Alcohol Consumption and Risk of Chronic Obstructive Pulmonary Disease

A Prospective Cohort Study of Men

Joanna Kaluza; Holly R. Harris; Anders Linden; Alicja Wolk


Am J Epidemiol. 2019;188(5):907-916. 

In This Article


In this large prospective study of men with relatively low-to-moderate alcohol consumption, both never/occasional drinkers (<1.0 drink/week) and heavy drinkers (>20.0 drinks/week) had an increased risk of COPD in comparison with moderate drinkers (7.0–14.0 drinks/week). A J-shaped association for beer consumption and a U-shaped association for wine consumption were observed. For liquor, we observed a suggestion of an increased risk starting at low liquor consumption (≥1.0 drinks/week).

To the best of our knowledge, this is the first prospective cohort study to have examined alcohol consumption in relation to COPD incidence. Our results are in agreement with those of studies which examined the associations between alcohol consumption and COPD mortality[13,14] and COPD-related deaths.[15] The observed protective association of moderate alcohol consumption with COPD incidence is in line with findings of previous prospective cohort studies on COPD mortality,[13–15] including a Chinese cohort study (<1, 1–21, and >21 drinks/week vs. never drinking: HR = 0.70 (95% CI: 0.57, 0.86), HR = 0.70 (95% CI: 0.53, 0.93), and HR = 0.95 (95% CI: 0.58, 1.55), respectively)[14] and a small cohort study (73 COPD deaths) from 3 European countries (>1 drink/week–≤3 drinks/day and >3 drinks/day vs. ≤1 drink/week: HR = 0.60 (95% CI: 0.33, 1.09) and HR = 1.25 (95% CI: 0.47, 3.31), respectively).[13] Similar results for the association between alcohol consumption and risk of COPD-related mortality have also been observed in a sample of 83,533 Chinese men.[15] Men who had consumed alcohol at least 12 times during the past year, as compared with men who drank less frequently, had a 16% (95% CI: 5, 26) lower risk of COPD-related mortality.[15]

In our study, we examined the consumption of specific types of alcohol. This is in contrast to previous prospective studies on alcohol consumption and COPD-related mortality, which only examined associations with total alcohol intake. In a cohort study by Tran et al.,[16] the authors mentioned that risk of hospitalization from COPD causes was lower among participants who consumed 1–2 alcoholic drinks/day than among nonconsumers, and associations were similar for wine, beer, and liquor. Our results are in line with findings obtained for other diseases, especially cardiovascular diseases.[25] In a recent review, Arranz et al.[26] highlighted that of the different types of beverages, moderate red wine consumption and, to a lesser degree, beer consumption were inversely associated with cardiovascular diseases, atherosclerosis, hypertension, certain types of cancer, and type 2 diabetes. However, results of a meta-analysis of intervention studies suggested that the associations between alcohol consumption and cardiovascular biomarkers are similar across beverage types.[27] We can hypothesize that the observed differences between type of alcoholic beverage and COPD incidence may partially be due to varying lifestyle factors, including drinking patterns that vary by alcoholic beverage preference. In this study, men who drank wine were more likely to have a university education, while those who drank liquor more frequently smoked cigarettes.

When we stratified results by smoking status, we observed that the lower risk of COPD incidence in men with moderate alcohol consumption was independent of smoking status. In contrast, in a prior study among Chinese men and women, occasional and moderate alcohol consumption were associated with reduced risk of COPD mortality only in ex-smokers.[14]

In a literature review, Sisson[28] concluded that the association between alcohol consumption and lung airway functions depends on the concentration and duration of exposure. Short exposure to moderate concentrations of alcohol may enhance mucociliary clearance, may stimulate bronchodilation, and probably attenuates the airway inflammation and injury observed in COPD, while prolonged and heavy exposure to alcohol impairs mucociliary clearance and may worsen lung function in COPD patients.[28]

The pathogenesis of COPD may involve several pathways of inflammation, and one of the most well-documented ones is that of oxidative stress.[29,30] Via these mechanisms, proinflammatory compounds present in cigarette smoke may trigger chronic inflammatory events, leading to local tissue remodeling that irreversibly impairs lung function over time.[29,30] There are data indicating that impaired antioxidant status correlates with increased severity of COPD.[31] Higher plasma lipid peroxidation has been associated with higher risk of progression in COPD patients, whereas higher activity of potentially protective compounds, such as catalase and erythrocyte glutathione, has been associated with decreased risk of progression.[32] It has been observed that moderate alcohol consumption is associated with reduced levels of systemic inflammatory markers[33] and no increased risk of acute exacerbations of COPD.[34,35]

