Heavy Drinkers Outlive Nondrinkers

This is the abstract of the study

From the Results section of the abstract:

Controlling only for age and gender, compared to moderate drinkers, abstainers had a more than 2 times increased mortality risk, heavy drinkers had 70% increased risk, and light drinkers had 23% increased risk
Click to expand...

Correlation vs. causation debate anyone?

From the Time article in the OP

That said, the new study provides the strongest evidence yet that moderate drinking is not only fun but good for you.
Click to expand...

John Cloud, the author of this article, clearly thinks it's causation. Anyone care to take a jab at it?
 
There are a lot of serious health problems, such as diabetes and heart disease, that would make a person not drink. People who take antidepressants are told not to drink. Those people would increase death rates in the non-drinking group.
 
I'd be interested in seeing a study of the "skeptical" reactions to this study based on culture. Americans are still pretty uptight about alcohol whereas Europeans are not. I'd say my European friends and family members would say, "Duh" while my American friends and family would find it highly suspect since we're only a few generations away from prohibition.

The abstract seems sound. They point out that studies so far show a benefit, but the studies were inadequate in regards to confounding factors. Their study addressed those factors. They pointed out that when they adjusted for confounding factors the benefits decreased significantly but still showed a significant positive effect. Their results are consistent with many other studies.

As for correlation versus causation, alcohol quite obviously has an effect on the body. Nearly every organ is affected. There have been studies showing positive effects when it comes to cardiovascular disease. Since that's a leading cause of death, then as long as the benefits are not offset by some increased risk of another deadly disease, we could see a difference in mortality rates.

Of course, nothing about quality of life is mentioned. Alcohol could have all sorts of negative effects on health that don't increase mortality rates. A study like this doesn't address that.
 
Scott Haley said:
There are a lot of serious health problems, such as diabetes and heart disease, that would make a person not drink. People who take antidepressants are told not to drink. Those people would increase death rates in the non-drinking group.
Click to expand...

Yes, but (from the Results section again):

A model controlling for former problem drinking status, existing health problems...
Click to expand...

... they controlled for that.
 
Last edited:
Scott Haley said:
There are a lot of serious health problems, such as diabetes and heart disease, that would make a person not drink. People who take antidepressants are told not to drink. Those people would increase death rates in the non-drinking group.
Click to expand...

The abstract reads, "A model controlling for former problem drinking status, existing health problems, and key sociodemographic and social-behavioral factors, as well as for age and gender, substantially reduced the mortality effect for abstainers compared to moderate drinkers. However, even after adjusting for all covariates, abstainers and heavy drinkers continued to show increased mortality risks of 51 and 45%, respectively, compared to moderate drinkers."
 
UncaYimmy said:
As for correlation versus causation, alcohol quite obviously has an effect on the body.
Click to expand...

General life-style benefits causing light drinking and health-beneficial choices in general, maybe.

Of course, nothing about quality of life is mentioned. Alcohol could have all sorts of negative effects on health that don't increase mortality rates. A study like this doesn't address that.
Click to expand...

True. It did conclude that heavy drinkers and abstainers are both likely to hit the ground earlier than light drinkers, however. Presumably, heavy drinkers because of the negative effects of alcohol. I don't think that light drinkers would suffer from quality-of-life issue, IMO.
 
Last edited:
godofpie said:
I'm not able to view the study. Have you read it?
Click to expand...

I unfortunately do not have access to the paper. What I know is that there is much room for error in these alcohol studies. They simply cannot pin down causation; and the significant amount of uncertainty in the results suggests it's more of a correlation.

Alcohol's effects on the body and much of its effects on the brain are well documented. I have yet to see any scientific evidence suggesting that starting an alcohol habit is healthier than ending one.
 
I'll drink to that!

Other studies also show that drinking lowers the risk of becoming diabetic.

I'm diabetic II, and I drink. Usually a large glass of wine with dinner, then take insulin later. The wine seems to lower my sugar, about 25 points.

