SEARCH

SEARCH BY CITATION

Keywords:

  • Alcohol;
  • causallity;
  • confounders;
  • diseases;
  • prevention

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

Background

In observational studies, moderate drinking is associated with a reduced risk of more than twenty different diseases and health problems. However, it would be premature to conclude that there is a causal relationship.

Method

This paper critically reviews the evidence for such associations.

Findings

It was found that reasons for questioning the causal association of moderate drinking and a reduced health risk are: the lack of dose-response relationships; the characteristics and lifestyles of today's abstainers and moderate drinkers; the lack of plausible biological mechanisms; the problems in the classification of drinking groups, and; the general limitations of observational studies.

Conclusions

The evidence for the harmful effects of alcohol is undoubtedly stronger than the evidence for beneficial effects.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

After adjusting for many confounders, observational studies have demonstrated a significant correlation between light or moderate alcohol use and reduced risk of several diseases and health problems, as shown in the following list:

  • Alzheimer's disease/dementia [1-3]
  • Asthma [4]
  • Autoimmune hypothyroidism [5]
  • Colorectal cancer [6]
  • Common cold [7]
  • Coronary heart disease [8, 9]
  • Diabetes (type 2) [10]
  • Gallstones [11]
  • Hearing loss [12, 13]
  • Intermittent claudication [14]
  • Liver cirrhosis [15]
  • Low birth weight, prematurity [16]
  • Lower urinary tract symptoms (in men) [17]
  • Metabolic syndrome [18]
  • Negative child development [19-21]
  • Osteoporosis [22-25]
  • Overweight [26-28]
  • Psychiatric disorders [29-32]
  • Renal cell cancer [33]
  • Rheumatoid arthritis [34-36]
  • Stroke (ischaemic) [37]
  • General health status [38, 39]
  • All cancer mortality [40]
  • Total mortality [41]

For the majority of these diseases and health problems, the correlation has been confirmed in more than one study. For seven of these, several studies and meta-analyses have been published.

In addition, Danish researchers have found that moderate wine consumption is associated with lower frequency of cancer in the upper digestive tract [42]; gastric cancer [43]; hip fracture [44]; and lung cancer [45].

A conspicuous fact is that light or moderate drinking apparently prevent such a large number of unrelated diseases. It is even more striking that beneficial effects for the different diseases all seem to peak at approximately the same consumption level. For almost all the diseases there is a remarkable absence of dose–response effects; protective and harmful factors more often have a dose–response effect within threshold limits. This is easier to explain biologically, and is regarded frequently as one of the criteria for inferring causality. However, in practically all these studies, certain small doses of alcohol appear to be protective and somewhat larger doses are apparently harmful.

Characteristics of Drinking groups

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

The diseases which moderate drinking can apparently prevent are largely diseases that we know are influenced by several life-style factors. Alcohol apparently prevents 12 diseases among the 16 listed by Wikipedia as ‘lifestyle diseases’.

The level of consumption which seems to have a preventive effect is the level that is most accepted in our present-day societies, more socially accepted than both abstention and higher consumption. In order to feel as a ‘normal’ citizen, as a ‘well-integrated member of society’, one should drink alcohol, but in moderation.

Thus, the group consuming alcohol at this level is conforming to the currently prevailing norms. Not surprisingly, several favourable characteristics have been demonstrated for this group. It differs from abstainers in many life-style factors which are relevant for health.

The confounders taken into account in the observational studies vary widely. Most common are age, sex, body mass index (BMI), smoking, pre-existing diseases and blood pressure. Today, abstainers differ from light, moderate and heavy drinkers on a number of demographic, physical and psychosocial characteristics. A hundred years ago, especially in the Nordic and English-speaking countries, a large number of people were ideological teetotallers, in order to contribute to the fight against inebriety. Today's abstainers are, however, not an average group of people who happen to have a temperance conviction, but are more often a somewhat deviant and marginalized group. This is reflected in their life-styles in several ways, as abstainers generally have less favourable life-styles than light or moderate drinkers.

An analysis [46] of the drinking groups concluded that ‘moderate alcohol consumption is a powerful general indicator of optimal social status’, and that ‘Our multivariate analysis showed that several risk factors which are rarely or never taken into account in previous observational studies are associated with alcohol intake independently of traditional risk factors.’ Even respiratory function, which has never been included as a confounder, was most favourable in moderate drinkers.

Abstainers as a group have a less healthy diet and exercise less than moderate drinkers [47-51]. These key confounders are rarely mentioned in the observational studies.

Even when all known confounders are taken into account, psychosocial factors seem to have a considerable impact on morbidity and mortality, probably through mechanisms which are still unknown. One study concluded that abstainers have more of several psychosocial risk factors such as low education, passive life-style, being unmarried, disabled or depressed [52]. This was confirmed strongly in a larger study. Among 30 possible psychosocial risk factors, abstainers had a higher score on 27 items [53].

