A public health expert is calling on the National Institutes of Health to change information on one of its websites to more accurately reflect scientific findings about the risk of alcohol. In a letter sent to NIH Director Dr. Francis Collins on Thursday, alcohol researcher Dr. Michael Siegel of Boston University School of Public Health called on NIH to retract and apologize for a statement on the website of NIH’s National Institute on Alcohol Abuse and Alcoholism that says, “Drinking too much alcohol can increase your risk of developing certain cancers,” especially as it regards breast cancer.

“This is in direct contradiction with the scientific evidence which shows that even light to moderate drinking increases women’s risk of breast cancer,” Siegel told STAT. By implying that only “too much” drinking is a cancer risk, “the NIAAA is deceiving and potentially harming women while furthering the agenda of the alcohol industry.”

The NIAAA has been repeatedly taken to task for its ties to the alcohol industry. In June, the NIH terminated an NIAAA-sponsored study on the cardiovascular benefits of “moderate drinking” after an investigation found that NIAAA officials had “early and frequent engagement” with the industry, including soliciting funding for the $100 million project from liquor companies that were told the study “represents a unique opportunity to show that moderate alcohol consumption is safe and lowers risk of common diseases.”


NIAAA’s warning about drinking “too much” alcohol is at odds with how its sister institute, the National Cancer Institute, describes the risk of cancer from alcohol consumption. The NCI notes that alcohol is “a known human carcinogen,” and emphasizes that “the more alcohol a person drinks,” particularly over time, “the higher his or her risk of developing an alcohol-associated cancer.” (Those include head and neck, esophageal, liver, breast, and colorectal.)

The cancer institute also explains, however, that more than 100 epidemiologic studies of alcohol consumption and breast cancer in women show that even low consumption raises the risk. Women who drank more than 45 grams of alcohol per day (about three drinks) had a 50 percent higher risk of developing breast cancer than nondrinkers.

But for every 10 grams of alcohol consumed per day there was a 12 percent increase in the risk of breast cancer, according to a study of 1.2 million women in the United Kingdom. That agrees with a more recent U.S. study from researchers at the Harvard T.H. Chan School of Public Health, which found that “for women who have never smoked, risk of alcohol-related cancers (mainly breast cancer) increases even within the range of up to one alcoholic drink a day.”

Most people reading the NIAAA website, Siegel argued, would not interpret “drinking too much” as less than one drink a day.

“Every person I’ve spoken with has told me they interpret ‘too much’ as meaning enough to impair them,” Siegel told STAT. “But there is a second problem: There is no known threshold for the risk of breast cancer introduced by alcohol consumption. So strictly speaking, any alcohol consumption is ‘too much’ in terms of raising one’s risk of breast cancer.”

In his letter to Collins, Siegel asked, “Why would an agency whose mission is to reduce the harms of alcohol put millions of women at risk by lying to them and telling them that there is no increased breast cancer risk unless they drink excessively? Could it be that the NIAAA is in bed with the alcohol industry?”

NIAAA did not immediately respond to a request for comment.

Leave a Comment

Please enter your name.
Please enter a comment.

  • It really is the end of science. The phrase “Too much drinking” came straight from the alcohol lobby. The NIH is in the pocket of the profiteers. The NIH is also promoting alternative medicine on their site. A few billionaires decided there was money to be made misleading the ignorant. The studies they posted to promote the alternative medicine industry are unscientific and misleading, yet they feature prominently and are often reprinted to sell quack cures, and promote dangerous things like Battlefield Acupuncture.

    • Mavis, there are not any definitive studies one way or the other. Epidemilogic studies and anecdotal evidence suggest that alcohol causes cancer. That is the best evidence we have so far, but its weak evidence.

      If you have any credible evidence, then please post a link.

  • One of this country’s biggest issues nobody wants to confront. The alcohol industry has relinquished responsibility from the danger of its product by focusing on impaired driving. As if that’s the only danger to be concerned with, while placing responsibility of their product on the consumer. Big alcohol is as dangerous as big Pharma and big tobacco.

  • “The cancer institute also explains, however, that more than 100 epidemiologic studies of alcohol consumption and breast cancer in women show that even low consumption raises the risk. Women who drank more than 45 grams of alcohol per day (about three drinks) had a 50 percent higher risk of developing breast cancer than nondrinkers.”

    I don’t think anyone would define 3 drinks a day (21 per week) as “low consumption.”

  • Re: Alcohol/Breast Cancer:
    There are numerous risk factors associated with breast cancer, and other cancers. However, risk does NOT mean causative. The methods used to attribute risk are statistical at best and we should consider them.
    It seems wasteful to me to spend thousands if not millions to make risk associations, when that money could be used to find the VIRUS causative for this disease. Like many others, viruses that are causative are sexually transmitted. The fewer sex partners one has in a lifetime the better. Always use protection.
    Today’s hook-up culture is a grave mistake.
    To the naysayers I say consider the following: HPV -causative for numerous genital & reproductive cancers, Ebstein-Barr appropriately named the “cancer virus” for Hodgkins and Non-Hodgkins lymphomas, Nasal cancer, certain stomach cancer, Herpes virus 8 for Kaposi’s sarcoma, Polyoma virus for Merkel cell cancers..I could continue. My point is, if all these cancers have viral cases, what makes anyone think that breast cancer is different??? Someone out there please find that virus, and please make a vaccine…and while you are at it, please make one for the entire Herpes Group. It would be a good start in stopping most if not all cancers.

  • Epidemiologic data do suggest that alcohol is a cause of cancer, and epidemiologic data also suggest that alcohol lowers the risk of heart disease. However correlation does not necessarily imply cause and effect. The epidemiologic data are the best data that we have, but these results are weak because they are not backed up by randomized controlled studies.

    People are looking for conclusive or near-conclusive results, but they do not yet exist.

Sign up for our Daily Recap newsletter

A roundup of STAT’s top stories of the day in science and medicine

Privacy Policy