My best reading is that the concern about diabetics and alcohol is that, should a diabetic drink enough to lower his sugars excessively, others may dismiss a hypoglycemic episode as drunkenness, and not seek treatment for the low sugar.

So, drink or insulin, not both. Be a responsible drunk. ;)

Also, re the benefits of alcohol: fat burning and alcohol both give off acetaldehyde as a by-product. I believe this is a sign to the homeostatic systems that you have used up sugar, and to get back into balance, at which point the body does some kind of beneficial tuning up. Perhaps simultaneous bursts of sugar AND insulin to replenish cells?
 
AaronMHatch said:
I have yet to see any scientific evidence suggesting that starting an alcohol habit is healthier than ending one.
Click to expand...
But to end one you have to have started one first. Adding insult to injury can't possibly make something healthier, can it?! :)
 
My two cent , probably the same effect as the french paradox. Drinking moderately a bit of alcohol, bring into your body other substance suspected to be positive to your health (tanin for example) I dunno if it was falsified or not. So it isn't surprising to hear healthy moderate drinker live longer in average than healthy non drinker. But it is quite surprising to hear that heavy drinker have a lower mortality than non drinker.

I think I will have to take over drinking 1 glass of alcohol per day or two :).
 
AaronMHatch said:
I unfortunately do not have access to the paper. What I know is that there is much room for error in these alcohol studies. They simply cannot pin down causation; and the significant amount of uncertainty in the results suggests it's more of a correlation.
Click to expand...
So, how do you respond to the repeated positive correlations regarding mortality and moderate alcohol consumption? It's not like the studies are all over the place. The results are consistent.

What room for errors do you know about? Are researchers unaware of them?

Alcohol's effects on the body and much of its effects on the brain are well documented. I have yet to see any scientific evidence suggesting that starting an alcohol habit is healthier than ending one.
Click to expand...
It's interesting you should call it a habit. If we were talking about vitamin C, would you call it a habit?

Define healthier. The studies have looked at mortality. I would say that pretty much every drug on the market has some negative side effect, some of which are serious. Doctors often recommend courses of drugs that prolong life at the expense of overall health. It seems to me this discussion so far sounds very different than one about a drug with an unknown mechanism of action.
 
*Did not join to make this post*

I don't have any statistics to hand, so there are several reasonable assumptions in what I'm about to say. If anyone has the statistics regarding how much alcohol the average person consumes, the general age range for alcohol consumption and how many people within that test were affected with health issues prohibiting alcohol consumption; I'd be grateful. I'm aware the last one is a bit of a stretch.

I think it is reasonable to say that on a weekly basis the majority of people would fall into the "moderate drinker" category. Lets throw a figure at it and call it 80% of people in any given collection of 1,000 people collected off of the street would be "moderate drinkers" or would have been moderate drinkers in the past, given "abstaining from alcohol does actually tend to increase one's risk of dying even when you exclude former drinkers" those that were drinkers and now are not did not contribute to the "non drinkers" mortality statistic.

The other two groups, the "heavy drinkers" and the "non drinkers" are an even split of the last 20%. Further assumptions incoming, I think it is reasonable to say that those who drink heavily or not at all do so under such a huge variety of conditions and for so many different reasons that its illogical/a waste of time to attempt to correlate reasoning with either stance.

Getting back to the "moderate drinkers"... moderation in all things. Without seeing the actual statistics of that study I could just as easily say "Heavy Drinkers Suicide Ratio 1:3" and you'd not be able to get back to much on it.

My point is that within reason you can't just take a group of people, segregate them based on one thing and then attribute the resulting divisions with amazing properties. For instance, people eat oranges, I take 1,000 people off of the street and ask them how many oranges do they eat a day. Dividing them up into three categories, "Monkey", "Fruit Bat" and "Carnivore" where "Monkey" is 5+ and "Carnivore" is 0. I find that almost miraculously those in the "Monkey" category can jump higher than those in the "Carnivore" category and then attribute this to their consumption of oranges. Thus, eat oranges and jump higher.