Both these studies showed that abstainers have fewer and poorer social relationships. A meta-analysis of 148 studies of social relationships concluded: ‘Data across 308 849 individuals, followed for an average of 7.5 years, indicate that individuals with adequate social relationships have a 50% greater likelihood of survival compared to those with poor or insufficient social relationships. The magnitude of this effect is comparable with quitting smoking and it exceeds many well-known risk factors for mortality (e.g. obesity, physical inactivity)’ [54].

Non-drinkers also have lower socio-economic status and education [55]. These factors seem to have an independent impact on morbidity and mortality, especially coronary heart disease [56-59].

Danish researchers have demonstrated clearly that different drinking habits may reflect different status and life-styles. While showing that moderate wine drinking is related to lower morbidity of various diseases, they have also shown that Danish wine drinkers live healthily in several ways [60-62]. Analysing the characteristics of drinking groups, the conclusion was: ‘wine drinking is a general indicator of optimal social, cognitive, and personality development in Denmark. Similar social, cognitive, and personality factors have also been associated with better health in many populations. Consequently, the association between drinking habits and social and psychological characteristics, in large part, may explain the apparent health benefits of wine’.

An American study of a national sample of young adults had a similar conclusion: ‘Wine drinkers generally had more formal education, better dietary and exercise habits, and more favorable health status indicators’ [63].

Altogether, there is ample evidence that groups with different drinking habits differ in several other ways than their drinking, making it difficult to separate the effects of drinking habits from other factors. Only a few studies have attempted to study more homogeneous groups, mainly by including only health professionals.

Almost all the studies have been performed in English-speaking countries or western Europe. It would be interesting to study populations where light or moderate drinkers are not a group with higher status, healthier life-style and better social integration. India may be such a country, as religious Hindus are not expected to drink alcohol; rather, a study in India concluded that light drinking increased the risk of coronary disease [64]. Conversely, a study in China showed a correlation between light drinking and less risk [65]. Whether light drinking is most frequent among those better off in China, as in western countries, is unknown. This topic should be studied further.

Limitations of Observational Studies

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

Due to research performed during the last 15 years, the general belief in non-randomized studies has been reduced. Most well known is the issue of hormone replacement therapy (HRT) for preventing coronary heart disease. In 1991, a review of the epidemiological studies [66] concluded that: ‘the bulk of evidence strongly supports a protective effect of estrogens that is unlikely to be explained by confounding factors…’. Observational studies had tried to take all relevant confounders into consideration. Their best estimate of the relative risk was calculated using meta-analytical techniques applied to the epidemiological studies they deemed to be of high quality based on their designs—prospective studies, even with internal controls and angiographic studies. The estimate of the relative risk for HRT users was 0.50. The confidence interval was narrow: 0.43–0.56. This was a clearer indication of a causal relationship than any of the studies apparently showing that alcohol reduces the risk of disease or death.

The biological effects of the hormones added to the certainty that HRT prevented heart disease. A vast literature documented a favourable effect of HRT on coronary risk factors, such as the cholesterols (as for alcohol today). The US Food and Drug Administration (FDA) accepted the drug industry's request to approve a label change that permitted prevention of heart disease to be included as an indication for HRT. Half the female American doctors past 50 years of age reportedly began to take HRT to prevent coronary heart disease. However, subsequent randomized trials demonstrated consistently that HRT did not prevent heart disease at all [67-69]. The reason for women taking HRT to perform better in observational studies was obviously some feature of these women which was not covered by all the confounders taken into account in the observational studies.

Another example is vitamin E. Observational studies showed an association between vitamin E consumption and reduced risk of coronary heart disease [70, 71]. Vitamin E's antioxidant properties made it seem plausible that vitamin E had a preventive effect; and several cardiologists began to consume vitamin E [72]. Other studies showed that consumption of beta-carotene was also associated with a lower risk of heart disease [73]. Nevertheless, subsequent randomized studies showed that neither vitamin E nor beta-carotene actually reduced the risk [74, 75].

Our experience with HRT, vitamin E and beta-carotene teaches us three important lessons: (1) adjusting for known confounders is not sufficient to prove causality; (2) do not be blinded by biological mechanisms that might explain a possible causal relationship; and (3) observational studies are valuable for generating hypotheses, but insufficient as final proof.

The number of observational studies is not an indicator of the strength of the evidence for causal effect. Observational studies remain second-class evidence. This is also confirmed by the many studies giving unreasonable results. One example is that television viewing is associated with higher cardiovascular and total mortality, independent of physical activity, gender, age, education, smoking, alcohol, medication, diabetes history, family history of cardiovascular disease and cancer and body mass index [76]. Why is TV viewing more harmful than, for instance, book reading? Because TV radiation is harmful only at an extremely close range, it must be due to some unknown health-related characteristics within the group watching more TV.

