Share This Page
No animal ever invented anything so bad as drunkenness—or so good as drink.
—G. K. Chesterton
As hinted in the epigram by Chesterton, alcohol is a Janus-faced substance, and this two-sided nature seems to stem largely from its effects on the brain. In Western culture, alcohol is associated with celebration, imbibed during many rites of passage and family gatherings, and taken as a sacrament in Christian churches and Jewish temples and homes. Yet, alcohol can and does lead millions of people to severe addiction with life-destroying consequences.
Scientists who ﬁrst studied alcohol dubbed it a “dirty” drug. Unlike other psychoactive substances such as cocaine and heroin, alcohol did not seem to affect particular neurotransmitters and their receptors but instead to “soak” the entire brain. Only later did researchers begin to learn that alcohol could have speciﬁc receptor effects, and the focus of investigation shifted to understanding them. Now, it has been shown that alcohol has both far-ﬂung effects in the brain and highly targeted ones, further complicating an already complex research arena.
The effects of consuming alcohol are identiﬁable at every stage of life but can be radically different at different stages. Used by a woman during pregnancy, alcohol can profoundly impair fetal brain development, but, lately, having a glass of wine has even gotten some surprisingly good press. At ﬁrst, the data on possible health beneﬁts of consuming alcohol came haltingly, hedged with disclaimers, from scientists and physicians afraid to be perceived as approving a substance responsible each year in the United States alone for more than 100,000 deaths, a huge toll of disability, and billions of dollars in medical costs. But the reports on the health beneﬁts of drinking alcohol for some people, in strictly limited amounts, keep coming and are more speciﬁc, with information on the particular substances involved and possible mechanisms of action.
By ﬁlling out the life-span proﬁle of alcohol use, scientists are getting a new handle on both the effects of alcohol and on the nature of brain development.
Drinking in the Womb
The effects of alcohol consumption begin in the womb, and this ﬁrst chapter of the alcohol story, as far as we know, is uniformly grim.
The human nervous system begins its life as a subset of stem cells, which differentiate themselves from other parts of the embryo at about two weeks after fertilization. At this time, the embryo has three distinct layers: the endoderm, or inner layer, which will primarily develop into organs such as the lungs and digestive system; the mesoderm, or middle layer, which will make up the cardiovascular system and many bones, connective tissues, and muscles; and the ectoderm, or outer layer, which will construct the skin, skull, and nervous system.
At about three weeks after conception, a set of cells in the mesoderm, called the notochord (which will ultimately become the spine), pushes parts of the ectoderm above it, moving it upward to produce two ridges. These ridges climb higher, like waves rising out of the ocean but facing each other.
When they peak, the two waves begin to curve inward. Eventually, the tips meet and form a tube. Many of the cells here will then become the precursors of the brain and spinal cord. Cells formed where the ridges touched for the ﬁrst time are called neural crest cells and can become either brain neurons or parts of the skull and connective tissue of the face.
This delicate process is so vulnerable to alcohol that researchers have found no level of alcohol consumption by an expectant mother that is without some effect on her baby. Unfortunately, according to a recent national survey, 9.8 percent of pregnant women reported using alcohol and 4.1 percent reported binge drinking in the month before the survey. Maternal alcohol use has varying results for the fetus. When the effects on the baby are the most severe, they are called fetal alcohol syndrome (FAS). Children who have some, but not all, of the clinical signs of FAS suffer from a variety of disorders such as alcohol-related neurodevelopmental disorder (ARND) and alcohol-related birth defects (ARBDs). As a group, these defects are called Fetal Alcohol Spectrum Disorders (FASD).
When people think of FAS, they may visualize a child with distinctive facial malformations: skin folds at the corner of the eyes, a noticeably smaller than normal head, an abbreviated nose, a thin upper lip, and a missing or indistinct philtrum (the central indentation between the nose and upper lip). These facial abnormalities are associated with FAS because the neural crest cells appear to be uniquely vulnerable to the toxic effects of alcohol. But the facial deformities warn of much deeper problems. Those neural crest cells that do not become the structure of the face and skull become brain cells, and these cells, too, can be killed or damaged by alcohol during fetal development.