I would probably fall into the "moderate drinker" category myself and I could say that of all the moderate and heavy drinkers I know/knew, not alot of them are dead. I could say the same about the people who abstain.
 
Last edited:
Koyaanisqatsi said:
I would probably fall into the "moderate drinker" category myself and I could say that of all the moderate and heavy drinkers I know, not alot of them are dead. I could say the same about the people who abstain.
Click to expand...
I wonder what you'd say if you fell into the "heavy drinker" category?! I'm a light (not "lite"!) drinker and I know even fewer people than you who are dead. In fact none. I knew some people who are now dead. Perhaps that's the correlation - the more you drink the more dead people you know!
 
Well it could be said that if volume of consumption over time is the deciding factor of "heavy drinker", then obviously the more you drink the more people you've known who are now dead. Because you've been around longer.

Grammatical error noted.
 
Last edited:
"I have always taken more from alcohol than alcohol has taken from me."-
Sir Winston Churchill.
 
UncaYimmy said:
The abstract reads, "A model controlling for former problem drinking status, existing health problems, and key sociodemographic and social-behavioral factors, as well as for age and gender, substantially reduced the mortality effect for abstainers compared to moderate drinkers. However, even after adjusting for all covariates, abstainers and heavy drinkers continued to show increased mortality risks of 51 and 45%, respectively, compared to moderate drinkers."
Click to expand...
It probably doesn't take many drunk drivers killing a van-load of abstainers while walking away from the accident themselves to skew the results in favor of heavy drinking. Controlling for "social-behavioral factors" would seem to imply a preference for abstainers who are on the roads when the bars close.

I'm not an abstainer, but I might as well be. In the past year, I've had one actual hard drink because I was visiting a friend, and a couple of sips of wine when I was preparing a recipe that called for it. Every few years, when 60 Minutes runs their "red wine extends lifespans" segment, I'll buy a couple of bottles, drink one over the course of a week, and let the others age on the shelf until the next time the segment airs. I know, one data point, but I won't die until I'm 106...
 
I notice that it is just a study of older adults - between the ages of 55 and 65. Perhaps more heavy drinkers die before the age of 55 than do non-drinkers so I don't think we can generalise this study to the drinking habits of all adults.
 
This study is a good example of the caution needed to interpret observational studies. Without controlling for factors related to alcohol consumption, the mortality rate for abstainers was higher than the rate for moderate and heavy drinkers. Once you controlled for some readily identified factors, the mortality rate in heavy drinkers and abstainers was pretty much the same. It's not hard to see that residual confounding factors which were less obvious or which weren't so easily measured (and therefore remained uncontrolled) could account for the remaining residual differences. After all, we have discovered on numerous occasions, when moving from observational studies to interventional studies, that even reasonably large effects were entirely due to uncontrolled and unrecognized confounders.

Linda
 
Last edited:
AaronMHatch said:
http://news.yahoo.com/s/time/08599201433200;_ylt=AvD8AXFcmqI31pbmrgLLks5v24cA;_ylu=X3oDMTM1czYwYWVjBGFzc2V0A3RpbWUvMjAxMDA4MzAvMDg1OTkyMDE0MzMyMDAEY2NvZGUDbW9zdHBvcHVsYXIEY3BvcwM0BHBvcwM0BHNlYwN5bl90b3Bfc3RvcmllcwRzbGsDaGVhdnlkcmlua2Vy

From the article: "The most shocking part? Abstainers' mortality rates are higher than those of heavy drinkers."

Abstainers die more often than heavy drinkers? That is shocking!
Click to expand...
I drink five brews a week. I enjoy everyone of them.
 
Professor Yaffle said:
I notice that it is just a study of older adults - between the ages of 55 and 65. Perhaps more heavy drinkers die before the age of 55 than do non-drinkers so I don't think we can generalise this study to the drinking habits of all adults.
Click to expand...

That's a good point. Is there anything in the study to show that it's not just evidence that those who drink heavily and survive to 55 have bodies which can repair themselves better than others?