Classification Problems

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

In addition to the huge problem of identifying and covering all relevant confounders, there are classification problems when subjects are referred to drinking groups. A well-known problem concerns the ‘sick quitters’, as people tend to reduce their drinking when health problems occur. Ex-drinkers have considerable excess cardiovascular and total mortality [77]. In a meta-analysis of 54 studies, the excess total or coronary mortality of non-drinkers was found only in studies where ex-drinkers or occasional drinkers were included in the non-drinker group [78].

Another meta-analysis, covering 35 studies of mortality from coronary heart disease and 54 studies of total mortality, also concluded that the apparent protective effect of alcohol disappeared when ex-drinkers and occasional drinkers were excluded from the non-drinker group [79]. A third meta-analysis of 35 prospective studies reached the same conclusion [80].

For Comparison: the Evidence for Harmful Effects

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

We must, of course, require the same degree of evidence for concluding that alcohol increases the risk of health problems.

Decades ago, it was believed commonly that cirrhosis of the liver was not caused by alcohol but by insufficient nutrition and/or by other substances in the beverages. Charles S. Lieber demonstrated, with studies on baboons, that alcohol causes cirrhosis even with optimal nutrition [81]; in addition, there is a clear dose–response relationship, so for cirrhosis there seems to be sufficient evidence for a firm conclusion. Surprisingly, no animal studies using coronary heart disease as an end-point have been published, in spite of the almost 5000 publications on alcohol and heart disease since the Lancet launched the prevention theory in 1979. Animal studies were critical for the recognition of alcohol as a teratogen [82, 83].

Countless experimental studies have demonstrated impaired coordination, increased reaction time and reduced skills when under the influence. The dose–response relationship is very clear. This evidence seems sufficient for stating that alcohol increases the risk of accidents and injuries, although the well-known tendency to show more risky behaviour after drinking may also contribute.

The issue of alcohol and cancer is more complicated. The most authoritative review is probably the large report from the World Cancer Research Fund/American Institute for Cancer Research on prevention and risk factors for cancer [84]. The chapter on alcohol concludes: ‘The evidence that alcoholic drinks are a cause of cancers of the mouth, pharynx, and larynx, oesophagus, colorectum (men), and breast is convincing. They are probably a cause of colorectal cancer in women, and of liver cancer.’ Assessing the evidence, the conclusion for most of these types of cancer is: ‘There is ample and consistent evidence, both from cohort and case–control studies, with a dose–response relationship. There is robust evidence for mechanisms operating in humans.’

Differing from the claim that moderate alcohol use prevents diseases, in addition to observational studies and plausible mechanisms, is that a large number of studies have demonstrated a dose–response relationship for many types of cancer. This corresponds to the general pattern for exposure to risk factors and strengthens the case for a causal relationship.

For cancer of the breast, the report states that ‘most experimental studies in animals have shown that alcohol intake is associated with increased breast cancer risk’. Thus, the evidence seems clear for breast cancer. In the absence of animal studies on other types of cancer, the evidence for a causal relationship for these other types is more weak. However, the dose–response relationships and the known biological mechanisms strengthen the evidence.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References

There is solid evidence that light or moderate drinkers have a reduced risk of several diseases which are influenced by life-style factors. Whether or not the lower risk is due to alcohol is a more complicated issue.

Taken together, the existing evidence does not seem to meet the criteria for inferring causality, such as the classical Bradford Hill guidelines for causation in epidemiology [85]. For almost all the diseases, we do not know of any plausible biological mechanism explaining a causal preventive role for alcohol. In theory there might be a common, as-yet undiscovered mechanism, but the diverse nature of the diseases makes it unlikely. Another criterion for causality which is almost completely lacking is the biological gradient, or dose–response relationship.

In order to draw reliable conclusions from the observational studies all relevant confounders should to be taken into account, so that the groups differ only with respect to consumption level and the confounders accounted for. Some recent studies attempt to overcome the problem by including more confounders than previously [86]. Like others before them, the authors seem to be confident that all relevant confounders have been taken into account.

The problem remains that we do not fully know what is implied by ‘all relevant confounders’. This means that, in practice, we are unable to find a perfect reference group. Today, the FDA requires randomized studies to approve new therapeutic or preventive drugs. It may be argued that such studies should also be required to recommend alcohol as a protective agent. Thus, alcohol's possible ability to prevent diseases should probably not be considered as an established fact. The issue remains: is a causal relationship more probable for some diseases than for others?

