Almost everyone has an apocryphal story about a friend or relative who has only "one drink a day" which turns out to be a liter of vodka. The definition of a "drink" confounds researchers who conduct epidemiological studies of drinking patterns. These self-reported statistics are usually under-reported although not usually to the same extent as the liter of vodka in the example above.
It's important to recognize that the definition of "moderate" as it relates to your maximum health may be too much for some people.
Moderate alcohol consumption is that range that maximizes the known cardio-protective and other health benefits without substantially increasing the risk from other factors. Put another way, moderate consumption should be at the level where your risk of dying from all causes is at its lowest.
Physicians and medical researchers say that two or three "drinks" per day falls into the moderate category without argument.
A 1991 study done at the Harvard School of Public Health by a team led by Eric Rimm found that people who averaged up to half a drink per day decreased their risk of heart attacks by only 1 percent -- essentially the same as abstainers. From one half to one drinks per day decreased it 21 percent; one to one and a half drinks, decreased risk by 32 percent and those who drank four to five or more drinks per day increased their risk by 50 percent.
But most medical researchers agree that four to five drinks per day push the envelope of moderation into an area where body size and other personal characteristics become more important.
A 1989 study of the relationship of alcohol use and later hospitalization, conducted by the Kaiser Permanente Medical Center in Oakland, Calif., found that of the 82,430 people studied, people who drank more than one but less than three drinks per day spent the least time in the hospital.
The study's authors, Mary Ann Armstrong and Dr. Arthur L. Klatsky, M.D., concluded that, "This study suggests that alcoholic beverage use has little impact on total hospitalization...In general, the favorable experience of the numerous lighter drinkers balances the unfavorable experience of the heavier drinkers and ex-drinkers."
In this study, heavier drinkers are those who consume more than three drinks per day. Like most studies done in the past five years, this one corrected its data analysis for ex-abusers who no longer drink because of poor health. It also suggested that under-reporting of alcohol consumption by participants could have biased the results.
Dr. Curtis Ellison, Chief of the Department of Preventive Medicine and Epidemiology at the Boston University School of Medicine, wrote in the September 1990 issue of Epidemiology "...given that participants in epidemiological studies tend to underestimate their usual amounts of alcohol intake, the actual amounts associated with decreased risk could be somewhat greater."
Most medical authorities today say that "moderate" consumption lies between one and five drinks per day and that beneficial effects fade rapidly at less than one drink per day.
Several other studies, however, have suggested that the definition of moderate may be higher. The phenomenon of under-reporting may mean that the wide gaps in the definition of moderate are reconcilable.
The key to interpreting this data is trying to standardize the alcohol content of a "drink." While scientists have generally defined a drink as 8 to 10 grams of alcohol, many researchers feel that with the emotional (and often guilt-associated) nature of alcohol, the under-reporting phenomenon affects not only reports of drinking frequency but amounts. For this reason, the drink that people report may be larger and contain substantially more alcohol than the 8 to 10 grams which is recorded by the scientific study.
According to a 1977 article in the Johns Hopkins Medical Journal, chronic ill effects are rare below a daily intake of 80 grams of pure alcohol although this amount varies according to the sex, body weight and other individual differences of the consumer. This parallels a statement from the British Royal College of Psychiatrists whose 1976 report, Alcohol and Alcoholism, estimated the upper limit of moderation as eight drinks per day, now generally accepted as too high.
The Johns Hopkins article, on the other hand, proposed that to qualify as a moderate drinker, a person should consume no more than 0.8 grams of alcohol per kilogram of body weight on any given day and no more than a daily average 0.7 grams per kilogram of body weight in any three-day period.
A kilogram is approximately 2.2 pounds; there are approximately 12 grams of alcohol in 4 ounces of wine, 12 ounces of beer or a mixed drink with 1 ounce of 80-proof distilled spirits. In research circles, 10 to 12 grams of alcohol is considered a "drink."