In other words, once you eliminate all the heavy drinkers who died before 55, those who are left may be people who can survive insults to their bodies slightly better than others, whether from alcohol, disease, injury, etc.
 
Marduk said:
I would think that the 24% of suicide deaths associated with alcohol more than balance those figures
http://www.medscape.com/viewarticle/704837
;)
Click to expand...

A link that you don't need a login for would be nice... not that I doubt you since I helped publish some of the relevant figures myself.

And many of these alcohol related suicides would be before the age of 55.
 
Last edited:
ElMondoHummus said:
Now only if binge drinking had any positive effects on my longevity...

;)

:boxedin: Was a joke... please don't throw things...
Click to expand...

I expect if you donate your liver to medical science it will last quite a while in a jar full of formaldehyde as an example of what binge drinking does
:D
 
UncaYimmy said:
So, how do you respond to the repeated positive correlations regarding mortality and moderate alcohol consumption? It's not like the studies are all over the place. The results are consistent.
Click to expand...

The disregard for sufficient controls in these alcohol experiments might explain the "consistent" results.

UncaYimmy said:
What room for errors do you know about? Are researchers unaware of them?
Click to expand...

The errors I'm talking about are the factors the researchers do not control for. For example, many alcohol studies of this kind state that light drinkers live longer than people who don't drink. Who are the kind of people who drink in moderation? Alcohol is an addictive substance that tests our constitution. People who can drink in moderation (and let's get real here - many of these studies I'm referring to have said one drink can be beneficial whereas 2 removes all beneficial results and increases our health risk) have to be in very good control of themselves. Having one drink suggests a stereotypical middle class / upper class person (I'm referring to the USA) who has a good head on their shoulders and knows how to take care of his or herself. Perhaps the only reason moderate drinkers live longer is because they're wise in all areas of their life.

Researchers are usually well aware of their inability to control for all of these factors all of the time. It's not uncommon to see it mentioned in the discussion section.

UncaYimmy said:
It's interesting you should call it a habit. If we were talking about vitamin C, would you call it a habit?
Click to expand...

Moderate drinking is considered by the US government to be no more than two drinks a day. I define a habit as being the regular occurrence of an action. If someone has a drink 5 days or more a week, that's regular occurrence, and thus a habit. I think a moderate drinking habit is just as much a habit as taking a vitamin each day.

UncaYimmy said:
Define healthier. The studies have looked at mortality. I would say that pretty much every drug on the market has some negative side effect, some of which are serious. Doctors often recommend courses of drugs that prolong life at the expense of overall health. It seems to me this discussion so far sounds very different than one about a drug with an unknown mechanism of action.
Click to expand...

I think you made my point for me. These studies give off the vibe that drinking can be healthier than not drinking by using mortality numbers as their proof. They haven't proved that drinking is healthier; and they've only showed a slight correlation for moderate drinking and living longer. Until they control for a sufficient number of variables, big deal.

Doctors also do not advise that we begin a moderate drinking habit if we currently do not drink. I think that's enough information for me.
 
fls said:
This study is a good example of the caution needed to interpret observational studies. Without controlling for factors related to alcohol consumption, the mortality rate for abstainers was higher than the rate for moderate and heavy drinkers. Once you controlled for some readily identified factors, the mortality rate in heavy drinkers and abstainers was pretty much the same. It's not hard to see that residual confounding factors which were less obvious or which weren't so easily measured (and therefore remained uncontrolled) could account for the remaining residual differences. After all, we have discovered on numerous occasions, when moving from observational studies to interventional studies, that even reasonably large effects were entirely due to uncontrolled and unrecognized confounders.

Linda
Click to expand...
Yes, I also don't think the sample size is large enough to draw any conclusions from a 9% difference. 1824 participants total, 69% of the non-drinking group was dead after 20 years, 60% of the heavy drinkers, moderate drinkers clearly the winners with only 41% dead. Still suprising, I would have expected the heavy drinkers to have a mortality high enough to swamp any other factors.
 