By far the largest number of studies has concerned the prevention of coronary heart disease. In addition to observational studies, several controlled studies have demonstrated alcohol's beneficial effects on biological markers such as high-density lipoprotein (HDL) and fibrinogen [87]. These studies have used 15–90 g alcohol a day, on average c. 40 g. However, the maximum protective associations in meta-analyses of cohort studies with coronary heart disease as the outcomes are based usually on much smaller doses. In numerous studies, one or two units a week (2–5 g alcohol a day) is sufficient to achieve the apparently preventive effect. For these doses, significant positive effects have not been shown in the experimental studies.

The absence of dose–response effects always weakens the case for a causal relationship. For coronary heart disease, a recent meta-analysis did find a dose–response effect [9]. This may be due to a causal effect, but there is, however, another reasonable interpretation: this review of research on the characteristics of drinking groups has indicated strongly that light or moderate drinkers have healthier life-styles and fewer risk factors, in ways not accounted for in observational studies. When the meta-analysis indicates the lowest risk for coronary mortality at 31 g/day for men and 11 g/day for women, it may also reflect that the effect of the other beneficial factors most probably peak around these consumption levels.

The evidence for alcohol's preventive effect on coronary heart disease is similar to the evidence we had for a preventive effect of hormone replacement therapy 15 years ago. The number of studies is larger for alcohol, but as long as the type of evidence is largely the same, regarding the number of studies as decisive is questionable. Altogether, the evidence for alcohol's ability to prevent diseases is considerably weaker than that for alcohol causing several kinds of harm.

Alternatively, the large number of unrelated diseases which are apparently prevented may also be interpreted as evidence of confounders not accounted for. In Clayton & Hills’ widely used textbook, Statistical Models in Epidemiology, the authors recommend omitting the zero exposure group from studies on the effect of various doses [88]. Their argument is that: ‘the level of exposure which is coded zero, is often qualitatively different from the other levels’. This presumption may seem to fit well with the ‘zero exposure to alcohol’ group.

Unfortunately, we cannot expect a controlled study based on randomized groups committed to a desired drinking practice over years. Thus, the decisive evidence would probably require an experimental study with animals, ideally primates, and would require a considerable number of animals and a long observation time.

Whether alcohol prevents several diseases and health problems at a level of consumption where health benefits outweigh harm has obvious consequences for public health education. The absence of definite knowledge leaves plenty of room for wishful thinking, which we observe frequently on this topic.

