There was roughly a 15 per cent drop in mortality for every one degree Celsius rise in temperature, the researchers said.
In contrast, the severity of symptoms and mortality rate remained constant in China during the first wave of the pandemic, which occurred solely throughout the winter.
The study also looked at data from more than 37,000 UK users of the Covid Symptom Study app reporting symptoms consistent with Covid-19.
It showed a similar decrease in the severity of reported symptoms from March through May as UK temperatures rose.
According to the researchers, the changes were too large to be explained by improvements in treatment of the disease, patient age or hospitals becoming overwhelmed during this time, suggesting there is a seasonal influence on the virus.
Scientists have called for caution over a new study that suggests the severity of Covid-19 may be reduced during the warmer months of the year, and that dry indoor air may encourage its spread.
Researchers from King’s College London analysed data from 6,914 patients admitted to hospital with Covid-19 in Croatia, Spain, Italy, Finland, Poland, Germany, the UK and China.
They mapped this against local temperature and estimated indoor humidity and found that severe outcomes – being taken to hospital, admittance to ICU or the need for ventilation – dropped in most European countries over the course of the pandemic, covering the transition from winter to early summer.
However, experts have challenged the findings, pointing to a number of methodological caveats within the study – which has not yet been peer-reviewed.
Sir Timothy O’Shea, a professor of veterinary epidemiology and data science at the University of Edinburgh, stressed that the paper only focuses on hospitalisations and failed to fully take into account additional factors which could explain the falling number of severe outcomes.
“It does not address the question of whether or not Covid-19 would transmit more readily in those conditions [outside of hospitals], and therefore it should not be inferred that the result would be a larger epidemic.”
He added: “A further caveat, is that it is difficult to assess whether any decline in deaths or severity over time is associated with improved methods for in-hospital treatment over time, and as staff become more experienced in working with Covid-19 cases, and/or as numbers of infected fall off over time (reducing strain on hospital staff).
“Also not explored is the potential role for co-morbidities, if for example the attack rate of Covid-19 is higher in individuals whose health is already compromised as, if true, this would mean that many of those more likely to be severely affected would also be more likely to be infected earlier.”
Dr François Cohen, an environmental economist at the University of Oxford, with expertise in the impact of the weather on socioeconomic factors, echoed this sentiment, saying: “The decrease in hospitalisation times found in this paper are very interesting. However, the conclusion that seasonality would be the main driver of this change is interpretative.
“I believe that many confounding factors could explain the decrease in hospitalisation times, beyond the weather, and would therefore recommend being very cautious when mentioning the results of this study.”
Dr Simon Clarke, an associate professor of cellular microbiology at the University of Reading, highlighted the variation in outbreaks that had been reported across the world – irrespective of regional temperatures.
“On the one hand we have a current, summer outbreak in the United States but we’ve also seen an unusual number of infections occur in workplaces where cold conditions occur, which is well established as being association with colds and flu.
“This study shows an association between temperature and severity of symptoms, but it does not demonstrate why that happens. It may be because low temperatures slow down the normal mucus clearing of infecting viruses in our nasal passages.”
The study suggests indoor heating during the winter months may also contribute to the spread of the disease by drying out the protective mucus barrier within the nose and airways, making viral infection easier.
Should the results of the paper hold, said Sir Timothy, then “this represents an important outcome – namely, showing that in a variety of settings, there is strong evidence for an increase in severity of Covid-19 with lowered temperature.”