In Temperance America and beyond, it seems no amount of evidence will be accepted as proving the health benefits of consuming beverage alcohol.
The Global Burden of Disease Study of alcohol, the latest in a series finding alcohol’s net effects worldwide to be destructive, calculates the optimal level of drinking to be exactly zero—a figure that embodies the concept “neo-Temperance” and reflects the views of its originators (we’ll review it later).
This is the vision of alcohol preferred by many in the political and public health arenas. A recent Mother Jones piece by Stephanie Mencimer (whose article on the subject I address here) declared: “Cancer is only one of the ways it [alcohol] can kill you. Drunk driving, alcohol poisoning, injuries, domestic violence, liver disease.”
My goodness—that seems like a substance to avoid at all costs! Notice that neither source mentions the overwhelming reason people offer for drinking: pleasure. The concept of pleasure is always treated negatively in classic Temperance literature, and doesn’t occur at all in epidemiology.
But what if all these ills—whose existence I don’t deny—were balanced, even outweighed, by benefits far more respectable than pleasure, those to our physical health?
By contrast, the 2010 U.S. Dietary Guidelines for Americans, published jointly and regularly by the Departments of Agriculture and Health and Human Services—and despite constant political pressure to warn against drinking—released this description of alcohol:
“there is strong evidence showing that moderate alcohol consumption is associated with a lower risk of cardiovascular disease. Moderate alcohol consumption also is associated with reduced risk of all-cause mortality among middle-aged and older adults and may also help to keep cognitive function intact with age.”
This radical notion is surely worthy of investigation. So here goes my further effort to resolve the irresolvable on the benefits of individuals drinking. I’ll examine five false assertions made by the heirs of Temperance, with reference to the most relevant research.
Myth 1: Research showing alcohol’s benefits is overwhelming funded and spurred by the alcohol industry.
While Mencimer and others cite books and individual research studies funded by alcohol producers, these comprise a mere handful out of hundreds. In 2011, the British Medical Journal published a meta-analysis of 84 studies of alcohol and cardiovascular disease and death that included two million men and women.
Remarkably, the BMJ investigators selected this research as the best among 4,235 studies they reviewed. These studies included scores of rock-solid government and nonprofit-funded research projects that identify alcohol’s benefits. Indeed, their findings often distress the anti-alcohol funders of these projects. Collectively the studies found a 25 percent reduction in death from heart disease, as calculated by the Harvard Health Blog.
Proponents of campaigns against alcohol (like Mencimer) routinely fail to cite any of these “pro-alcohol” studies.
I describe in my response to the Mother Jones piece how the government’s Dietary Guidelines for Americans struggled for decades to embody the state of the research indicating that alcohol lowers heart disease rates (and those of other cardiovascular conditions, such as stroke) and overall mortality, which it finally admitted in 2010.
In that article, I also cite the famed Framingham Heart Study, whose government funding agency refused to allow the researchers to publish, and is still reticent about, what the Framingham investigators concluded after a half-century of research:
“In the community, alcohol consumption is not associated with increased risk for congestive heart failure, even among heavy drinkers [more than 15 drinks weekly for men and eight for women]. To the contrary, when consumed in moderation, alcohol appears to protect against congestive heart failure.”
In the U.S. Government’s view, the public should never be informed of such results.
Now, in addition to the Dietary Guidelines and Framingham, here are three other extremely important studies by groups and organizations—sources that, far from being alcohol industry stooges, are instead antagonistic to the industry.
The summation below of the large bodies of research investigating alcohol’s effects on heart disease and dementia was written in 2009 by a consortium of researchers under the aegis of the Research Society on Alcoholism in the journal, Alcoholism: Clinical and Experimental Research. As the term “alcoholism” in both the organization’s and the journal’s names indicates, these are not groups devoted to touting alcohol’s benefits.
“A considerable body of epidemiology associates moderate alcohol consumption with significantly reduced risks of coronary heart disease. . . . Also, brain functional comparisons between older moderate alcohol consumers and nondrinkers have received more recent epidemiological study. In over half of nearly 45 reports since the early 1990s, significantly reduced risks of cognitive loss or dementia in moderate, non-binge consumers of alcohol (wine, beer, liquor) have been observed. . . .