We can hypothesize that the protective association for moderate alcohol consumption, especially beer and wine consumption, relates to the antioxidant impact of polyphenols present in alcoholic beverages. Compared with gin (a polyphenol-free alcoholic beverage), red wine (a polyphenol-rich beverage) has been shown to reduce plasma superoxide dismutase activity and malondialdehyde levels and to increase the lag phase time of low-density lipoprotein oxidation.[36] In contrast, when alcohol is consumed in high amounts over extended time periods, increased production of free radicals and decreased levels of the antioxidant glutathione within the alveolar space (by as much as 80%–90%) is observed.[37] As a consequence, the production of alveolar epithelial surfactant and barrier integrity may be impaired and alveolar macrophage function may be decreased—phenomena that render the lung susceptible to oxidant-mediated injury.[37] Moreover, alcohol may affect bronchial epithelial cells, one of the most critical immune barriers in the lungs. Alcohol also targets generic intracellular signaling pathways in key structural lung cells, such as the guanosine triphosphatase RhoA (Ras homolog gene family, member A)[38] and the glucocorticoid-induced leucine zipper gene (GILZ) in bronchial epithelial cells.[39] All of these processes are likely to increase the tissue damage caused by the airway infections that repeatedly occur in COPD.

A main strength of our study was its population-based prospective design, where ascertainment of alcohol exposure was independent of outcome ascertainment, as well as the collection of detailed information on diet and potential risk factors for COPD. Moreover, our analytical cohort was well representative of the Swedish male population in terms of age, education, prevalence of overweight/obesity, and smoking. Furthermore, a large number of incident COPD cases allowed us to conduct separate analyses for specific alcoholic beverages.

A main limitation of our study was the lack information on factors which may affect COPD risk apart from cigarette smoking. These include occupational and environmental exposure to dust, gases, and vapors and history of severe respiratory infection. In addition, we could not examine cigarette smoking intensity or cigar or pipe smoking. Estimates of alcohol consumption and other factors were based on self-reports made at baseline, and the exposures could have changed during the follow-up period. A possible source of nondifferential misclassification of the exposure was also likely underreported alcohol consumption. Self-reported alcohol consumption may be burdened by recall bias due to social desirability and may introduce bias away from the null, even with nondifferential misclassification. It is probable that men with heavy alcohol consumption and binge drinkers were underrepresented in the study. Moreover, the lack of a J-shaped association between COPD risk and liquor consumption may have been not only the result of the absence of antioxidants in this type of alcohol but also a consequence of a harmful drinking pattern.

In our study, we observed that men with higher wine consumption were more likely to have a university education than those with lower wine consumption, while men with higher liquor consumption were more likely to be current smokers and less likely to be never smokers. To address these issues, we adjusted the models for many potential confounders and adjusted intakes of specific alcoholic beverages for each other. Further, the food frequency questionnaire used in this study, including our exposure variable (ethanol), had high validity. However, unmeasured or residual confounding cannot be completely ruled out. In a 2014 Mendelian randomization study that utilized the alcohol dehydrogenase 1B gene (ADH1B) variant rs1229984 as a proxy for alcohol consumption, Holmes et al.[40] reported that persons genetically predicted to be nondrinkers and to consume less alcohol had a reduced risk of coronary heart disease compared with those without this variant. The authors suggested that studies on the health-promoting role of alcohol consumption for moderate drinkers may be influenced by residual confounding or selection bias.[40] Future Mendelian randomization studies of variant ADH1B rs1229984 and COPD could provide more clarity with regard to this issue.

Given that COPD as a diagnosis was formally defined during the 1970s[40] and that awareness of this disease among health-care professionals continued to evolve during the 1990s and later,[23] we cannot rule out underascertainment of COPD in the early years of follow-up. Underascertainment of cases may lead to attenuation of risk estimates. Indeed, we observed that risk estimates for the first years of follow-up among men with high alcohol consumption were attenuated in comparison with the later period. However, we found that the associations remained similar after excluding the first 5 years of follow-up. In addition, we cannot rule out the possibility that some patients classified as having COPD had obtained the diagnosis without the correct spirometry assessment, even though this investigation is formally required for a clinical diagnosis. We cannot exclude possible misdiagnosis of COPD as well; patients with heart failure are relatively frequently misdiagnosed with COPD,[41,42] and this could result in an inverse alcohol-COPD association.

In conclusion, our results are consistent with the inverse associations observed between moderate alcohol consumption and risk of cardiovascular disease and suggest that moderate alcohol consumption may protect against risk of COPD. Our findings warrant confirmation in future prospective studies—specifically whether low-to-moderate consumption of beer and wine may decrease risk of COPD and whether even small amounts of liquor intake may increase the risk of COPD among men. In future research, investigators could consider a Mendelian randomization approach, which is generally less susceptible to confounding and selection bias than observational studies.[43]