The death toll from the novel coronavirus outbreak has now topped 2,000 people. But while cases have spread to at least 28 countries beyond China, the vast majority of those deaths—all but six of them—have been in mainland China.
That’s not surprising, to some extent, given that the virus causing the respiratory illness COVID-19 originated in China, and given the nation’s high population density, among other factors. But there may be another underlying cause behind the severity of symptoms and number of coronavirus deaths in China: the country’s sky-high smoking rate among men.
More than 52% of Chinese males ages 15 and over are regular smokers compared with just 2.7% of females, according to the World Health Organization (WHO). There are more than 300 million smokers in the country, which accounts for 40% of total worldwide tobacco consumption.
Chinese men make up a disproportionate share of coronavirus infections and deaths, according to several recent studies. One report found that nearly 60% of examined cases occurred in men; others have pegged the number at anywhere between the mid- to high- 50s. The median age of patients ranges from about 47 years to 56 years, depending on the report.
A new review of infections published Monday also found a disparity in death rates. Among men, the coronavirus was found to have a 2.8% fatality rate. For women, the number was 1.7%.
While there are multiple risk factors for coronavirus infection, medical experts told Fortune that smoking may also be fueling the asymmetric number of COVID-19 deaths in men. Smoking could potentially make someone both more susceptible to infection and less able to fight it off given the medical conditions associated with tobacco use.
There are three possible ways that smoking may be driving these disparities, Dr. Peter Hotez, dean for the National School of Tropical Medicine at Houston’s Baylor College of Medicine, said: Patients with hypertension or heart disease (both associated with smoking) appear more susceptible to the virus; smokers are more prone to contracting the flu, and coronavirus infection can occur simultaneously with influenza; and there is growing evidence that a key biological marker that the current coronavirus strain binds to is far more prevalent in smokers than in nonsmokers.
“The question is whether smoking has a direct impact on the coronavirus,” said Hotez. “And preliminary evidence suggests that may be the case.”
A biological marker (or biomarker) is, simply put, an indicator of some kind that can be measured in biological samples like blood work that would show something abnormal like an infection. A study published last week by University of South Carolina researchers supports the thesis that smoking is linked with a higher amount of the biomarker associated with coronavirus infection. “We observed significantly higher ACE2 gene expression in smoker samples compared to nonsmoker samples,” wrote the authors. “This indicates the smokers may be more susceptible to 2019-nCov and thus smoking history should be considered in identifying susceptible population and standardizing treatment regimen.”
Viruses bind to specific receptors, and in the case of coronavirus, ACE2 is a key target. “ACE2 is more prevalent in the lower part of the respiratory tract, it’s more prevalent in men than in women, and it’s more prevalent in smokers,” Albert Rizzo, chief medical officer for the American Lung Association, said. “If you’re older, have bad lungs, smoke, you’re not going to fare as well with the coronavirus.”
There was a similar disparity between men’s and women’s death rates during the SARS outbreak of 2003, which was also caused by a strain of coronavirus.
The question is, How can such information be leveraged to tackle the outbreak or determine which populations are most in need of a vaccine—once one is actually developed?
Rizzo doubts that smoking status will be a direct factor in whether or not someone is vaccinated. Given the difficulty of developing a vaccine during an ongoing outbreak and inevitable supply constraints, high-risk populations will likely be prioritized during a vaccination drive.
“I don’t think smoking by itself will be a good figure to use,” said Rizzo. “It’s going to be whether these people are impaired on an immune system level.” That might mean having an underlying respiratory condition like chronic obstructive pulmonary disease (COPD) or a similar ailment—conditions that may well be caused by smoking.
More must-read stories from Fortune:
—How the coronavirus may spread around the world
—MWC trade show customers seek refunds after last-minute cancelation
—In China, oil won’t be the only energy sector battered by the coronavirus
—Why it’s so hard for China’s factories to get back to work
—Coronavirus may be the straw that breaks the back of oil fracking
Subscribe to Fortune’s Brainstorm Health for daily updates on biopharma and health care.