Critique 227: A reduced risk of chronic obstructive pulmonary disease (COPD) associated with the intake of some alcoholic beverages – 29 May 2019

Kaluza J, Harris HR, Linden A, Wolk A.  Alcohol Consumption and Risk of Chronic Obstructive Pulmonary Disease: A Prospective Cohort Study of Men.  Am J Epidemiol  2019;188:907–991.

Authors’ Abstract

Studies indicate an inverse association between moderate alcohol consumption and chronic inflammatory diseases; however, the association between alcohol consumption and chronic obstructive pulmonary disease (COPD) incidence has not been widely studied.  We investigated the associations of total alcohol consumption and intake of specific alcoholic beverages with risk of COPD in a population-based prospective cohort study, the Cohort of Swedish Men (n=44,254).   Alcohol consumption was assessed with a self-administered questionnaire in 1997.

During follow-up (1998–2014), 2,177 COPD cases were ascertained.  Moderate alcohol consumption was associated with the lowest risk of COPD.  A J-shaped association was observed for ethanol consumption (P for nonlinearity =0.003) and beer consumption (P for nonlinearity < 0.001); for wine consumption, a U-shaped association was observed (P for nonlinearity <0.001).  Defining a “standard drink” as 12 g of ethanol, the multivariable-adjusted hazard ratios were 0.77 (95%confidence interval (CI): 0.66,0.90) and 0.92 (95% CI: 0.81,1.05) for beer consumption of 4.1–6.0 and >6.0 standard drinks/week(SDW) versus <1.0 SDW, respectively; 0.80 (95%CI:0.69,0.93) and 1.00 (95%CI:0.83,1.21 ) for wine consumption of 2.0 – 4.0 and >4.0 SDW versus <1.0 SDW, respectively; and 1.10 (95% CI: 0.98, 1.24) and 1.20 (95% CI: 0.99, 1.44) for liquor consumption of 2.0–4.0 and >4.0 SDW versus <1.0 SDW, respectively.

In conclusion, our findings suggest that moderate beer and wine consumption, but not liquor consumption, may decrease risk of COPD.  Additional studies are needed to confirm these associations.

Forum Comments

Chronic obstructive pulmonary disease (COPD) is a major health problem and cause of death throughout the world.  While cigarette smoking is known to be the most important risk factor, there are limited data available on the relation of alcohol consumption to the condition.  Given that moderate alcohol intake has been shown to decrease the risk of many inflammatory conditions, it may also affect the risk of COPD, which is associated with marked inflammation of lung tissues.

The present analyses are based on a large population-based cohort study of more than 44,000 Swedish men who were followed for up to 17 years for the development of COPD.  An assessment of alcohol consumption, as well as beverage-specific intake, was made in 1997, and was related to the subsequent diagnosis of COPD, which occurred in 2,177 subjects during follow up.  Overall, there was a clear J-shaped curve between alcohol intake and COPD.  The authors conclude from their beverage-specific analyses that the moderate intake of beer or wine, but not liquor, may decrease the risk of COPD.

Among the strengths of the study is that it is population based and involves a large number of subjects, with more than 2,000 developing COPD.  The authors had data permitting them to evaluate a number of demographic and lifestyle factors as potential confounders (including smoking status and pack-years of smoking, education, physical activity, BMI, and some indices of diet).   However, there were very few ex-smokers or current smokers among the non-drinkers, making it somewhat difficult to adequately adjust for cigarette use as a cause of COPD.  (The authors attempted to adjust for this by also reporting stratified analyses of smokers versus non-smokers, but the numbers of COPD cases among non-smokers was very low.)  Further, their measure of socio-economic status was based on education, which varied markedly between consumers of each beverage, with wine consumers being more likely to have a university education than consumers of other beverages.  Such wide differences between categories of exposure always raise questions as to the adequacy of evaluation for confounding by usual statistical techniques.

The authors also describe a number of potential biologic mechanisms by which alcohol and/or polyphenols might affect lung tissue and increase or decrease the risk of COPD.  While there is increasing research on such mechanisms, the ultimate reasons why consumers of wine or beer may have a lower risk of disease remain undetermined, and residual confounding by a number of lifestyle factors may still play a role.

Comments of specific Forum members:  Reviewer Van Velden wrote: “This is indeed an interesting study that demonstrated a clear beneficial influence of antioxidant-rich alcoholic beverages on COPD.  The investigators observed other lifestyle factors such as smoking and exercise as well, but educational factors and occupational exposure to pollutants may also influence the results.  It is interesting to note that they also mention genetic factors that may have an influence on the outcome.  This may warrant future studies to investigate the influence of genetics on COPD.”

Forum member Finkel noted: “Regarding the drinking/COPD relationship as studied by Kaluza, et al, I believe we and others reading the paper should exercise caution in drawing conclusions because of the small and complex differences being measured in the face of variables notoriously at risk of influence by a host of potential confounding factors, especially those of numerous elements of what may be called lifestyle.”

Forum member Skovenborg commented: “I agree with Finkel, and I also think that Van Velden  has a good point.  The study is well done and reported with due diligence; however, the problem is that we at this point do not know whether the results are due to anti-inflammatory effects of polyphenols in beer and wine, or to the different drinking patterns of beer-, wine-, and liquor-drinkers or residual confounding.”

