The URL for the current French article is:
    http://www.ama-assn.org/sci-pubs/journals/most/recent/issues/inte/ioi81189.htm
While I'm at it, another article comparing the relation of different beverage-types with mortality, this time with US aggregate level data, has just also just appeared on the web.  It's by Ron Roizen et al.; the abstract can be found at:
    http://www.bmj.org/cgi/content/short/319/7211/666
with the full text available by a click from there.  Robin  
    -----Original Message-----
    From: Robin Room <[log in to unmask]>
    To: [log in to unmask] <[log in to unmask]>; [log in to unmask] <[log in to unmask]>
    Date: den 13 september 1999 17:09
    Subject: wine vs. beer as a preservative
    
    
    (Susanna Barrows told me the story once, complete with photos, of a French study prior to the First World War which compared the marching and fighting abilities of French poilus (infantry soldiers) when their rations included wine vs. when their rations included beer.  Wine won, with a patriotic flourish in the study's reporting, as I remember.  The authors of that study would appreciate the one below.
       "Wine drinkers were less likely to die violent deaths" is the giveaway line that there must be uncontrolled social differences between the wine and beer drinkers in the study, which could also be affecting the other findings.  Robin)
    
    
    Yahoo!/Reuters - Monday September 13, 1999
    
    Wine Drinking Beats Beer In French Health Study
    
    CHICAGO (Reuters) - Drinking a few glasses of wine a day proves to be more of a
    life-saver than beer drinking, perhaps because of cancer-fighting compounds in
    wine, a study of more than 36,000 middle-aged Frenchmen said Sunday.
    
    As previous studies have shown, drinking one to three glasses of wine daily can
    lower blood pressure and cholesterol levels and lessen the chances of suffering
    a heart attack.
    
    The French study showed moderate wine drinkers had a death rate from heart
    attacks that was between 45-48 percent lower than those who abstained from
    drinking. Red wine particularly contains antioxidants that may help prevent
    heart disease.
    
    Beer drinkers were found to have a 42 percent lower rate of fatal heart attacks
    than non-drinkers, but not as low as the wine drinkers.
    
    But the wine drinkers in the study also had a significantly lower risk of dying
    from all causes than either beer drinkers or nondrinkers, when adjusted for
    factors such as age and smoking.
    
    Wine drinkers were less likely to die violent deaths, or suffer fatal bouts with
    cancer or other diseases.
    
    While drinking alcohol from any source increases the risk of cancer, the
    researchers said, substances found in wine such as resvaratrol may more than
    compensate by inhibiting development of the disease.
    
    The study found beer drinkers did not enjoy the same protection from cancer as
    wine drinkers. Drinking beer might even have posed an increased risk of cancer
    due to the presence of substances called nitrosamines found in beers brewed
    before 1990.
    
    Most types of beer have since been made with much lower levels of nitrosamines
    but the impact on cancer rates will not be known until the next decade, the
    researchers said.
    
    They noted that excessive drinking has been found to double the risk of cancer.
    Drinkers were also more likely to smoke than nondrinkers.
    
    All 36,250 men in the study ranged in age from 40 to 60 when data collection
    began between 1978 and 1983, and the researchers later examined death
    certificates.
    
    They noted statistical comparisons could have been skewed by the control group
    of nondrinkers, some of whom might have stopped drinking due to health problems
    that could have ultimately hastened their deaths.
    
    As with other health studies involving alcohol, the researchers were careful to
    issue a cautionary note.
    
    ``Because of the potential health hazards of any alcoholic beverage, our results
    suggesting that a few glasses of wine may have salutary effects should not be
    interpreted as an incentive to drink,'' wrote study authors Serge Renaud and
    Roger Salamon of the National Institute for Health and Medical Research in
    Bordeaux, and Gerard Siest and Rene Gueguen of the Center of Preventive Medicine
    in Vandoeuvre-les-Nancy.
    
    ``If you drink alcoholic beverages, do so in moderation, with meals and when
    consumption does not put you or others at risk,'' they wrote.
    
    The study appeared in the Archives of Internal Medicine, a journal published by
    the American Medical Association.