"Drinking at least 4 cups a day could slash your risk of rosacea by 20%," reports the Mail Online.
Rosacea is a common and poorly understood inflammatory skin condition that causes flushing and irritation of the skin, usually on the face.
The condition is marked by flare-ups, where symptoms can be troublesome for a few weeks before fading.
The precise causes are unknown, although various things have been suggested as triggers for the condition, including sunlight, heat, spicy foods, alcohol, and hot and caffeinated drinks.
In this study, researchers used data from the ongoing US Nurses' Health Study.
They found women who reported drinking coffee regularly were less likely to be diagnosed with rosacea, compared with those who rarely, if ever, drank coffee.
The biggest decrease in risk was seen among women who drank 4 or more cups a day, who had a 23% lower risk of rosacea.
Other sources of caffeine, such as tea, soda and chocolate, weren't linked to a risk of rosacea. This could be because they contain only a small amount of caffeine compared with coffee.
These results are surprising, as many health organisations and websites, including this one, recommend avoiding coffee as a potential trigger for the condition.
If these results are replicated in further studies, advice to avoid coffee to lower the risk of rosacea may need to be reconsidered.
The researchers who carried out the study were from Qingdao University in China, and Harvard University and Brown University in the US.
The study was funded by the Dermatology Foundation and Brown University.
The study was reported with reasonable accuracy in the Mail Online, although the headline stating that "coffee is good for your skin" is speculation.
The Sun's reporting was of poorer quality. The newspaper's claim that "Drinking coffee could be key to 'curing rosacea'" is an incorrect interpretation of the study.
The study found people were less likely to get rosacea if they drank coffee, not that people with rosacea could be cured by drinking coffee.
Researchers carried out a cohort study using data from a big long-running survey of female nurses in the US, the Nurses' Health Study II.
They wanted to investigate links between caffeine consumption from different sources and rosacea.
Observational studies like this are good for spotting links between factors, but can't prove that one directly causes another. Other factors may be involved.
The study began in 1989, and women were asked about their diet every 4 years, including how often they ate or drank certain foods and drinks containing caffeine.
In 2005, they were asked whether they'd ever been diagnosed with rosacea.
Researchers looked to see how many women had rosacea, and how this compared with their average consumption of:
They took into account these potential confounding factors:
They calculated the chances of women having rosacea at different levels of caffeine and coffee consumption.
Of the 82,737 women who took part, 4,945 were diagnosed with rosacea (59 for every 1,000 women) during the 16-year study period.
Among women who drank 4 or more cups of caffeinated coffee daily, there were roughly 4 diagnoses of rosacea for every 1,000 women each year.
Among women who drank 1 or less cup a month, there were roughly 5 diagnoses of rosacea for every 1,000 women each year.
Researchers saw a similar reduction when comparing women who consumed the most and least amounts of caffeine (HR 0.76, 95% CI 0.69 to 0.84).
But when they looked at caffeine from foods and drinks other than coffee, no link was found.
They also checked whether decaffeinated coffee consumption was linked to a reduced risk of rosacea, and found it wasn't.
The researchers said: "We provide evidence that caffeine intake and caffeinated coffee consumption are associated with a decreased risk of incident rosacea."
They say this "may have implications for the causes and clinical approach to rosacea" and that their findings "do not support limiting caffeine intake as a preventive strategy for rosacea".
The study adds to the evidence looking at how coffee or caffeine consumption may be linked to rosacea.
Contrary to other studies suggesting that it may be a trigger for rosacea, this study finds the reverse.
It doesn't prove that coffee is protective against rosacea, but the results are interesting. The study shows a clear dose-response – that is, the more coffee people drank, the lower the risk.
Although the suggestion coffee might lower the risk of rosacea sounds odd, there are possible reasons why it might.
Caffeine affects the blood vessels, which might prevent them widening, as they do during a rosacea flush.
Caffeine and other compounds in coffee are also antioxidants, which might suppress inflammation.
It may also affect hormone levels. Hormones are a possible trigger for rosacea, which is more common among women after the menopause.
But the study does have limitations to be aware of:
If you have rosacea, you may have been advised to avoid coffee. Doctors will need to consider this evidence alongside other research in the field to see whether this study is enough to change advice.
It's not possible to say at this stage whether this will change understanding around the causes of rosacea.
For now, if you find hot drinks including coffee trigger flares of rosacea, they're clearly best avoided.
But if you haven't had any problems after drinking coffee, this study suggests there appears to be no reason to stop.
One option if you're affected by rosacea is to keep a trigger diary, in which you record your exposure to potential triggers and then see how it impacts your symptoms.
It could be the case that certain triggers differ from person to person.
Read more self-help advice about rosacea.