FAS is the number one known preventable cause of mental retardation in the United States. The Centers for Disease Control and Prevention estimate 800 to 6,000 new cases of FAS a year, but, when we consider FASD as a whole, that ﬁgure jumps to 40,000 new cases a year. Although only 25 percent of people with full FAS and 10 percent of those with ARND are mentally retarded, a large proportion of these children will go on to have lifelong learning and behavioral problems. Many will never be able to live independently. Children with FAS tend to have attention difﬁculties, poor impulse control, associated behavior problems, and sensory disorders such as hearing loss. Interestingly, facial abnormalities appear only if a woman drinks heavily early in her pregnancy. If she does not drink until later, this can still cause severe brain damage, but the disability will not be visible in the baby’s face, which had already developed when alcohol was introduced into the womb.
James West, M.D., who heads the department of anatomy and neurobiology at Texas A&M College of Medicine, describes these timing issues. “The brain is one of the ﬁrst organs to [begin to] develop and one of the last to be completed. Data say that [fetuses are] vulnerable throughout pregnancy. If anything, in the third trimester, they are most vulnerable.” The third trimester is the period of most rapid brain growth and also the period when connections between brain cells—synapses—are most rapidly forming. In its effects on the fetus, alcohol is relatively indiscriminate, says West. “It can kill cells after they begin forming. It can interfere with the cell cycle [which determines when cells divide and reproduce]. It can prevent the normal number of cells from being generated. It can cause some cells to migrate to the wrong places. It can kill cells that are already formed.”
As complex and multifaceted as they are, many of these effects are the last link in chains of events that start with just one or two speciﬁc actions of alcohol. Researchers have noted, for example, that the disabilities and deformities associated with FAS resemble those seen in children with genetic defects linked with a protein called L1 cell adhesion molecule. “There are a number of similarities in the brain lesions,” says Michael Charness, M.D., professor of neurology at Harvard Medical School. For example, in both FAS and what are called “L1 gene mutations,” the corpus callosum, which connects the two brain hemispheres, is either missing or severely undersized. Both conditions are marked by enlarged ventricles, the ﬂuid-ﬁlled spaces inside the brain, and by abnormal development of parts of the cerebellum.
Behavioral manifestations, including mental retardation and other cognitive and neurologic deﬁcits, are similar for the two types of disorders, although people with L1 mutations tend to have more severe retardation than people with FAS. This difference occurs because in FAS the protein is disrupted only while the child is in utero and exposed to alcohol, whereas, in the genetic disorder, the molecule fails to function throughout life. During development, L1 appears to serve as a kind of guide for neurons, telling them where to stick together and where their projections (axons and dendrites) should go. Alcohol strongly inhibits these functions, so the cells and their projections wind up in the wrong places with the wrong connections.
Protecting the Developing Brain
With the potential catastrophe of FAS looming, why do so many women continue to consume alcohol during pregnancy? The majority of women who give birth to full FAS babies are chronic alcoholics, often with life histories marred by violence, mental illness, personal tragedies, and acute stress. Many use illegal drugs as well as alcohol. A large percentage of these women are from backgrounds of poverty, but the statistics on this may be somewhat skewed; often, a diagnosis of FASD is not even considered for middle or upper class children.
Mothers I [Joseph Volpicelli] have seen who gave birth to FAS babies often were sexually abused as children, repeatedly raped, or witness to events such as seeing their fathers murder their mothers—experiences few people can imagine surviving. They do not stop drinking because for them life without anesthesia is experienced as unbearable and their own day-to-day emotional survival is all that they can consider. In less extreme cases of FASD, the mothers’ stories may be less extreme, as well. Ignorance of alcohol’s potential effects can play a role. But most mothers place their children’s health above their own; and those who can stop drinking while they are pregnant almost always do so.
For mothers who cannot stop, there may soon be hope for protecting the fetus. One avenue of prevention relies on our knowledge of how alcohol affects the action of L1. In 2003, Charness and his colleagues discovered that a molecule called NAP can block the disruptive effects of alcohol on the L1 protein during development and, at least in mice, mitigate much of the damage associated with fetal exposure to alcohol. Charness was amazed at how well NAP worked. “We were a little ﬂabbergasted that [the protein] would have such a complete effect on measures of developmental delay and embryotoxicity [killing cells in the embryo].” He hypothesizes that this effect occurs because the protein, by safeguarding the action of L1, interrupts a chain of events that otherwise results in large numbers of cells dying. In particular, because L1 enables cells to connect to each other, and cells that do not connect are programmed to kill themselves, preserving the activity of L1 may prevent much of the cell death that is caused in this indirect way by alcohol.