Let's see what that might mean: Example 1: A 180-pound man weighs about 82 kilograms; 0.7 times 82 yields 57 grams of alcohol (or 5.75 drinks) as the upper limit of daily consumption for this drinker to be considered moderate. Example 2: A 120-pound woman weighs 54.5 kilograms; 0.7 times 54.5 yields about 38 grams of alcohol or just over three drinks per day.
Women have about 10 percent less blood and other fluids than men of the same size, as well as a higher percentage of body weight as fat (25 percent versus 10 percent). Since alcohol does not concentrate in fat, this means that a woman drinking the same amount of alcohol as a man of the same weight, will have a higher blood alcohol concentration than the man.
In addition, one controversial study suggests that women create less of the enzyme that digests alcohol than men of the same weight. But conclusions cannot be made on the basis of a single study.
However, based on size differences and body composition, moderate for women will be about 25 to 30 percent less than for a man of the same size.
It's important to remember that these Johns Hopkins figures should be considered the upper limit of moderate consumption; indeed, many authorities would consider this limit as overlapping somewhat with the heavy drinking category.
This range may or may not be too high. On the one hand, most epidemiological studies put the maximum cardio-protection (and minimum cancer and cirrhosis increases) at between one and three drinks per day. However, medical experts have shown that consumption is typically under-reported with the result that the actual optimum health effects may be closer to the Johns Hopkins article than might be apparent on its face.
Some of the best evidence to support this under-reporting hypothesis comes from Rimm's 1991 Harvard study. He studied physicians who, it is assumed, are more aware of scientific methods and the need to report accurately. Rimm's research found that risks of death from heart attacks continued to decrease even when daily consumption exceeded 50 grams per day. Rimm said in a telephone conversation that he believes his reported consumption figures are probably the most accurate of all the epidemiological studies.
Since heavy drinking is known to increase hypertension (high blood pressure) and since hypertension increases the risks of heart disease, the fact that heart attack deaths continued to decrease with consumption of more than 50 grams of alcohol per day tends to support the Johns Hopkins somewhat higher definition of moderation.
Under-reporting may mean we need to re-define moderate drinking.
Most epidemiological studies indicate that the decrease in heart disease deaths starts to level off at about two drinks per day and begins to increase thereafter, eventually catching up and surpassing abstainers. But Rimm found steady decreases at four or more drinks per day with no sign that heart attack death rates were rising.
This is an indication that alcohol consumption in the other studies may have been under-reported since they showed cardiac death rates beginning to increase at two or three drinks (where death rates in the people Rimm studied were still decreasing.)
This could mean that in those other studies, a reported two glasses per day might, in reality, reflect the actual consumption of three or four. In other words, people who said they were consuming two drinks per day (25 grams) but who were actually drinking three or four, artificially biased the results toward the low end of the scale. The J-shaped curve is still valid and doesn't change its shape, but the number of drinks that define each point on the curve may need to be doubled.
If this is true, then the alcohol consumption levels at which cancers and cirrhosis begin to increase may be likewise higher than the earlier studies indicated. That means these negative health effects would not begin to show up until consumption was much higher than is now thought. Overall, this would shift the optimum consumption ahead by one or two drinks, putting it somewhere between one and five per day, which is consistent with a large number of medical studies.
But, since Rimm's study, with presumably more accurate subjects included only heart disease and alcohol consumption and did not also study cancer and cirrhosis, this can only be called an hypothesis.
The Johns Hopkins information should be considered an upper limit only -- one which should not be approached on a daily basis.
A better rule-of-thumb is the generally accepted, 25 gram (two drink) limit. Indeed, while Rimm did not define moderate in the Harvard study, when asked, he said that 25 grams (two drinks) per day was a "conservative" definition.
While "drinks per day" is the most commonly used indicator to measure health effects, most researchers hypothesize that peak blood alcohol levels may be an even more important indicator of health effects. As you can read in this book's chapter on liver disease and cirrhosis, the liver's inability to metabolize alcohol at a rate that keeps it from accumulating in the blood is reached at a blood alcohol concentration of about 46 mg/dl (.046 BAC).