This group of physiological researchers explained these accumulated results:
“moderate alcohol exposure appears to trigger analogous mild stress‐associated, anti‐inflammatory mechanisms in the heart, vasculature, and brain that tend to promote cellular survival pathways.”
Let’s turn now to a fourth such resource, the largest prospective study—following a single group of people over years—of alcohol consumption ever conducted, with roughly half a million Americans, ages 50 and up, at risk for heart disease or not, published in the New England Journal of Medicine in 1997.
The study found that abstainers had the highest death rate, while “this rate was lowest among men and women reporting about one drink daily.” (See tables, one for men and women, in Figure 1.)
Mencimer and others draw parallels between the obfuscation of research by American tobacco companies and alcohol producers. Yet the research in the NEJM report was conducted by Michael Thun for the American Cancer Society. Thun made his bones conducting anti-cigarette cancer research for the Society. He and his funders were expecting, or at least hoping, to point the finger of death at alcohol in the same way they had with cigarettes.
Myth 2: Alcohol outcome research and researchers understate or ignore breast cancer.
You don’t miss outcomes like dying as you follow people’s lives in studies like that by the American Cancer Society. And Thun et al. did find increased breast cancer death rates for women drinkers, which are reported in the lower table under “alcohol-augmented conditions” in Figure 1.
It’s just that breast cancer deaths paled in comparison to the death rates from heart disease for these women.* I’m not minimizing death due to breast cancer. But summing deaths from all causes due to different levels of drinking, Thun and ACS found that a woman’s odds of living longer are improved if she drinks.
Myth 3: Abstainers from alcohol are usually ill, so it is wrong to conclude that abstinence causes higher mortality rates.
In her Mother Jones piece, Mencimer doubts the existence of the J-shaped heart disease/mortality curve—which is received opinion across a broad swath of heart researchers—in which abstainers show a slightly elevated death rate, light-to-moderate drinkers a somewhat reduced death rate, and then heavier drinkers (more than 1-2 drinks daily) a sharply rising death rate. Mencimer rejects the early bump in the curve because she believes that people assigned to the abstainer group quit drinking due to pre-existing illnesses.
This legitimate concern mirrors research conducted by British investigator A. Gerald Shaper in the 1970s. He, as Mencimer summarizes, found that: “71% of nondrinkers in the study were actually former drinkers who had quit,” most due to illness. The percentage of quitters that is sick is higher in the UK than the US, since the UK has fewer ideological teetotalers. Nonetheless, for Shaper and now Mencimer, it was unfair to tag their abstinence as having any role in their earlier deaths.
In making this assertion in regard to all research since then, anti-alcohol campaigners dismiss scientists as educated idiots, incapable of taking such a prominently identified factor into account.
Instead, researchers—who as a group are timorous about, or resistant to, positive alcohol findings—either eliminate sick ex-drinkers or, if their sample is large enough, include only lifelong abstainers.
A fifth example of current research about drinking and mortality—the second largest single American study ever conducted of life, death, and drinking—separated lifetime abstainers from those who had quit drinking for comparison with drinkers. This investigation, described in an editorial and press release by the American College of Cardiology in 2017, is based on data from 333,247 participants in the U.S. National Health Interview Surveys from 1997 to 2009.
The title of the editorial—“Light-to-moderate alcohol consumption may have protective heart effects”—expresses caution in a way that does not fully reflect the research results. The researchers likewise stated their conclusions in the most cautious way possible, while warning against exceeding moderate drinking limits:
“Light and moderate alcohol intake might have a protective effect (relative to lifetime abstainers in this study) on all-cause and CVD-specific (heart) mortality in U.S. adults. Heavy or binge drinking was associated with increased risk of all-cause and cancer-specific mortality.”
Myth 4: Many drinkers, especially women, consume more than recommended limits, so in practice they would be better off abstaining.
And now let’s take such caution—expressed for example in the National Health Interviews Survey analysis, and understandable though it is—to task.