Even at best, though, NAP may be only part of the answer. Charness thinks that inhibiting the cell-connecting function of L1 is only one way that alcohol hinders fetal development. “It is unlikely to be the only story,” he says. Nonetheless, because there is no way at present to prevent any of the fetal harm caused when mothers do not stop drinking during pregnancy, if NAP were proven safe, it could be given to pregnant women at high risk to reduce the effects of their drinking on their children. The protein would be given to the pregnant mother as a drug, reaching the fetus through the placenta.
What about other systems known to be affected by alcohol but that probably do not involve L1? For example, alcohol interferes with the development of the serotonin system, which is believed to be involved with the regulation of mood and aggression. Like L1, the neurotransmitter serotonin plays a role both in development and in later brain activity. (In the case of serotonin, its developmental role is to promote the growth of some neurons and receptors.) Interestingly, a drug already marketed for the treatment of anxiety, buspirone, can prevent serotonin-related damage to the fetal brain caused by alcohol—at least in rodents.
Although any drinking during pregnancy seems to have measurable effects on the baby, the worst damage—both to fetuses and to the brain at any time of life—appears to be caused by binge drinking. In this context, binge drinking is deﬁned as heavy intermittent drinking in which large quantities of alcohol are rapidly consumed. The rapid rise in the blood alcohol concentrations of the binge drinker delivers high amounts to the brain, and these high concentrations seem to do the most harm. By contrast, consuming alcohol with food or drinking slowly do not seem to cause harmful amounts to reach vulnerable cells as readily.
From Infancy to Adolescence
Late in fetal development and just after birth, the human brain grows exuberantly. Aaron White, Ph.D., assistant research professor in psychiatry at Duke University, suggests that nature’s “strategy seems to be to create as many opportunities for circuits as possible. As soon as the baby is born, the brain seems to wire itself based on interaction with the outside world.” This wiring involves a ruthless pruning process in most areas of the brain, whereby millions of cells created during the period of extensive growth are killed off when they fail to make appropriate connections. Thus, in most parts of the brain, during most of childhood, brain cells and extraneous connections between them are culled, not created.
In one area, though, circumstances are much different. “In the frontal lobes, this overproduction continues through infancy and during the ﬁrst decade of life,” White says, explaining that, although extra cells are not produced at this time, the volume of tissue increases and millions of new synapses are formed.
That changes during adolescence. The teenage brain undergoes a burst of neural reorganization so profound that it is only paralleled by the massive overproduction and subsequent pruning that occurred in late fetal life and early infancy. As puberty starts, the pruning of the frontal lobes, the seat of planning, judgment, and self-control, commences. Not surprisingly, those are the areas in which adolescents are least skilled. To make matters potentially more alarming for Mom and Dad, during the teen years frontal development is accompanied by a change in the brain’s reward and pleasure centers that makes “routine” less satisfying. This change spurs teens to push away from their parents and gravitate to their peers. It also encourages them to take risks. All of this makes good sense, from the point of view of evolution, for a creature preparing to ﬁnd a mate who is not a close relative. “It has to do with getting kids to swim in the deeper end of the gene pool,” White suggests.
The immaturity of judgment, scorn of routine, penchant for risk taking, and orientation to peers may make adolescents especially susceptible to addiction, including alcoholism. The lessening ability to take pleasure in ordinary life may make highs more attractive, parental disapproval may add the thrill of the forbidden, and the lack of ability to defer gratiﬁcation may open the door to the repeated binging that can lead to addiction. Research ﬁnds that if people do not use alcohol or other addictive drugs heavily in their teens or early 20s they usually do not develop an addiction later in life.
Teens who drink heavily but do not become alcoholics may nevertheless be affecting brain circuits being laid down that will guide them throughout life. “The key point,” says White, “is that we’re not talking about the difference between brilliant and below-normal IQ for most people. If teens drink heavily during high school, they might be just ﬁne, but maybe they could have been ‘ﬁner.’ It may not be enough to mess up their future, but that doesn’t mean the effect isn’t there and there is certainly evidence that the effects are there.”