Declaration of interests

None.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Characteristics of Drinking groups
  5. Limitations of Observational Studies
  6. Classification Problems
  7. For Comparison: the Evidence for Harmful Effects
  8. Conclusions
  9. References
  • 1
    Anstey K. J., Mack H. A., Cherbuin N. Alcohol consumption as a risk factor for dementia and cognitive decline: meta-analysis of prospective studies. Am J Geriatr Psychiatry 2009; 17: 542555.
  • 2
    Neafsey E. J., Collins M. A. Moderate alcohol consumption and cognitive risk. Neuropsychiatr Dis Treat 2011; 7: 465484.
  • 3
    Kim J. W., Lee D. Y., Lee B. C., Jung M. H., Kim H., Choi Y. S. et al. Alcohol and cognition in the elderly: a review. Psychiatry Invest 2012; 9: 816.
  • 4
    Lieberoth S., Backer V., Kyvik K. O., Skadhauge L. R., Tolstrup J. S., Grønbæk M. et al. Intake of alcohol and risk of adult-onset asthma. Respir Med 2012; 106: 184188.
  • 5
    Carle A., Pedersen I. B., Knudsen N., Perrild H., Ovesen L., Rasmussen L. et al. Moderate alcohol consumption may protect against overt autoimmune hypothyroidism—a population-based case–control study. Eur J Endocrinol 2012; 167: 483490.
  • 6
    Kontou N., Psaltopoulou T., Soupos N., Polychronopoulos E., Xinopoulos D., Linos A. et al. Alcohol consumption and colorectal cancer in a Mediterranean population: a case–control study. Dis Colon Rectum 2012; 55: 703710.
  • 7
    Cohen S., Tyrrell D. A., Russell M. A., Jarvis M. J., Smith A. P. Smoking, alcohol consumption, and susceptibility to the common cold. Am J Public Health 1993; 83: 12771283.
  • 8
    Ronksley P. E., Brien S. E., Turner B. J., Mukamal K. J., Ghali W. A. Association of alcohol consumption with selected cardiovascular disease outcomes: a systematic review and meta-analysis. BMJ 2011; 342: d671.
  • 9
    Roerecke M., Rehm J. The cardioprotective association of average alcohol consumption and ischaemic heart disease: a systematic review and meta-analysis. Addiction 2012; 107: 12461260.
  • 10
    Baliunas D. O., Taylor B. J., Irving H., Roerecke M., Patra J., Mohapatra S. et al. Alcohol as a risk factor for type 2 diabetes: a systematic review and meta-analysis. Diabetes Care 2009; 32: 21232132.
  • 11
    Walcher T., Haenle M. M., Mason R. A., Koenig W., Imhof A., Kratzer W. The effect of alcohol, tobacco and caffeine consumption and vegetarian diet on gallstone prevalence. Eur J Gastroenterol Hepatol 2010; 22: 13451351.
  • 12
    Popelka M. M., Cruickshanks K. J., Wiley T. L., Tweed T. S., Klein B. E., Klein R. et al. Moderate alcohol consumption and hearing loss: a protective effect. J Am Geriatr Soc 2000; 48: 12731278.
  • 13
    Fransen E., Topsakal V., Hendrickx J. J., Van Laer L., Huyghe J. R., Van Eyken E. et al. Occupational noise, smoking, and a high body mass index are risk factors for age-related hearing impairment and moderate alcohol consumption is protective: a European population-based multicenter study. J Assoc Res Otolaryngol 2008; 9: 264276.
  • 14
    Garcia-Diaz A. M., Marchena P. J., Toril J., Arnedo G., Muñoz-Torrero J. F., Yeste M. et al. Alcohol consumption and outcome in stable outpatients with peripheral artery disease. J Vasc Surg 2011; 54: 10811087.
  • 15
    Rehm J., Taylor B., Mohapatra S., Irving H., Baliunas D., Patra J. et al. Alcohol as a risk factor for liver cirrhosis: a systematic review and meta-analysis. Drug Alcohol Rev 2010; 29: 437445.
  • 16
    Holman C. D., English D. R., Bower C., Kurinczuk J. J. NHMRC recommendations on abstinence from alcohol in pregnancy. Med J Aust 1996; 164: 699.
  • 17
    Rohrmann S., Crespo C. J., Weber J. R., Smit E., Giovannucci E., Platz E. A. Association of cigarette smoking, alcohol consumption and physical activity with lower urinary tract symptoms in older American men: findings from the third National Health and Nutrition Examination Survey. BJU Int 2005; 96: 7782.
  • 18
    Alkerwi A., Boutsen M., Vaillant M., Barre J., Lair M. L., Albert A. et al. Alcohol consumption and the prevalence of metabolic syndrome: a meta-analysis of observational studies. Atherosclerosis 2009; 204: 624635.
  • 19
    Robinson M., Oddy W. H., McLean N. J., Jacoby P., Pennell C. E., de Klerk N. H. et al. Low–moderate prenatal alcohol exposure and risk to child behavioural development: a prospective cohort study. Br J Obstet Gynaecol 2010; 117: 11391150.
  • 20
    Testa M., Quigley B. M., Eiden R. D. The effects of prenatal alcohol exposure on infant mental development: a meta-analytical review. Alcohol Alcohol 2003; 38: 295304.
  • 21