Anti-alcohol claimants are right, in a sense. There is no J-shaped mortality curve with drinking (meaning there is a modest rise in death rates for abstainers and a radical rise for heavier drinkers). But this is the case for a completely different reason than they assert—because in fact, even heavier drinkers don’t die more rapidly than abstainers.
Here’s that radical discovery by the Framingham researchers:
“In the community, alcohol consumption is not associated with increased risk for congestive heart failure, even among heavy drinkers [more than 15 drinks weekly for men and 8 for women].” [My emphasis]
In a parallel case, here’s the stated conclusion of the American Cancer Society article in the New England Journal of Medicine: “Mortality from all causes increased with heavier drinking [i.e., more than one drink daily].”
This statement is true. But, if we return to the tables in Figure I, we see something else.
The data aren’t J-shaped, but rather like an elongated U in this immense study. Although the lowest mortality rate was at one drink daily, that rate increased only slightly at higher levels of drinking. The highest level of drinking that could be assessed reliably was six and more drinks daily, combined into a single category. This group of heaviest drinkers, both men and women, died more frequently than lower-level drinkers. Yet the heaviest drinkers in the study, male and female, nonetheless died at a slightly lower rate than abstainers.
Let’s return to National Health Interview Surveys of a third of a million of American men and women from 1997 to 2009, as its results on drinking and mortality were summarized by the American College of Cardiology.
First came the good news for moderate drinking relative to lifetime abstinence:
“Moderate drinking was associated with a 13 percent and 25 percent decreased risk of all-cause mortality, and 21 percent and 34 percent decreased risk of cardiovascular disease mortality, respectively, in both men and women.”
That is, women benefited more from moderate drinking in terms of reduced death rates than men.
Now, onto heavier drinking:
“Researchers found that male heavy drinkers had a 25 percent increased risk of mortality (compared to lifetime abstainers) due to all-causes and a 67 percent increase in mortality from cancer. These increases were not significantly noticed in women.” (My emphasis)
Needless to say, neither the American College of Cardiology nor the research on which their analysis is based, the National Health Interview Surveys, are industry-funded. Nonetheless, in this huge government study, significant increases in overall mortality weren’t noted for women due to heavier drinking, and moderate drinking was found to reduce overall mortality significantly for both men and women, but more so for women.
In order to make their key point that subjugation to the alcohol industry causes researchers to understate alcohol’s dangers, the anti-alcohol lobby must not only ignore crucial, landmark data, but also malign those who generate the research, including—along with the government agencies responsible for Dietary Guidelines for Americans and the Framingham Heart Study—the American Cancer Society, the Research Society on Alcoholism, and the American College of Cardiology.
But what dangers and health risks is the anti-alcohol lobby itself short-shrifting? Let’s consider what it would mean to accept the view of reality generated by these unbiased research findings. If a woman, based on the Mencimer-Mother Jones analysis, concluded that she should stop drinking alcohol altogether, she would stand a greater risk of dying—according to the consensus of these five prominent research and health groups.
It is true that Mencimer’s chances of getting breast cancer would have been somewhat lower* if she hadn’t drunk alcohol at all, as she seems now to recommend. Nonetheless, two massive studies, conducted decades apart, along with other reliable evidence-based sources, tell us that women who drink alcohol—preferably moderately, but even if not perfectly moderately—don’t have higher death rates, and more usually reduce their chances of dying, than women who abstain from alcohol.
Myth 5: Promoting alarmist views of alcohol to curtail drinking, while discounting health and cultural benefits, increases safe drinking and enhances public health.
I certainly don’t recommend heavy drinking. My online Life Process Program (LPP) is dedicated to helping people move to moderate drinking, if they choose that goal. But I—and LPP—likewise discourage working from a place of fear. Because, even more than consumption levels, patterns of drinking determine health outcomes. That is, avoiding binges and unsafe behaviors like driving and fighting while drinking will save more people than cutting back their sum total of drinks.