The amount, timing, and setting of drinking by young people are all relevant to the effects of alcohol on the brain. Teens who are exposed to low amounts in family settings, religious services, or on other occasions where drinking is moderate actually appear to be at lower risk of alcohol problems than teens never exposed to drinking or exposed at drinking parties with their peers. This effect is probably related to social factors, such as the modeling of moderation by parents. Teens who drink heavily—particularly in binges followed by periods of abstinence—from a young age are at much higher risk of becoming alcoholics. But here it has been difﬁcult to sort out cause and effect. Teens who start binging long before their peers (at ages such as 11 or 12 instead of 16 or 17) already tend to be predisposed to alcohol problems for reasons that go beyond starting to drink early. They often suffer depression or other mental illnesses, have parents who are alcoholic, and live in distressed neighborhoods where they are involved with antisocial behavior. In other words, whatever drives people to drink earlier and more than their peers, rather than the age of ﬁrst exposure to high amounts, could account for their later increased risk of alcoholism.
Recent brain research on rats, however, suggests that something more is going on; early exposure to high amounts of alcohol in and of itself seems to cause brain changes that make later heavy drinking more likely. It appears that, although adolescent rats are more susceptible than adult rats to the memory-impairing effects of alcohol, they are actually less affected by alcohol’s sedative and coordination-reducing qualities than are adult rats. “Just naturally,” says White, “as rats progress through adolescence, something unfolds that makes them more vulnerable to the effects of alcohol on balance and motor coordination and more sensitive to the sedative effects of alcohol.”
This means, roughly, that after they graduate from Rat U. and get on with the main business of life—reproduction—rats ﬁnd that, just like adult humans, they simply cannot “drink the way they used to.” Instead of ﬁring them up, excessive drinking now makes them tired and uncoordinated. Yet here we encounter another complication. Binge drinking during adolescence seems to thwart this maturational pathway leading to the adult’s altered reaction to alcohol. In a fascinating study, White and his colleagues found that when adolescent rats are repeatedly exposed to high amounts of alcohol in a “binge” pattern, they do not outgrow their adolescent resistance to the balance and motor coordination or the sedative effects.
Could this help to explain why some heavy college drinkers move on to lifelong alcoholism? While their peers—either because they did not start to drink as early, did not binge as heavily or frequently, did not have a genetic susceptibility, or, most likely, some combination of factors—become less able to stay awake and coordinated during heavy binges, the alcoholics retain their adolescent ability to withstand those effects. This, in turn, could lessen their appreciation of the negative consequences associated with drinking and keep them drinking longer.
Further supporting this idea, research on alcoholism also ﬁnds that being able to “hold your liquor” or “having a hollow leg” puts people at high risk of the disorder and that the characteristic is highly heritable. In the late 1970s, Mark Schuckit, M.D., professor of psychiatry at the University of California, San Diego, and his colleagues proposed this as one important risk factor for alcoholism. His research and that of others has since borne it out. “Resistance to the effects of alcohol is an excellent predictor of later alcoholism,” says Schuckit.
Brain Changes in Alcoholism
Being resistant to the effects of alcohol is not the only genetic pathway to alcoholism: People whose genes predispose them to take more risks are more vulnerable, as are people with mental illnesses like depression and schizophrenia. In some case, though, alcoholism appears to be the result of environment alone: for example, chronic high stress, heavy exposure during adolescence, or the two in combination. Almost certainly, there are other as-yet-undiscovered genetic and environmental inﬂuences at work.
Research ﬁnds that children of alcoholics are about four times more likely to become alcoholics than are people whose parents do not have the disorder. If both parents are affected, the risk is six times as great. Interestingly, Schuckit found that children of alcoholics who were adopted and raised by nonalcoholic parents had an even higher risk of the disorder, although the heightening of risk was slight. He hypothesizes that perhaps because the children were not exposed to their biological parents’ problem, they did not recognize their own elevated risk and its potential consequences.
Schuckit and the National Institute on Alcoholism and Alcohol Abuse are involved in a project to tease out the various genes involved in alcoholism, especially looking for genes that determine how the brain reacts to alcohol. How does this popular substance cause a “high” and how might this experience lead to addiction in some people? Here again, the effects of alcohol on the brain are complex, with large, global effects and small, highly speciﬁc ones. Alcohol seems to simultaneously enhance the action of inhibitory circuits in the brain, mediated by the neurotransmitter gamma aminobutyric acid (GABA), while reducing the action of excitatory circuits, mediated by another neurotransmitter, glutamate. The net result of these conﬂicting inﬂuences is an overall slowing of brain activity that we may experience as sedation.