    Kelly Y., Sacker A., Gray R., Kelly J., Wolke D., Quigley M. A. Light drinking in pregnancy, a risk for behavioural problems and cognitive deficits at 3 years of age? Int J Epidemiol 2009; 38: 129140.
  • 22
    Holbrook T. L., Barrett-Connor E. A prospective study of alcohol consumption and bone mineral density. BMJ 1993; 306: 15061509.
  • 23
    Siris E. S., Miller P. D., Barrett-Connor E., Faulkner K. G., Wehren L. E., Abbott T. A. et al. Identification and fracture outcomes of undiagnosed low bone mineral density in postmenopausal women: results from the National Osteoporosis Risk Assessment. JAMA 2001; 286: 28152822.
  • 24
    Rapuri P. B., Gallagher J. C., Balhorn K. E., Ryschon K. L. Alcohol intake and bone metabolism in elderly women. Am J Clin Nutr 2000; 72: 12061213.
  • 25
    Sommer I., Erkkilä A. T., Järvinen R., Mursu J., Sirola J., Jurvelin J. S. et al. Alcohol consumption and bone mineral density in elderly women. Public Health Nutr 2012; 19. [Epub ahead of print].
  • 26
    Wang L., Lee I. M., Manson J. E., Buring J. E., Sesso H. D. Alcohol consumption, weight gain, and risk of becoming overweight in middle-aged and older women. Arch Intern Med 2010; 170: 453461.
  • 27
    Arif A. A., Rohrer J. E. Patterns of alcohol drinking and its association with obesity: data from the Third National Health and Nutrition Examination Survey, 1988–1994. BMC Public Health 2005; 5: 126.
  • 28
    Thomson C. A., Wertheim B. C., Hingle M., Wang L., Neuhouser M. L., Gong Z. et al. Alcohol consumption and body weight change in postmenopausal women: results from the Women's Health Initiative. Int J Obes (Lond) 2012; 36: 11581164.
  • 29
    Skogen J. C., Harvey S. B., Henderson M., Stordal E., Mykletun A. Anxiety and depression among abstainers and low-level alcohol consumers. The Nord-Trøndelag Health Study. Addiction 2009; 104: 15191529.
  • 30
    Caldwell T. M., Rodgers B., Jorm A. F., Christensen H., Jacomb P. A., Korten A. E. et al. Patterns of association between alcohol consumption and symptoms of depression and anxiety in young adults. Addiction 2002; 97: 583594.
  • 31
    Flensborg-Madsen T., Becker U., Grønbæk M., Knop J., Sher L., Mortensen E. L. Alcohol consumption and later risk of hospitalization with psychiatric disorders: prospective cohort study. Psychiatry Res 2011; 187: 214219.
  • 32
    Bots S., Tijhuis M., Giampaoli S., Kromhout D., Nissinen A. Lifestyle- and diet-related factors in late-life depression—a 5-year follow-up of elderly European men: the FINE study. Int J Geriatr Psychiatry 2008; 23: 478484.
  • 33
    Song D. Y., Song S., Song Y., Lee J. E. Alcohol intake and renal cell cancer risk: a meta-analysis. Br J Cancer 2012; 106: 18811890.
  • 34
    Maxwell J. R., Gowers I. R., Moore D. J., Wilson A. G. Alcohol consumption is inversely associated with risk and severity of rheumatoid arthritis. Rheumatology (Oxf) 2010; 49: 21402146.
  • 35
    Pedersen M., Jacobsen S., Klarlund M., Pedersen B. V., Wiik A., Wohlfahrt J. et al. Environmental risk factors differ between rheumatoid arthritis with and without auto-antibodies against cyclic citrullinated peptides. Arthritis Res Ther 2006; 8: R133.
  • 36
    Di Giuseppe D., Alfredsson L., Bottai M., Askling J., Wolk A. Long term alcohol intake and risk of rheumatoid arthritis in women: a population based cohort study. BMJ 2012; 345: e4230.
  • 37
    Patra J., Taylor B., Irving H., Roerecke M., Baliunas D., Mohapatra S. et al. Alcohol consumption and the risk of morbidity and mortality for different stroke types—a systematic review and meta-analysis. BMC Public Health 2010; 10: 258.
  • 38
    Sun Q., Townsend M. K., Okereke O. I., Rimm E. B., Hu F. B., Stampfer M. J. et al. Alcohol consumption at midlife and successful ageing in women: a prospective cohort analysis in the nurses’ health study. PLoS Med 2011; 8: e1001090.
  • 39
    Powers J. R., Young A. F. Longitudinal analysis of alcohol consumption and health of middle-aged women in Australia. Addiction 2008; 103: 424432.
  • 40
    Jin M., Cai S., Guo J., Zhu Y., Li M., Yu Y. et al. Alcohol drinking and all cancer mortality: a meta-analysis. Ann Oncol 2012 [Epub ahead of print].
  • 41
    Di Castelnuovo A., Costanzo S., Bagnardi V., Donati M. B., Iacoviello L., de Gaetano G. Alcohol dosing and total mortality in men and women: an updated meta-analysis of 34 prospective studies. Arch Intern Med 2006; 166: 24372445.
  • 42
    Grønbæk M., Becker U., Johansen D., Tønnesen H., Jensen G., Sørensen T. I. A. Population based cohort study of the association between alcohol intake and cancer of the upper digestive tract. BMJ 1998; 317: 844848.