While still a theory, this could be the line that separates the good health effects from the bad. It would explain why weekend binge drinking patterns (seen in many societies) are less healthy than drinking the same amount of alcohol, but doing it a little each day. This could also answer why wine -- which is most often consumed with meals -- may also be healthier than beer and spirits.
Certainly 0.046 is a good point to stay below: it seems to be the liver's saturation point and it's about half of the 0.08 to 0.10 maximum limits that most states use to determine legal intoxication for drivers.
The blood alcohol concentration level theory is just that, not proven fact. As you've seen, it has been very hard to conduct accurate epidemiological studies of large numbers of people because of under-reporting and other errors. And while relating health effects to blood alcohol levels would make this much more precise, it is impractical to even consider keeping constant track of peak blood alcohol levels in thousands of people. And to date, no research has been funded -- either by the government, university or private sources -- to investigate this link.
That's why, for your everyday decisions, the 25-gram level (two drinks) is so important: the overall body of the best science shows that this level definitely puts you into the lower cardiac- risk category (even if the under-reporting hypothesis is true) without expanding the limits of moderation.
It's important to note here that many alcohol control advocates who oppose any form of alcohol consumption, frequently distort research data, quote results out of context and rely upon flawed research to support their ideological cause. As stated above (and elsewhere in this book) scientifically valid and intellectually honest conclusions cannot be based on a single study. Data must be considered in the context of the whole body of knowledge and not taken out of context or over-extrapolated to unsupported conclusions.
This is said to prepare you for alcohol control advocates who will cite pieces of seemingly credible research to "prove" that any alcohol at all is harmful and that alcohol abuse starts at one drink per day. Interestingly enough, almost all of these advocates are not M.D.s; a great number of them are social scientists with a poor understanding of medical science and the scientific method. They seem to feel that it's intellectually acceptable to use bits and pieces of research in order to support their social or ideological goals. The overall body of research simply does not support their positions.
Significantly, the people behind the science of alcohol consumption, who recognize the beneficial effects of moderate alcohol consumption, are mostly physicians and scientists who are trained to interpret scientific data. These people are cautious and unemotional, and typically don't project well on television. This may be one good reason why their messages are so often unheard or drowned out in the din of anti-alcohol fervor.
Dr. John P. Callan, M.D., a widely-published Illinois psychiatrist with extensive experience in treating chemical dependency (including alcoholism) writes in the Winter 1992 issue of the journal, Priorities: "These Neoprohibitionistic `drys' are not only generating intemperate propaganda against overindulgence, but also against the publication of healthful effects of moderate drinking. Alcohol has been shown, in reputable scientific studies, to be beneficial for good health. The one-sided attacks should not go unanswered. Legitimate voices should speak out lest unscientific opinion prevail and Prohibition be re-enacted."
"Safe" as applied to alcohol consumption must be defined in relation to: (1) here-and-now and (2) later.
The "later" definition applies to lifetime consumption and drinking patterns. The overall body of scientific research shows that people who are regular, moderate (two drinks per day) consumers of alcohol have an approximately 10 percent lower risk of dying from all causes (and 40 percent less heart attack risk) than either heavy drinkers or abstainers. Put another way: the risk of dying -- period -- is 100 percent for all of us; moderate drinkers delay the inevitable and have a greater likelihood of living longer.
Here-and-now, things are a bit different. The single most serious result of drinking too much is death -- most often from automobile accidents. It is unsafe to drive drunk and the safest rule is not to drive at all when you drink.
