.4-4 range.

But something didn’t fit.

For instance, how is it possible that people in the same house have only about a 20% chance of passing the virus to each other? According to the Korean study that analyzed this, even people sleeping in the same bed sometimes don’t pass the virus to each other. So how can this be reconciled with, for instance, one single singer that infected 87% of the choir in just one session?

No matter where you look, discrepancies like this pop up and there’s no one explanation to settle things. Presumably, it’s many different things add up. Some people might have high viral loads or might simply cough more. Others might not show any symptoms and have an active social life, passing the virus on to many others without even knowing it. Culture, testing, transportation habits, weather — these and many more parameters can complicate the issue.

The 80-20 rule has been used as a ballpark approximation in the pandemic: 20% of individuals may be responsible for 80% of the disease spread. These individuals are called super-spreaders and similarly, events that play an oversized role in spreading the pandemic are called super-spreading events.

So the average transmission, R0, just won’t cut it. Instead, there’s a different number we should be looking at.

Bill Gates walks into a bar

They say when Bill Gates walks into a bar, everyone’s suddenly a millionaire. That is, his net worth is so high that if you average it to everyone in the bar, it’s enough to make everyone a millionaire.

A similar thing happens with viral transmission. If you average it out, it seems like we’re all millionaires. But in reality, a few are billionaires and many are struggling to make a living.

The number that sums up this distribution is called the “dispersion parameter” — or K for short.

It works like this: a low K value means that a small number of infected people are responsible for large amounts of transmission. The lower K is, the more superspreaders are super spreaders. For influenza, K is approximately 1, as is also thought to be the case for the 1918 Spanish Flu. Meanwhile, some research has shown that the previous outbreaks of SARS and MERS had lower K values: 0.16 and 0.25 respectively.

This translates into something like this: for the flu, 40% of infected people might not pass on the virus to anybody else. For SARS or MERS, that figure is closer to 70%.

In the case of SARS-CoV-2, the K factor seems to be even lower, closer to 0.1. This has been suggested by multiple studies from the beginning of the pandemic, which suggests that most people don’t transmit it at all, while 10-20% may be responsible for 80-90% of transmission.

What this all means

The UK claims it was the best-prepared country for the pandemic — xcept the early stages of the pandemic were a disaster in the UK and nothing really recommends its approach over those of its western European peers. The UK did indeed have a good plan, but it was for the wrong pandemic. The UK had a flu-based pandemic plan, and it acted on it. But COVID-19 isn’t the flu. The flu has a K of 1, while COVID-19 has a value closer to 0.1. So instead of blanket measures that apply to as many people as possible, you need to identify and target superspreader events with accuracy.

Several Asian countries identified this early on. Just look at Japan, a country with an aging population and major crowded urban clusters. Japan was widely thought to be extremely vulnerable to the coronavirus, and when the cases started to rise, it seemed Japan was headed for disaster. But Japan managed to flatten the curve through a clever approach that focused on targeting these superspreaders and super-spreading events.

If you look at R0 and neglect K, it’s like not seeing the forest for the trees, says Oshitani Hitoshi, Professor of Virology at the Tohoku University Graduate School of Medicine.

“Data clearly indicates that the measures taken by Japan have been more effective than those taken in Western countries. The difference was the strategy to suppress the transmission. In short, Japan’s strategy was “to see the forest to understand the whole”. Western countries, including New York City, focused more on seeing the trees instead.”

The core of Japan’s strategy (and that of other countries like South Korea) was to trace large sources of transmission and take measures to tackle these clusters. They eased restrictions in some areas, tolerating small degrees of transmission, but nipped the bud of large transmission. For a disease like the flu, it’s not the best approach. For this particular coronavirus, however, it seems like the way to go.

To detect these clusters, having a robust contact tracing system is absolutely mandatory — but we’ve seen many countries fumble, either due to inability or lack of political will. The way contact tracing is carried out in some places could also use a revamp.

Right now, many areas are doing forward contact tracing: you see who contracted the disease and you try to see who that person has been in contact with. But with such an overdispersed virus, that may not be the most productive way to do it, as most people don’t transmit the virus at all — so you’re spending a lot of time and effort tracing the contacts of people who will not infect others. Instead, perhaps we should be doing backward contact tracing. That is, you try to see where people contracted the virus and work on the assumption that it may have been an important infection cluster.

This type of backward tracing is more challenging and also controversial. It also only works if the number of infections is relatively low, otherwise the whole country becomes one big cluster.

But there’s also good news. Now that we know how skewed transmission is and how important cluster detection is, we know that we don’t need to close everything down. If we’re precise, we can preserve some parts of normal life that have a low risk of spreading the virus while keeping transmission in check.

There’s no easy way out. But once we start seeing the forest from the trees, we can make our lives a bit easier. It’s a war of attrition and we need all the help we can get.

Leave a Reply

Your email address will not be published. Required fields are marked *