It’s Not “No Safe Level” — But Not Moderation Either
And a blistering takedown of an alcohol report.
Today, the US government told Americans how to eat — and how to drink.
“If adults age 21 years and older choose to drink alcoholic beverages, drinking less is better for health than drinking more,” according to the new Dietary Guidelines for Americans (DGAs).
The House Oversight Committee also released a damning report on how one of the alcohol studies meant to inform the DGAs was assembled.
The secretive second study
The development of the alcohol guidance has been mired in controversy for almost two years. Although the DGAs are meant to rest on a single scientific review, the process began to fall apart when two competing studies were commissioned.
The first — authorised by Congress and funded with $1.3 million — was by the National Academies of Sciences, Engineering, and Medicine (NASEM) and published in December 2024. It concluded there was “moderate certainty” that people who drink alcohol in moderation have lower all-cause mortality than those who don’t.
The second was an Alcohol Intake and Health (AIH) study run by the Interagency Coordinating Committee on the Prevention of Underage Drinking (ICCPUD), whose mandate is underage drinking, not dietary guidance.
In stark contrast to NASEM’s report, an early AIH draft warned that “the risk of dying from alcohol use begins at low levels of average use” and that “higher levels of alcohol consumption are linked with progressively higher mortality risk.” The study was never published in final form.
The two reports couldn’t have been more different. One said moderate drinking was fine. The other said it was a potential death sentence.
The dispute quickly spilled into the press, with media running competing claims about which study was biased and who was really influencing the outcome.
The second study triggered a subpoena fight
In April 2024, the House Oversight Committee demanded to know why the ICCPUD study was running at all. After all, that’s what the NASEM report was there to do.
After months of requests and a subpoena, the Biden Administration coughed up just 31 pages of documents.
After the administration changed, Health Secretary Robert Kennedy handed over 82 pages of internal documents. The Oversight Committee read them carefully and then produced a report about the ICCPUD work. They called it A Study Fraught With Bias.
“The evidence points to the AIH study group having a pre-determined goal — to publish a biased study that parroted a ‘Canadian model’ conclusion that no amount of alcohol consumption is safe,” it says.
The Oversight Committee interprets this as evidence that the outcome had been decided in advance.
The Canadian model, released in January 2023, recommended that Canadians cut alcohol altogether, or at least limit themselves to just two drinks a week.
Not only that, but the Oversight report accuses the AIH group of being secretive and taking “active steps to conceal their study from Congress and the public”.
What Congress says really happened
The Oversight Committee report says all six members of the Scientific Review Panel “are affiliated with U.S. and international anti-alcohol advocacy groups”.
The scientists, it adds, were largely selected by Alicia Sparks, described as “a long-time anti-alcohol activist and board member of the anti-alcohol U.S. Alcohol Policy Alliance”.
That connection was not disclosed when the US Alcohol Policy Alliance publicly attacked the competing NASEM report.
In January, USAPA co-founder Diane Riibe told The New York Times that the NASEM review was “a thinly veiled effort to undo the growing evidence that alcohol causes cancer and is increasingly associated with serious health outcomes” — without revealing that a board member of her organisation had helped select the panel for the rival ICCPUD study.
The Committee further concluded:
ICCPUD “hid this information from Freedom of Information (FOIA) requestors and Congress,”
Its behaviour was “likely unethical” and “contrary to federal law,”
Those involved were “intentionally hiding information from FOIA requestors” while planning to
“push a political agenda based on the ‘Canadian model’ of a similar alcohol study concluding no amount of alcohol consumption was safe.”
And it says the ICCPUD panel tried to bypass scrutiny by taking its conclusions to an academic journal.
By submitting to BMJ Open while Congress was demanding documents, the panel could potentially convert an unauthorised interim draft into something that looked like independent, peer-reviewed science.
“The ICCPUD study group used an outside academic journal to prematurely release their predetermined ‘Canadian model’ findings the group planned to include in the original Biden Administration study,” the report states.
In short, there were lots of problems with the ICCPUD report. What’s even more problematic is that this might have become the basis for government policy. What stopped it wasn’t better science — it was a subpoena and a change of government.
So what’s the new drinking rule?
The new guidelines stop short of telling people not to drink at all.
“Alcohol is a social lubricant that brings people together,” said Dr Mehmet Oz, administrator of the Centers for Medicare & Medicaid Services, speaking at today’s White House press conference on the DGAs.
But while he noted that small amounts of alcohol play a role in conviviality, he advised people to drink less. As for the previous guidelines about two glasses a day for men and one for women, he said there was a move away from that because “there was never really good data. It was primarily confused with broader data about social connectiveness”.
So while the NASEM report has shaped the 2025-2030 guidelines, and the ICCPUD study has been left in limbo, the new guidance avoids a direct collision with the WHO’s “no safe level” line — recasting alcohol as something that may have social value, yet which should be approached with caution.
The alcohol industry should not mistake this for a reprieve. The direction of travel is clear: moderation is no longer being framed as protective, only as less harmful.
This post has been lightly edited since it was first published. In the rush to get it done, it was a bit rough.