The specter of dying due to exceeding minimal drinking levels is a Temperance shibboleth. In Temperance terms, the tippler inevitably becomes the drunkard. And the danger of drinkers becoming alcoholics is still claimed by those refusing to publicize alcohol’s benefits. In this world-view, even slightly exceeding recommended limits is lethally dangerous. Yet warnings like the Mother Jones piece are likely to cause some women to cease drinking altogether, with net negative results.
Such fear-mongering—which Mother Jones elsewhere deplores as a political tool—accomplishes more harm than good. You might think such irresponsible messaging would be decried. But, in Temperance America, you can never go wrong by attacking alcohol, however irrationally.
And what of the Global Burden study that I referenced at the start of this piece? As I describe here, the first systematic assessment of drinking across Europe showed that Northern European countries, while drinking less than Southern countries, exceeded them six-fold in alcohol-related deaths: 18 vs. 3 per 100,000 for men; 3 vs. 0.5 for women.
An independent Italian researcher, Franca Beccaria, has brilliantly described the warring worlds of alcohol regulation—dominated by the fear-based, highly regulated paradigm of Northern Europe and English-speaking countries, from where the Global Burden researchers hail—from the perspective of Italy, in which wine is viewed as a cultural and family heritage.
Beccaria challenges “the universal application of the public health approach in alcohol policy” (that is, limiting total consumption) while she offers “support for the drinking patterns paradigm and argues for an increased contribution from the social sciences to understanding and addressing the alcohol issue.” This approach is of one piece with the international harm reduction movement in regards to drugs. Her efforts with alcohol, alas, have been overrun.
The social historian Harry Levine notes that 19th century-era Temperance advocates despised no one more than the moderate drinker, who disproved their basic belief that alcohol was inherently evil. Neo-Temperance acolytes have now come full circle by denying that such a thing as moderate drinking exists. By obfuscating the value of moderate drinking and destroying it as a public health goal, the Global Burden research actually contributes to worsening drinking outcomes worldwide.
The Global Burden of Alcohol study homogenized data from a staggering 195 countries. Global’s chief author, Max Griswold, an American, decisively declared, “we found that the combined health risks associated with alcohol increase with any amount of alcohol.” But glomming together data from so many nations, many in the developing world, involving a multitude of unknown factors, doesn’t actually shed light on Americans’ decision-making around drinking.
Drinking by poorer people has more negative consequences than for better-off drinkers. In the U.S., where drinking (versus abstinence) is far more common among higher-income people, drinking is associated with smoking only for poorer and less well-educated drinkers. Likewise, worldwide, poorer countries have different disease profiles in relation to drinking. Walter Willett, professor of epidemiology and nutrition at the Harvard School of Public Health, argues: “Our decisions about drinking in the United States shouldn’t be influenced by what alcohol does to tuberculosis.”
The Global Burden study was not the only one combining subjects from different countries to appear in Lancet this year. Another study, based on 600,000 subjects from “high-income countries,” found (this result was highlighted in a critique of the Global Burden study by David Spiegelhalter) that, “compared to moderate drinkers, ‘never drinkers’ experience 30% more heart disease and strokes, and a 20% higher overall death rate.” These figures are close to the results of the 2011 BMJ meta-study reported in the Harvard Health Blog, involving two million men and women, that found a 29 percent reduction in the diagnosis of coronary artery disease and 25 percent reduction in coronary deaths for moderate drinkers.
These figures—moderate drinkers have 30 percent less coronary artery disease in middle age and 20 percent lower death rates—have been consistently replicated. They are the best estimates of the neighborhood of the health benefits of drinking alcohol.
But, recall my opening sentence: “No amount of evidence will be accepted as proving the health benefits of consuming beverage alcohol.”
Strange to say, none of the modern Temperance advocates I know follow their own advice to abstain. That they don’t actually makes sense since, as Gallup points out, “drinking is highest among educated, upper-income Americans.” And they’ll probably outlive abstainers. Might they take secret pleasure in doing what they would have others not do?
* This is because many more women, like men, die of heart disease than any other cause. Yet cancer is more feared. As columnist Mimi Swartz noted in an article titled “Our Coronary Complacency”: “One of the greatest terrors of women’s lives is breast cancer, which kills about one in 30 women a year, and yet one in every four female deaths stem from heart disease.”