Alcohol also inﬂuences activity of the neurotransmitter serotonin, and, in combination with alcohol’s effects on GABA, this seems to affect the circuits in the center of the brain for yet another neurotransmitter, dopamine, which is associated with pleasure. Alcohol also may act directly on this reward center by affecting dopamine receptors themselves. The experience of the alcohol high is also mediated in part by opioid receptors, which are linked with the pain relief and euphoria associated with substances such as heroin.
This last connection has provided a lead for my own research [Joseph Volpicelli] at the University of Pennsylvania, where my colleagues and I have developed an opioid-blocking drug called naltrexone for the treatment of alcoholism. We have known for some time that rats, like humans, often consume more alcohol after stressful experiences. Back in the 1970s, we found that when we gave rats naltrexone after such an experience, they did not show this increased drinking. In humans, research by Christina Gianoulakis, Ph.D., found that drinkers with a family history of alcoholism reported higher levels of euphoria from drinks given in the laboratory than people without a family history of the disorder. But when the children of alcoholics were given naltrexone, they experienced the same lower level of euphoria from alcohol as the people without a family history. One can conclude from this that people with a predisposition to alcoholism experience more pleasure from drinking and that naltrexone could possibly normalize this excessive response, perhaps reducing their risk of drinking problems.
In 1994, the Food and Drug Administration (FDA) approved naltrexone for the treatment of alcoholism. In our ﬁrst clinical trial of the drug, we found that it cut down the number of drinking days of alcoholic subjects by half, as compared with subjects given a placebo. Serious relapse, deﬁned as having ﬁve or more drinks on one occasion, was also reduced by half, although minor slips were not reduced dramatically. Other studies since conﬁrmed our research.
It appears that naltrexone may reduce craving for the second or third or fourth or ﬁfth drink more than it reduces desire for the ﬁrst. Making imbibing alcohol less rewarding, it makes excessive drinking less attractive. But naltrexone is not a panacea. As with any drug, many people ﬁnd it does not work for them or has too many side effects. In 2004, the FDA approved another drug, acamprosate, for treating alcoholism; it seems to cut craving by reducing the overactivity of NMDA glutamate receptors. It works for about as many people as are helped by naltrexone and has a similar level of effectiveness. Both drugs work best if given in the context of a comprehensive treatment program, which helps people discover what triggers their excessive drinking, treats them for any co-occurring mental disorders, and helps them ﬁnd alternative sources of meaning and joy.
When the Brain’s Own Defense Becomes the Problem
There is still much we do not know about how alcohol affects the brain and what speciﬁc changes lead to alcoholism. Clearly, some people have genetically based differences in their brain chemistry that make them more prone to alcoholism. Stressful experiences also seem able to change the brain in ways that make heavy drinking more likely. More studies are ﬁnding long-lasting adaptations made in the brain’s reward center in response to repeated exposure to alcohol and other addictive substances. However, we still do not know which of these many factors are critical to addiction.
There is also no doubt that serious brain damage can result from alcoholism. Studies ﬁnd that the frontal lobes are especially vulnerable; they are visibly smaller in long-term alcoholics than in nondrinkers. Frontal lobe dysfunction is linked with poor judgment and lack of impulse control. Curiously, it appears that most of the damage to the brain caused by alcohol occurs during withdrawal from alcohol use, when the brain tries to adjust to the absence of alcohol. Even if someone drinks heavily for just one night, the brain begins to adapt to the presence of alcohol. When it is no longer present, these adaptations themselves can be harmful to the brain. White compares the situation to a tug of war between brain and alcohol: “When the buzz wears off, it’s as if the other side lets go of the rope. There ensues a hyperactive phase [of continued adaptation but in the absence of alcohol] in some regions such as the hippocampus, where the over-activity can be neurotoxic.” The hippocampus is critical to memory, so blackouts from drinking probably stem from interference of the alcohol here. Hangovers, White explains, are signals that acute tolerance and withdrawal have occurred; they may, in fact, be a sign that damage has been sustained.
Studies of teen alcoholics by Susan Tapert, Ph.D., and her colleagues found that repeated withdrawals were associated with memory problems. Other researchers found that adult alcoholics who had been through repeated detoxiﬁcation attempts performed more poorly on cognitive tests of frontal lobe function than those who had not been repeatedly detoxiﬁed. One implication of this for treatment is that detoxiﬁcation should be done as slowly and gently as possible, with medications such as antianxiety drugs that block the rebound effect. Unfortunately, current practice in most detoxiﬁcation programs emphasizes withholding antianxiety medications for fear of creating alternate addictions. But if the anxiety and sleep problems reﬂect “overshoot,” because the brain’s counterpunch to alcohol is now occurring in the absence of alcohol, then relieving them could not only make detoxiﬁcation more comfortable but also could potentially mitigate associated brain damage.