  • 43
    Barstad B., Sørensen T. I., Tjønneland A., Johansen D., Becker U., Andersen I. B. et al. Intake of wine, beer and spirits and risk of gastric cancer. Eur J Cancer Prev 2005; 14: 239243.
  • 44
    Høidrup S., Grønbæk M., Gottschau A., Lauritsen J. B., Schroll M. Alcohol intake, beverage preference, and risk of hip fracture in men and women. Copenhagen Centre for Prospective Population Studies. Am J Epidemiol 1999; 149: 9931001.
  • 45
    Prescott E., Grønbæk M., Becker U., Sørensen T. I. A. Alcohol intake and the risk of lung cancer: influence of type of alcoholic beverage. Am J Epidemiol 1999; 149: 463470.
  • 46
    Hansel B., Thomas F., Pannier B., Bean K., Kontush A., Chapman M. J. et al. Relationship between alcohol intake, health and social status and cardiovascular risk factors in the Urban Paris-Ile-de-France cohort: is the cardioprotective action of alcohol a myth? Eur J Clin Nutr 2010; 64: 561568.
  • 47
    Ruidavets J. B., Bataille V., Dallongeville J., Simon C., Bingham A., Amouyel P. et al. Alcohol intake and diet in France, the prominent role of lifestyle. Eur Heart J 2004; 25: 11531162.
  • 48
    Barefoot J. C., Grønbaek M., Feaganes J. R., McPherson R. S., Williams R. B., Siegler I. C. Alcoholic beverage preference, diet, and health habits in the UNC Alumni Heart Study. Am J Clin Nutr 2002; 76: 466472.
  • 49
    Piazza-Gardner A. K., Barry A. E. Examining physical activity levels and alcohol consumption: are people who drink more active? Am J Health Promot 2012; 26: e95104.
  • 50
    McCann S. E., Sempos C., Freudenheim J. L., Muti P., Russell M., Nochajski T. H. et al. Alcoholic beverage preference and characteristics of drinkers and nondrinkers in western New York (United States). Nutr Metab Cardiovasc Dis 2003; 13: 211.
  • 51
    Burke V., Lee A. H., Hunter E., Spargo R., Smith R., Beilin L. J. et al. Alcohol intake and incidence of coronary disease in Australian aborigines. Alcohol Alcohol 2007; 42: 119124.
  • 52
    Camacho T. C., Kaplan G. A., Cohen R. D. Alcohol consumption and mortality in Alameda County. J Chronic Dis 1987; 40: 229236.
  • 53
    Naimi T. S., Brown D. W., Brewer R. D., Giles W. H., Mensah G., Serdula M. K. et al. Cardiovascular risk factors and confounders among nondrinking and moderate-drinking U.S. adults. Am J Prev Med 2005; 28: 369373.
  • 54
    Holt-Lunstad J., Smith T. B., Layton J. B. Social relationships and mortality risk: a meta-analytic review. PLoS Med 2010; 7: e1000316.
  • 55
    Fillmore K. M., Golding J. M., Graves K. L., Kniep S., Leino E. V., Romelsjö A. et al. Alcohol consumption and mortality. I. Characteristics of drinking groups. Addiction 1998; 93: 183203.
  • 56
    Manrique-Garcia E., Sidorchuk A., Hallqvist J., Moradi T. Socioeconomic position and incidence of acute myocardial infarction: a meta-analysis. J Epidemiol Community Health 2011; 65: 301309.
  • 57
    Smith G. D., Shipley M. J., Rose G. Magnitude and causes of socioeconomic differentials in mortality: further evidence from the Whitehall Study. J Epidemiol Community Health 1990; 44: 265270.
  • 58
    Hemingway H., Shipley M., Macfarlane P., Marmot M. Impact of socioeconomic status on coronary mortality in people with symptoms, electrocardiographic abnormalities, both or neither: the original Whitehall study 25 year follow up. J Epidemiol Community Health 2000; 54: 510516.
  • 59
    Marmot M. The Status Syndrome: How Social Standing Affects Our Health and Longevity. London: Times Books; 2004.
  • 60
    Tjønneland A. M., Grønbæk M. N., Stripp C., Overvad O. K. Wine intake and diet in a random sample of 48 763 Danish men and women. Am J Clin Nutr 1999; 69: 4954.
  • 61
    Johansen D., Friis K., Skovenborg E., Grønbaek M. Food buying habits of people who buy wine or beer: cross sectional study. BMJ 2006; 332: 519522.
  • 62
    Mortensen E. L., Jensen H. H., Sanders S. A., Reinisch J. M. Better psychological functioning and higher social status may largely explain the apparent health benefits of wine: a study of wine and beer drinking in young Danish adults. Arch Intern Med 2001; 161: 18441848.
  • 63
    Paschall M., Lipton R. I. Wine preference and related health determinants in a U.S. national sample of young adults. Drug Alcohol Depend 2005; 78: 339344.
  • 64
    Roy A., Prabhakaran D., Jeemon P., Thankappan K. R., Mohan V., Ramakrishnan L. et al. Impact of alcohol on coronary heart disease in Indian men. Atherosclerosis 2010; 210: 531535.
  • 65