Moreover, researchers summarized, “Alcohol consumption is associated with [only] a modest increase in breast cancer risk. . . .In comparative European research, 5 percent of breast cancer incidence was attributed to current or former alcohol consumption.” Furthermore, “the majority of the studies to date do not support increased breast cancer risk with higher alcohol use in women at high risk of breast cancer. . . . Studies also do not suggest that alcohol affects breast cancer recurrence or survival.” Nonetheless, these authors indicate, women in these situations “could benefit from staying within the drinking guidelines of ≤1 drink a day for overall health, and consider stopping entirely to reduce breast cancer risk.”
My Own Role in This Debate
Like many skirmishes around the globe, there is a war over the meaning, value, and consequences of drinking alcohol. It is a long-standing war, and I have been a participant in it for much of my going-on 50 years in the field.
One crucial component in this war is whether drinking a lot in itself causes alcoholism. For most in the field, dependence is “simply another term for a sufficient level of use of a substance,” as Robin Room, a progenitor of the limit-alcohol-consumption model of public health, maintains. For decades Room and I have engaged in high-visibility debates on this and related questions, in which he is called on to counteract my radical views, to wit: “The limitations of control-of-supply models for explaining and preventing alcoholism and drug addiction” (1987); (with Archie Brodsky) “Exploring psychological benefits associated with moderate alcohol use” (2000); “Alcohol as evil—Temperance and policy” (2010).
My fundamental argument—expressed in my 1985 book, The Meaning of Addiction—is that an addictive involvement inescapably requires that the involvement have a special personal meaning for the user, one that occurs in a specific cultural context.
Room is a charter member of the Kettil Bruun Society, an influential group of epidemiologists who seek to limit alcohol consumption. Room, an Australian, is the group’s honorary president. A sort of one-person international Temperance movement, Room previously directed government-funded research organizations in the U.S., Canada, and Sweden.
The current officers of KBS come from New Zealand, Australia, the U.S., another from Australia, and Norway. These neo-Temperance acolytes control the public health field in alcohol, with the brief intercession of Franca Beccaria as KBS president from 2011-2013. Franca, who (like me) holds no permanent academic or institutional position, was elected president after I forwarded her views on the KBS listserv, including her feeling that she and fellow Southern Europeans were disregarded by KBS.
The Global Burden of Disease anti-alcohol campaign is the apotheosis of the ideas of anti-alcohol epidemiologists. In 2018, they decided for the world that there is no safe way to consume alcohol, a substance present at every cradle of civilization and one that has been enjoyed by billions of people, usually benefiting their health in the bargain. These neo-Temperance scientists to all appearances despise and seek to end this eternal human relationship with alcohol, the most ubiquitous psychoactive substance, one that is used almost universally to celebrate joyous occasions. I, despite being American, appreciate and try to refine this relationship.
In 1993, I published in the American Journal of Public Health “The Conflict Between Public Health Goals and the Temperance Mentality,” which detailed how the health benefits of alcohol had been ignored in the U.S. This article, as has been nearly all of my work, was self-funded. In 1999, Archie Brodsky and I were given unrestricted grants by the Distilled Spirits Council (DISCUS) and Wine Institute to assess psychosocial benefits of drinking, an investigation published in 2000 in Drug and Alcohol Dependence as “Exploring Psychological Benefits Associated with Moderate Alcohol Use.” The main benefit Archie and I identified was better cognitive functioning in older moderate drinkers relative to abstainers. This finding has since been affirmed by the 2010 Dietary Guidelines for Americans and a 2009 consortium of the Research Society on Alcoholism. Our research led me to organize the 1998 conference and 1999 volume, Alcohol and Pleasure: A Public Health Perspective, funded by the International Center for Alcohol Policies, an alcohol industry group.
I haven’t received funding of any sort from the alcohol industry for many years.
Image: Cartoon by George Cruikshank, a supporter of the British Temperance Movement, published between 1830 and 1840, Via Library of Congress