The Positive Side of Drinking
Perhaps Chesterton, quoted at the beginning of this article, was missing something when he wrote, in another context, “Men grow too old for love, my love, men grow too old for wine…” Excessive drinking is unquestionably a serious assault on the brain, but for those who can drink moderately, particularly as they grow older, research suggests that alcohol may help protect against the most common types of dementia and strokes. Kenneth Mukamal, M.D., assistant professor of medicine at Harvard, comments that “in the most recent studies, the ﬁndings are relatively consistent. There is a lower risk of dementia and generally a lower risk of cognitive decline with light to moderate drinking.” In comparison to both heavy drinkers and teetotalers, light-to-moderate drinkers have the lowest risk of both common types of dementia, Alzheimer’s disease and vascular dementia and stroke.
In a study published in the Journal of the American Medical Association in 2003, Mukamal and his colleagues reported that people who drink in moderation—from one to six drinks a week—had a 46 percent reduction in the risk of dementia as compared with people who did not drink. On the other hand, people who drank more than 14 drinks a week were 88 percent more likely to develop dementia than abstainers. Alcohol dose appears to be a critical variable.
In another study, Mukamal and his colleagues found that levels of brain atrophy rose with increased exposure to alcohol, even at the lowest doses, but brain lesions linked with certain kinds of strokes and dementia were reduced in moderate drinkers compared with abstainers and heavy drinkers.
Some studies show extra beneﬁts from red wine, which is rich in antioxidants, while others do not. Because these are not controlled studies, it is also possible that other factors that determine whether and what a person chooses to drink (such as social class or economic level), could actually account for what looks like protective inﬂuences from alcohol use. The 2003 Mukamal study, however, did control for variables such as education level and still found signiﬁcant beneﬁts for health from moderate drinking.
In January 2005, research reported in the New England Journal of Medicine and involving the famous Nurses’ Health Study found that consuming about one drink a day of wine, beer, or hard liquor had signiﬁcant beneﬁts in maintaining the mental acuity of older women. A second study, published in February 2005 by scientists at Wake Forest University, found similar beneﬁts. The lead author of the ﬁrst study, Meir J. Stampfer, M.D., told The New York Times that the ﬁndings had important implications for women’s health if alcohol could be recommended in a way that did not lead to abuse.
Although there is some conﬂicting data among the various studies about dosage (higher levels of moderate drinking may be better for the heart, while lower levels in the moderate range may be best for the brain), alcohol seems to protect the brain via its effects on the cardiovascular system. Moderate amounts of alcohol increase good cholesterol, which lowers the risk of both heart disease and ischemic stroke. Ischemic stroke, which is caused by blood clots and responsible for four-ﬁfths of strokes in the United States, can itself cause catastrophic brain damage. Smaller ischemic strokes, which may not even be noticeable at the time to their victims or those around them, can ultimately cause vascular dementia, in which many minor events add up to serious devastation. These small strokes also increase the risk of Alzheimer’s disease and may be involved directly in its pathology.
Light-to-moderate drinking, these studies suggest, could also reduce the risk of ischemic stroke and dementia by directly thinning the blood. Unfortunately, this increases the risk of the other kind of stroke, hemorrhagic stroke, in which blood vessels in the brain leak. And heavy drinking, which raises blood pressure, as well, dramatically increases the risk of hemorrhagic stroke.
Again and again, we see the complex effects on the brain and body. Indeed, those effects are so varied that at this point physicians do not recommend “medicinal drinking” to try to prevent dementia and heart disease. People who already drink moderately and can remain moderate may enjoy possible beneﬁts; people who habitually abstain should not start for supposed medical reasons; and heavy drinkers should cut back or quit because the harm outweighs any potential beneﬁts.
The reluctance to recommend drinking for one’s health surely reﬂects cultural and moral biases that things good for us and also pleasurable are rarely found in combination. At the same time, though, that reluctance reﬂects caution arising from the state of the science on this ancient and paradoxical substance, alcohol.