Can humid weather help keep coronavirus under control?

Plenty of people have been speculating about how COVID-19 might be affected by the weather. Finally, we have some peer-reviewed science to sink our teeth into.

ABC

Aug 25, 2020, updated Aug 25, 2020

Everyone from the President to the postie seems to have been speculating about how COVID-19 might be affected by the weather, but finally we have some peer-reviewed science to sink our teeth into.

A new study has found an inverse relationship between humidity and reported cases of COVID-19 in Sydney, both as the number of cases grew and declined, suggesting dry air could aid the spread of the virus.

What did the study do?

The study, conducted by Dr Michael Ward, professor of veterinary and public health at the University of Sydney, and international colleagues, took locally acquired COVID-19 cases in Sydney between February and May and used postcode data to link cases to the closest weather station.

They then compared the number of reported cases with a range of weather conditions as recorded by the Bureau of Meteorology including temperature, relative humidity and wind speed recorded at 9:00am and 3:00pm, as well as rainfall, in the 14 days before the reported case.

Of all the possible weather variables, they found relative humidity at 9:00am was the best predictor.

For every 1 per cent decrease in relative humidity, they found a 7 to 8 per cent increase in COVID-19 cases.

Dr Tim Inglis, medical microbiologist and head of pathology and laboratory medicine at the University of WA, and was not connected to the study, said it was a “very plausible” and “interesting” series of observations.

“Which, for those in the know who understand the behaviour of respiratory viruses and the environment, is maybe not too surprising,” Inglis said.

Those in the know will be aware of other studies looking into similar respiratory viruses such as SARS and MERS, which also found links to various weather indicators including humidity and sometimes temperature.

Another recently published study from Ward’s team linked COVID-19 to temperature and humidity as the disease took hold in mainland China, and another look into Sydney published in May, which also found only relative humidity to be a factor.

“The temperature thing seems to be a little bit complicated and depending on where you are,” Ward said.

But humidity seems to be a consistent thread.

“At least within the context of New South Wales, it really looks as if when the air is dryer there is a greater risk of COVID occurring,” Dr Inglis said.

Why would dry air help spread COVID-19?

The theory put forward by Ward and his colleagues is that when the air is more humid, aerosol particles of COVID-19 are large and act like droplets, falling out of the air.

When the air is drier, the aerosol particles shrink and can therefore stay suspended in the atmosphere for longer.

For this theory to hold, we need to accept that COVID-19 is spreading through aerosols via airborne transmission.

It is something the World Health Organisation (WHO) initially downplayed but now acknowledges there are cases in closed settings where aerosol transmission cannot be ruled out.

One study has showed COVID-19 can remain suspended as an aerosol for at least three hours and another for up to 16 hours under laboratory conditions.

“They see the same relationship in the laboratory we’re seeing out in the real world. It gives us confidence that maybe that’s something that’s going on,” Ward said.

Are the tropics off the hook?

No.

Before Queenslanders and those from the Top End get too cocky, this does not make you immune to COVID-19.

Just look at Florida or Indonesia, which have recorded thousands of cases in recent months despite high humidity.

If we accept the link between low humidity and increased aerosol transmission as the mechanism behind this relationship, that does not stop any of the other forms of transmission.

All of those COVID-19 droplets and direct-contact opportunities are still out there, regardless of how like a sauna it is.

The results from this latest paper only looked at, and directly apply to, Sydney.

Ward speculates that even in humid conditions, changes in humidity over short timeframes could affect COVID-19 rates — but he stresses that analysis has not yet been done.

There could be a completely different relationship in the tropics. For example, flu is not considered seasonal in the tropics like it is in temperate latitudes.

Things to consider

This paper was published in a scientific peer-reviewed journal, meaning it has been given the tick of approval by others in the field.

But even with the best science, there are always things to consider.

Obviously, the weather is not the only thing that could be contributing to the spread of COVID-19.

Observational studies like these cannot rule out the possibility of behaviour changes or other factors influencing outcomes.

Ward said they attempted to “crudely” take behavioural change into account by dividing the study period into before and after control measures were in place. They found the inverse relationship between humidity and COVID-19 cases held in both stages.

“That indirectly gives us confidence that even though the cases were going down in the second phase, there was still that relationship persisting between humidity and cases,” Ward said.

The study is also limited by its access to only home postcode data, not necessarily the location where COVID-19 was contracted.

Likewise, it uses outdoor weather variables, when the majority of transmissions are thought to take place indoors.

Dr John Mathews, professorial fellow in the school of population and global health at the University of Melbourne, and was not connected to the study, also points out this data is not measuring the actual spread of COVID-19 but when cases are detected through testing, and testing is more likely if someone has symptoms.

“So an alternative explanation for what’s been found is that the changes in humidity change the proportion of people who’ve got symptoms, rather than actually having an effect on the virus spread,” Mathews said.

“I can’t prove that’s true, but it just illustrates how difficult it is to draw valid conclusions from complex data like this.”

Despite his scepticism, Mathews says the proposed link between low humidity and reduced transmission is plausible.

“It’s an interesting paper and I guess one of the responsibilities of old people like me is just to remind people that you can get interesting findings which agree with your hunch but it doesn’t necessarily mean your hunch is right.

“There may be alternative explanations,” he said.

Should I buy a humidifier?

Not just yet.

The data has so far only shown a link between outdoor humidity and the number of COVID-19 cases; most transmissions are thought to be happening indoors.

Ward said it would be interesting now to look at indoor outbreak locations and see if there was a relationship with indoor humidity.

Especially for places like meatworks environments, which he said could be conductive to aerosol production, or in aged care homes.

“If there is, I think then you might have enough evidence to say, ‘well, maybe we should consider modifying humidity as one of our responses’,” Ward said.

But Inglis called for caution. He said a humidifier could well increase the water content, but it might also affect air currents indoors.

“It may be those air currents that are indoors that are more important to the transmission than just the humidity itself.”

“So these are the sorts of things that need to be unwrapped by people who’ve got the ability to do the detailed studies before we can provide specific guidance on whether people should go out and purchase humidifiers.”

COVID-19 warning days

Inglis said a warning system to alert the public on days with low humidity would be a good idea, “certainly on a trial basis”, like we have for fire danger or thunderstorm asthma.

“Those are the sorts of things that you’ll see the public health physicians and the people who do the detailed modelling looking at on a wider basis than just New South Wales or urban Sydney,” he said.

But he reiterated there was much work to do. “The sensible approach is to look to this group that have provided us with some useful information, for some follow-on studies as a matter of priority.”