A driver should never even approach the 0.08 to 0.10 blood alcohol concentration (BAC) which determines the levels of legal intoxication or impairment. For long-term health reasons, people should keep their BAC at 0.05 or below and for the purposes of driving, much lower than that. Alcohol's effects vary enormously from one person to another. But in general, people with a BAC of:
-- 0.02 to 0.03 begin to experience changes in behavior, coordination and mental acuity;
-- 0.05 may begin to feel tranquil or sleepy. The key liver enzyme that metabolizes alcohol becomes saturated at about 0.046. Above this level, BAC will continue to increase; below this level, the liver will keep the BAC from rising;
-- 0.05 to 0.15 experience a steady loss of coordination;
-- 0.15 to 0.20 are obviously intoxicated and may show signs of delirium;
-- 0.30 and 0.40 usually lose consciousness ("pass out");
-- 0.40 to 0.50 frequently die when their heart and respiration fail.
Alcoholics and chronic abusers will experience this same progression of symptoms, but usually at much higher BAC levels.
Charts which purport to show "safe" levels for people of various weights are inherently misleading because so many variables affect the intake of alcohol and their effects on a given individual. Most of these charts are based on an "average" 155 pound man and indicate that one drink per hour will keep BAC at a "safe" level. This may or may not be safe depending on the person and the person's sex (ovulation, menstruation and other factors further complicate things for women).
To be absolutely "safe" requires a designated driver or public transport.
Blood alcohol levels can sneak up on people, partly because the pleasure of intoxication makes many people ignore alcohol's effects and partly because most people have no way of relating the degree of intoxication they feel with the actual BAC. What is your BAC after a glass of wine; after two glasses? Knowing how your body handles alcohol is a vital part of not drinking too much -- both in terms of enhancing long-term health, and preventing short-term tragedies like drunk driving.
All alcohol is not equal.
Some alcoholic beverages are absorbed more quickly than others in ways that can significantly affect how much makes it into your bloodstream and how quickly it gets there.
Some factors are inherent in the drink itself.
-- Drinks between 15 and 30 percent alcohol are absorbed most quickly. This includes high-percentage table wines and most dessert wines, ports, madeiras and fortified wines.
-- Drinks with more than 30 percent alcohol (6O proof) tend to be somewhat delayed because the high alcohol concentration irritates the stomach lining, producing a protective layer of mucus that delays absorption.
-- Some research shows that carbon dioxide content induces the stomach to pass its contents more rapidly on to the intestines (which absorb alcohol more readily). Thus, beer, champagne or drinks with carbonated mixers may cause more rapid intoxication than those of the same alcohol content which lack carbon dioxide.
-- Sugar content will slow down absorption.
-- Eating food along with alcohol delays absorption.
-- High concentrations of non-alcohol organic compounds delay absorption. Wine, especially red wine, is absorbed more slowly than ethanol/water combinations of the same alcohol percentage because of the significant concentrations of hundreds of organic compounds. Likewise, traditionally brewed beer has very high concentrations of proteins, B-vitamins and complex carbohydrates. Unfortunately, in the pursuit of crystal-clear beer, most commercially available brews undergo intensive filtration that removes most of the proteins, vitamins and other organic compounds which are both healthy and retard alcohol absorption. Unmixed distilled spirits are basically water and alcohol with insignificant amounts of other organic compounds.
In addition to absorption factors inherent in the drink itself are myriad human factors that determine blood alcohol concentration. Among them:
-- Body mass; the same drink will produce a lower BAC in bigger people than in smaller ones.
-- Body fat; alcohol is not absorbed well by fat, so a fat person will tend to have a higher BAC for a given consumption level than a lean person of the same weight.
-- Sex; in addition to body mass and fat composition, a woman will experience different levels of absorption at different times in her menstrual cycle -- frequently with peaks at ovulation and at the onset of menstruation.
-- Food consumption; the more you eat with alcohol, the slower its absorption.
-- Illness; diseases can alter alcohol absorption.
-- Drugs; many pharmacueticals can alter alcohol metabolism. Tagamet and other ulcer medications can increase BAC rapidly; sedatives combined with alcohol abuse can kill. The effect of illegal drugs can be enhanced by alcohol; this combination can also kill.
-- Biorhythms; morning drinking produces a higher peak level of alcohol in the blood, but also disappears faster since the liver metabolizes it more quickly earlier in the day.