Explained: Seasonal UV radiation and global Covid-19 trends

The study compared various populations to themselves over time, and tested whether daily variations in UV radiation lead to changes in Covid-19 cases up to three weeks in the future.

The novel coronavirus has now seen all the seasons in both hemispheres, but the effect of seasonal factors such as temperature and humidity remains to be fully understood. New research has now looked at the influence of another seasonal variable — ultraviolet radiation from the Sun — and found evidence is that higher natural UV radiation reduces Covid-19 transmission. The study is published in the journal PNAS.

The caveat

The researchers stress that seasonal changes in UV are just one of many causes of change in Covid-19 case rates. And the influence of UV is modest compared to policy measures such as travel restrictions, school closures, or event cancellations. “On average across the northern hemisphere, we find that prior estimates of the effect of these social distancing policies are 3-6 times larger than our estimated effects of seasonal changes in UV,” economist Tamma Carleton of the University of California–Santa Barbara said, by email.

Because other research has looked at how artificial UV light can inactivate the coronavirus, it is also important to note that this is not the same as the effect of the sun’s UV radiation. UV disinfectants currently in use today rely on wavelengths in a range called UV-C. Such wavelengths do not reach us naturally from the Sun, as these are absorbed by the ozone layer. In sunlight that reaches the Earth, the wavelengths are in the range known as UV-A (and to an extent in the range UV-B).

📣 JOIN NOW 📣: The Express Explained Telegram Channel

The analysis model

The study avoided comparing Covid-19 trends in different locations because this approach would introduce confounding factors. “For example, comparing Covid-19 cases in India to cases in Norway and attributing the difference to large differences in average UV exposure would be erroneous — there are many differences between India and Norway, including population density, health services, testing rates, etc, and UV is just one of many factors that differs across these countries,” Carleton said.

Instead, the study compared various populations to themselves over time, and tested whether daily variations in UV radiation lead to changes in Covid-19 cases up to three weeks in the future. They gathered disparate datasets from different countries’ agencies, and harmonised them to create a global dataset.

The extent of variations

A change in UV exposure by 1 standard deviation, the analysis found, reduced the growth rate of new cases by around 1 percentage point over the following two weeks. A change of 1 percentage point, for context, is approximately equivalent to the change in UV between April and May in Delhi.

Based on changes in UV, the model predicted Covid-19 growth rates for the temperate zones north and south of the tropics. Between January and June, Covid-19 growth rates would increase by 7.3 percentage points in southern temperate locations and decline by 7.4 percentage points in the northern temperate ones. And when the seasons flip, growth rates in December compared with July were predicted to decrease by 7.7 percentage points in southern temperate regions, and jump by 7.8 percentage points in northern areas.

Carleton noted, however, that UV seasonality looks different in India than in other parts of the northern hemisphere due to the South Asian monsoon. “The onset of the monsoon lowers summer UV, which can raise the Covid-19 transmission risk relative to other times of the year and relative to other regions in the northern hemisphere, where summer tends to be the highest UV season,” she said.

Causal or correlational?

Indeed, infection rates did appear to have decreased in much of the northern hemisphere during the summer, the researchers noted, but observed that many parts of the northern hemisphere also relaxed their pandemic restrictions during the same period. Asked whether the correlation with UV can indeed be called causal, Carleton said: “In the paper, we use a statistical model that is designed precisely to isolate natural variation in UV exposure that is ‘quasi-random’, so that we can feel confident causally interpreting the effect of UV on COVID-19 growth rates.”

Source: Read Full Article