One big hole in the study is that heavy drinkers are defined as anyone who drinks more than 3 drinks a day. So someone who drinks 4 drinks a day is lumped in the same category as a 30 drink a day alcoholic, and we don't know the proportions of each, or the average drinks per day of those in the heavy drinker category.
 
Koyaanisqatsi said:
My point is that within reason you can't just take a group of people, segregate them based on one thing and then attribute the resulting divisions with amazing properties.
Click to expand...

Well, no. That's why you try to control for confounding variables, like these researchers did. Have you read the linked article?

AaronMHatch said:
The disregard for sufficient controls in these alcohol experiments might explain the "consistent" results.

The errors I'm talking about are the factors the researchers do not control for. For example, many alcohol studies of this kind state that light drinkers live longer than people who don't drink. Who are the kind of people who drink in moderation? Alcohol is an addictive substance that tests our constitution. People who can drink in moderation (and let's get real here - many of these studies I'm referring to have said one drink can be beneficial whereas 2 removes all beneficial results and increases our health risk) have to be in very good control of themselves. Having one drink suggests a stereotypical middle class / upper class person (I'm referring to the USA) who has a good head on their shoulders and knows how to take care of his or herself.
Click to expand...

Yeah, if only the researchers had controlled for socioeconomic status. Oh wait, they did -- as was made clear in the linked article.

I think you made my point for me. These studies give off the vibe that drinking can be healthier than not drinking by using mortality numbers as their proof. They haven't proved that drinking is healthier; and they've only showed a slight correlation for moderate drinking and living longer. Until they control for a sufficient number of variables, big deal.
Click to expand...

What would constitute a "sufficient number of variables" for you? And since you obviously didn't even bother to read the article about this study, how will you ever find out?

Doctors also do not advise that we begin a moderate drinking habit if we currently do not drink. I think that's enough information for me.
Click to expand...

Where do you think "doctors" get the information on which they base their advice?
 
Dunstan said:
Yeah, if only the researchers had controlled for socioeconomic status. Oh wait, they did -- as was made clear in the linked article.
Click to expand...
Since I'm a poor student and didn't pay for the article like you, can you tell me whether character was controlled for? Ah, you're referring to the Yahoo write-up. Not good enough.
Dunstan said:
What would constitute a "sufficient number of variables" for you?
Click to expand...
Character would be interesting.
Dunstan said:
Where do you think "doctors" get the information on which they base their advice?
Click to expand...
From research showing that starting a moderate drinking habit is not healthier than abstaining entirely.
 
Last edited:
AaronMHatch said:
Since I'm a poor student and didn't pay for the article like you, can you tell me whether character was controlled for? Ah, you're referring to the Yahoo write-up. Not good enough.

Character would be interesting.
Click to expand...

Since "character" is so vague as to be pretty much impossible to control for -- unless you have a suggestion to offer? -- you're essentially saying that no study will ever be good enough for you. Your belief that abstinence is healthier than moderate drinking is unfalsifiable. To which I say, ok, suit yourself, but it makes further discussion with you on this topic rather pointless.

From research showing that starting a moderate drinking habit is not healthier than abstaining entirely.
Click to expand...

See above.
 
fls said:
This study is a good example of the caution needed to interpret observational studies. Without controlling for factors related to alcohol consumption, the mortality rate for abstainers was higher than the rate for moderate and heavy drinkers. Once you controlled for some readily identified factors, the mortality rate in heavy drinkers and abstainers was pretty much the same. It's not hard to see that residual confounding factors which were less obvious or which weren't so easily measured (and therefore remained uncontrolled) could account for the remaining residual differences. After all, we have discovered on numerous occasions, when moving from observational studies to interventional studies, that even reasonably large effects were entirely due to uncontrolled and unrecognized confounders.
Click to expand...

If you had to place a bet with even odds on the results of a randomized intervention study, would you put money on alcohol increasing or decreasing lifespan? How much would that decision be influenced by this study?
 
You must log in or register to reply here.

Back
Top Bottom