If you want to reduce your cancer risk, completely eliminating alcohol consumption seems like the way to go.

Alcohol still plays a strange role in our society. We’ve known for a while that it’s bad for our health, we tax it quite heavily, and yet it remains incredibly popular. In most nations, it plays an important cultural role, it’s present everywhere and shows no real sign of declining. Even more, every once in a while, there’s the odd study saying that alcohol might good for you. The health impact of alcohol on our body is still controversial and probably won’t be settled by any single study, but there is growing evidence that alcohol is bad for you, even in low quantities.

A new study carried out in Japan, one of the countries with the best overall health, reports that every bit of alcohol increases your cancer risk.

Researchers looked at data gathered between 2005 and 2016 in 33 general hospitals in Japan. They examined 63,232 patients with cancer and took an equal number of participants in the control group. The participants reported their average daily intake of alcohol.

As soon as the alcohol consumption became larger than zero, the risk of cancer also increased. There was an almost linear association between cancer risk and alcohol — the more participants drank, the more their risk increased. For instance, a light level of drinking (one drink per day for 10 years) would increase overall cancer risk by 5%. Those who had an average of 2 drinks per day had almost double the risk of cancer.

Researchers say that this is particularly significant since heart diseases are relatively rare in Japan, and cancer is one of the biggest health problems in the country.

“In Japan, the primary cause of death is cancer,” said Masayoshi Zaitsu, one of the study’s authors. “Given the current burden of overall cancer incidence, we should further encourage promoting public education about alcohol-related cancer risk.”

Of course, this opens up the perennial discussion about association and correlation-not-causation. This study did not analyze causation, it only looked at a statistical relationship. Also, it is quite possible that Japanese people are not representative of the entire planet, so the findings might not carry over in other cultures. But it’s another piece of evidence suggesting that no level of alcohol consumption is without risk.

So if you’re thinking about a cup of wine… you may as well make an informed decision.

A previous study found that 50% of the cancer cases in Japan are preventable, and alcohol seems to fit in quite well in this pattern. Smoking and consumption of fried meat were also found to lead to an increase in overall cancer risk.

Journal Reference: “Light to moderate amount of lifetime alcohol consumption and risk of cancer in Japan.” Masayoshi Zaitsu, Takumi Takeuchi, Yasuki Kobayashi, and Ichiro Kawachi. CANCER. (DOI: 10.1002/cncr.32590).