Surgeon General Vivek Murthy’s recent advisory that drinking alcohol raises the risk of cancer is something of a gamble. It’s a bet that telling people to do less of something they enjoy will be taken in good faith, and not as a politically motivated judgment of their lifestyle choices. It also exemplifies some perennial challenges in public health. Communicating about risk in an intelligible, actionable way to the public is hard to do without oversimplifying things. Helping policymakers and people decide what to do with the information is even harder.
The advisory relies on decades of epidemiologic studies and experiments testing what happens to mice who are given alcohol. (The poor, drunk mice develop tumors.) It’s a synthesis of information gathered across time, nations and many hundreds of thousands of people.
Even though the advisory relies on mostly observational studies and not randomized controlled trials, the relationship between alcohol and cancer has been demonstrated consistently enough that we can have confidence that it’s reliable; it’s a recurring signal heard through so much noise. The medical community also has theories about how alcohol causes cancer, through DNA damage and inflammation. The case is strong. But in terms of how usefully and precisely the advisory elucidates cancer risk for various levels of drinking, I’m less sure.
The report describes the relationship between alcohol and cancer in different ways: the number of new cases of cancer a year in the United States potentially related to alcohol consumption (roughly 100,000); the number of annual cancer deaths that might be attributed to alcohol (roughly 20,000, compared to nearly 200,000 cancer deaths attributable to smoking); the increase in absolute risk for developing alcohol-related cancers (a 2.5-percentage-point increase for women and a 1.5-percentage-point increase for men); and the relative risk for specific cancers, such as breast cancer (one study suggests that a drink a day increases a woman’s risk by 10 percent).
But it’s hard for individuals to translate statistics to their own lives. A small increase in relative risk is difficult to make meaningful, even for people who understand what “relative risk” means. (It doesn’t mean a 10 percent risk of breast cancer; it means women who drink may be 10 percent more likely to get breast cancer than women who don’t.)
There are many other open questions that might seem important to a person deciding whether to change her habits: Is a glass of wine as carcinogenic as a daily martini? Does it matter how old you are when you start or stop drinking? And perhaps most important, do you lower your cancer risk if you quit drinking tomorrow, regardless of your age? The answers to all of these questions are unclear.
Translating statistical risk into practical advice will always run up against a person’s priorities and values, and different people might reasonably make different decisions. Some have interpreted the advisory as confirmation that no amount of drinking alcohol is safe — but I disagree. Evidence is information, not a prescription. For some people, fear of developing cancer might outweigh all other concerns. For others, the pleasure of drinking might outweigh their anxiety about a modest increase in long-term health outcomes. That’s not bad or self-destructive. It’s a personal decision about priorities.
When the alcohol advisory was released, some commenters responded with scorn. The libertarian magazine Reason published a response with the headline “Surgeon General Gins Up a Questionable Drinking Causes Cancer Scare.” In my own life, my husband rolled his eyes when I asked him what he thought about the advisory, and a group chat of women I participate in had a lively conversation about how the advisory seemed like “scolding B.S.” The handful of patients I’ve discussed it with said it didn’t make a difference to them.
The surgeon general’s report isn’t only, or even primarily, intended to speak to individual Americans, however. The majority of its recommendations, like the one to change alcohol labeling to highlight cancer risk, are policy ideas. The way that public health most effectively helps people change their habits is by changing the incentives, pressures and opportunities in the culture around them: Vaccine mandates help the hesitant decide to get shots, and speed limits help people inclined to drive fast to slow down. These types of policies are beneficially coercive, and they can evolve over time as people get used to new expectations and restrictions.
While some people in my life reacted to the advisory with derision or indifference, one friend saw the news and texted me in alarm. He drinks more than he’d like to. He asked what I thought and I told him that while I couldn’t confidently or precisely estimate his personal risk of getting cancer, cutting down on drinking was undoubtedly a good idea for his health if he wanted to try it. Since the advisory prompted our conversation, he says he’s begun drinking less, and even just a week in, he’s feeling good.
That even a majority of Americans might not be in the mood for the surgeon general’s advice may not matter. The first surgeon general’s report linking smoking and lung cancer was released in 1964, when smoking was ubiquitous; several others followed. Science is a body of knowledge built over time, and culture change is a long game. The alcohol advisory is likely only one in an evolving history of landmark public health documents about the dangers of drinking; it’s a document for the archives as well as the moment.