    Bazzano L. A., Gu D., Reynolds K., Chen J., Wu X., Chen C. S. et al. Alcohol consumption and risk of coronary heart disease among Chinese men. Int J Cardiol 2009; 135: 7885.
  • 66
    Stampfer M. J., Colditz G. A. Estrogen replacement therapy and coronary heart disease: a quantitative assessment of the epidemiologic evidence. Prev Med 1991; 20: 4763.
  • 67
    Hulley S., Grady D., Bush T., Furberg C., Herrington D., Riggs B. et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. Heart and Estrogen/progestin Replacement Study (HERS) Research Group. JAMA 1998; 280: 605613.
  • 68
    Rossouw J. E., Anderson G. L., Prentice R. L., LaCroix A. Z., Kooperberg C., Stefanick M. L. et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. JAMA 2002; 288: 321333.
  • 69
    Grady D., Herrington D., Bittner V., Blumenthal R., Davidson M., Hlatky M. et al. Cardiovascular disease outcomes during 6.8 years of hormone therapy: heart and Estrogen/progestin Replacement Study follow-up (HERS II). JAMA 2002; 288: 4957.
  • 70
    Rimm E. B., Stampfer M. J., Ascherio A., Giovannucci E., Colditz G. A., Willett W. C. Vitamin E consumption and the risk of coronary heart disease in men. N Engl J Med 1993; 328: 14501456.
  • 71
    Stampfer M. J., Hennekens C. H., Manson J. E., Colditz G. A., Rosner B., Willett W. C. Vitamin E consumption and the risk of coronary disease in women. N Engl J Med 1993; 328: 14441449.
  • 72
    Mehta J. Intake of antioxidants among American cardiologists. Am J Cardiol 1997; 79: 15581560.
  • 73
    Pandey D. K., Shekelle R., Selwyn B. J., Tangney C., Stamler J. Dietary vitamin C and beta-carotene and risk of death in middle-aged men. The Western Electric Study. Am J Epidemiol 1995; 142: 12691278.
  • 74
    Rapola J. M., Virtamo J., Ripatti S., Huttunen J. K., Albanes D., Taylor P. R. et al. Randomised trial of alpha-tocopherol and beta-carotene supplements on incidence of major coronary events in men with previous myocardial infarction. Lancet 1997; 349: 17151720.
  • 75
    Virtamo J., Rapola J. M., Ripatti S., Heinonen O. P., Taylor P. R., Albanes D. et al. Effect of vitamin E and beta carotene on the incidence of primary nonfatal myocardial infarction and fatal coronary heart disease. Arch Intern Med 1998; 158: 668675.
  • 76
    Wijndaele K., Brage S., Besson H., Khaw K. T., Sharp S. J., Luben R. et al. Television viewing time independently predicts all-cause and cardiovascular mortality: the EPIC Norfolk study. Int J Epidemiol 2011; 40: 150159.
  • 77
    Roerecke M., Rehm J. Ischemic heart disease mortality and morbidity rates in former drinkers: a meta-analysis. Am J Epidemiol 2011; 173: 245258.
  • 78
    Fillmore K. M., Kerr W. C., Stockwell S., Chikritzhs T., Bostrom A. Moderate alcohol use and reduced mortality risk: systematic error in prospective studies. Addict Res Theory 2006; 14: 101132.
  • 79
    Fillmore K. M., Stockwell T., Chikritzhs T., Bostrom A., Kerr W. Moderate alcohol use and reduced mortality risk: systematic error in prospective studies and new hypotheses. Ann Epidemiol 2007; 17: S1623.
  • 80
    Stockwell T., Chikritzhs T., Bostrom A., Fillmore K., Kerr W., Rehm J. et al. Alcohol-caused mortality in Australia and Canada: scenario analyses using different assumptions about cardiac benefit. J Stud Alcohol Drugs 2007; 68: 345352.
  • 81
    Rubin E., Lieber C. S. Fatty liver, alcoholic hepatitis and cirrhosis produced by alcohol in primates. N Engl J Med 1974; 290: 128135.
  • 82
    Cudd T. A. Animal model systems for the study of alcohol teratology. Exp Biol Med (Maywood) 2005; 230: 389393.
  • 83
    Warren K. R., Hewitt B. G., Thomas J. D. Fetal alcohol spectrum disorders: research challenges and opportunities. Alcohol Res Health 2011; 34: 411.
  • 84
    World Cancer Research Fund/American Institute for Cancer Research. Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective. Washington, DC: AICR; 2007.
  • 85
    Hill A. B. The environment and disease: association or causation? Proc R Soc Med 1965; 58: 295300.
  • 86
    Mukamal K. J., Chen C. M., Rao S. R., Breslow R. A. Alcohol consumption and cardiovascular mortality among U.S. adults, 1987 to 2002. J Am Coll Cardiol 2010; 55: 13281335.
  • 87
    Brien S. E., Ronksley P. E., Turner B. J., Mukamal K. J., Ghali W. A. Effect of alcohol consumption on biological markers associated with risk of coronary heart disease: systematic review and meta-analysis of interventional studies. BMJ 2011; 342: d636.
  • 88
    Clayton D., Hills M. Chapter 25. Models for dose–response. Statistical Models in Epidemiology. New York, NY: Oxford University Press; 1993.