A number of Forum members were also worried by potential inadequacies in estimating alcohol consumption.  Reviewer Skovenborg stated: “I think that this study is imperfect regarding one of the most important conditions for healthy enjoyment of alcohol: drinking pattern (as a regular, steady consumption of light to moderate amounts of alcohol, especially wine or beer, with meals generally relates to lower risk of disease).  Also, as stated previously by Zhang, ‘The differences in age and smoking status (the two strongest risk factors for COPD) among the drinking  categories are substantial, making it challenging to appropriately account for their confounding effect using the multivariable-adjusted regression models.’ Another weakness is that only alcohol consumption at baseline was assessed, with no ability to judge the effects of changes in consumption.”

Skovenborg continued: “I also notice the very narrow windows for a positive effect of beer and wine: up to 4 glasses of wine/6 glasses of beer per week.  Keeping in mind that this is a study from Sweden with a particular anti-alcohol tradition and anti-alcohol political climate, it might be of interest to take a look at studies of underreporting in Sweden.  One study found a coverage rate of 51% in Sweden which might be marginally better than in the other Scandinavian countries due to the Systembolaget organization of alcohol sales in Sweden; Mäkelä et al reported that the coverage rate was 45% for Denmark, 40% for Finland, 43% for Norway, and 51% for Sweden.

“Another study (Kühlhorn & Leifman) found differential underreporting regarding information about drinking pattern; in the present paper, Kaluza et al present a ‘normal week’s consumption’ category.  Two Swedish alcohol surveys were compared in a search for a reasonable explanation of the large difference in their coverage rates, namely 75% and 28%.  The technique used in the survey with a very high coverage rate (Survey A) takes into consideration the actual drinking pattern of the population studied (i.e., the concentration of drinking on weekends).  By dividing a ‘normal week’s consumption’ into four units (Monday-Thursday, Friday, Saturday and Sunday), the technique allows one to average periods with varying drinking habits.  In the survey with a low coverage rate (Survey B) a “normal week’s consumption” was not so divided.  A third study (Ramstedt) found differential coverage rates for respectively beer & spirits and wine.  Of the recorded amount of purchases at Systembolaget, 87% was reported in the survey, compared with the 40-60% usually found for self-reported consumption.  Significant differences across beverages were revealed, showing a lower coverage rate for beer and spirits and a higher coverage rate for wine and cider.”  It is not possible to determine the extent to which such problems in assessing actual alcohol intake may have played in the reported beverage-specific results in this study.

References from Forum critique

Kühlhorn E, Leifman H.  Alcohol surveys with high and low coverage rate: a comparative analysis of survey strategies in the alcohol field.  J Stud Alcohol 1993;54:542-554.

Mäkelä P, Fonager K, Hibell B, Nordlund S, Sabroe S, Simpura J.  Episodic heavy drinking in four Nordic countries: a comparative survey.  Addiction 2001;96:1575-1588.

Ramstedt M.  How much alcohol do you buy?  A comparison of self-reported alcohol purchases with actual sales. Addiction 2010;105:649-654.

Forum Summary

Chronic obstructive pulmonary disease (COPD) is a major health problem and cause of death throughout the world.  While cigarette smoking is known to be the most important risk factor, there are limited data available on the relation of alcohol consumption to the condition.  Given that moderate alcohol intake has been shown to decrease the risk of many inflammatory conditions, it may also affect the risk of COPD, which is associated with marked inflammation of lung tissues.

In the present study, the investigators related total alcohol consumption and intake of specific alcoholic beverages with the risk of COPD, based on data from more than 44,000 Swedish men in a population-based prospective cohort study.  A total of 2,177 COPD cases were ascertained during follow up that extended up to 17 years.  In their analyses, the authors had data permitting them to evaluate a number of demographic and lifestyle factors as potential confounders (including smoking status and pack-years of smoking, education, physical activity, BMI, and some indices of diet).   However, there were very few ex-smokers or current smokers among the non-drinkers, making it somewhat difficult to adequate adjust for cigarette use as a cause of COPD.

Forum members thought that there are a number of deficiencies in this study that somewhat weaken its conclusions; especially important was the lack of ability to consider the pattern of drinking (regular, moderate versus binge drinking).  Still, there is a strong consistency between the results of this study (indicating a “J-shaped” or “U-shaped” curve for alcohol intake and COPD) and results from extensive previous epidemiologic and experimental research.

Overall, the cumulative research findings now suggest that pulmonary disease might be included among the “diseases of ageing” that show a reduction in risk among moderate consumers of certain alcoholic beverages.  Based on some previous epidemiologic studies and results from numerous experimental studies, data now suggest that, in addition to alcohol, the polyphenols present in wine and some beers may offer protection against COPD.  While in the present study red wine consumption was associated with the lowest risk of COPD, Forum members acknowledge residual confounding by other lifestyle factors may still play a role.

*                      *                      *                      *                      *                      *

Comments on this critique by the International Scientific Forum on Alcohol Research have been provided by the following members:

Harvey Finkel, MD, Hematology/Oncology, Retired (Formerly, Clinical Professor of Medicine, Boston University Medical Center, Boston, MA, USA)

David Van Velden, MD, Dept. of Pathology, Stellenbosch University, Stellenbosch, South Africa

Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark

R. Curtis Ellison, MD, Professor of Medicine, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA

Pierre-Louis Teissedre, PhD, Faculty of Oenology–ISVV, University Victor Segalen Bordeaux 2, Bordeaux, France

Yuqing Zhang, MD, DSc.  Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA.

Adrew L. Waterhouse, PhD, Department of Viticulture and Enology, University of California, Davis, USA