Abstract
Background
COPD is characterised by multimorbidity and high mortality. Although alcohol consumption is associated with comorbidities common in COPD, its impact on COPD outcomes is uncertain. We investigated associations between baseline alcohol consumption and 10-year mortality, stratified by COPD status and examined whether comorbidity prevalence differed across alcohol categories.
Methods
COPD was defined by post-bronchodilator forced expiratory volume in 1 s/forced vital capacity <0.7; alcohol as ≤10 versus >10 units·week−1. Comorbidities were ascertained from national registries. Cox models estimated mortality hazard ratios (HRs) adjusting for age, sex, smoking, education, body mass index and comorbidities. Inverse probability weighting addressed confounding. Logistic regression examined alcohol–comorbidity associations with false discovery rate correction.
Results
In 2004–2006, 4528 Danish adults aged 45–84 years were recruited. Among 3468 participants with complete alcohol and COPD data (mean age 64 years; 65% male; 12% COPD), 593 deaths occurred over 10 years. Alcohol >10 units·week−1 was not associated with mortality in unadjusted (HR 0.97 (95% CI 0.80–1.18)) or fully adjusted (HR 0.87 (95% CI 0.71–1.07)) models. COPD was strongly associated with mortality (adjusted HR 2.00 (95% CI 1.64–2.43)). No alcohol×COPD interaction was detected. Cardiovascular comorbidities were generally more prevalent among participants with COPD, although patterns varied. No associations between alcohol and individual comorbidities survived multiple testing correction.
Conclusions
In this cohort with predominantly moderate consumption, alcohol intake >10 units·week−1 was not associated with 10-year mortality in adults with or without COPD, whereas COPD conferred excess risk independent of comorbidity and lifestyle factors. These findings should not be interpreted as evidence that alcohol consumption is safe.
Shareable abstract
In 3468 adults followed for 10 years, moderate-to-high alcohol consumption was not associated with mortality or comorbidity. COPD conferred two-fold increased mortality risk independent of alcohol, comorbidity and lifestyle factors. https://bit.ly/41r07tN
Introduction
COPD ranks third globally as a cause of death [1]. Beyond respiratory manifestations, COPD is increasingly recognised as a multisystem condition with prevalent comorbidities including cardiovascular disease, metabolic syndrome, depression and lung cancer that adversely affect morbidity and mortality [1, 2]. The COPD-specific Comorbidity Test (COTE) and population-based studies confirm that comorbidity burden influences mortality, beyond lung function impairment [3, 4].
Alcohol consumption contributes to more than 2.5 million deaths annually [5, 6]. While historically protective effects were attributed to moderate drinking, recent evidence suggests even low-level consumption may increase cancer risk and all-cause mortality [7, 8]. Harmful alcohol consumption contributes to hypertension, atrial fibrillation, liver disease, malignancy and depression, conditions also prevalent in COPD [9–12].
Biological mechanisms support a potential role for alcohol in respiratory health. Chronic alcohol consumption impairs mucociliary clearance, disrupts epithelial barrier integrity and alters immune responses, increasing susceptibility to respiratory infections, particularly community-acquired pneumonia, a major cause of acute deterioration and death in COPD [13–15].
Epidemiological findings on alcohol consumption and COPD outcomes remain limited and inconsistent. Most prior work focused on short-term outcomes or hospitalised populations, with limited characterisation of comorbidity and mortality patterns. National hospital data indicate alcohol use disorder is present in ∼10% of COPD admissions, with adverse outcomes [16]. Studies of high-risk COPD patients found no clear association between self-reported baseline alcohol consumption and future exacerbations [17, 18]. Population-based studies suggest non-linear associations between alcohol consumption and COPD incidence and mortality [19–21].
These findings highlight the need to clarify how alcohol consumption relates to long-term mortality and comorbidity burden in COPD. We assessed associations between baseline alcohol consumption and 10-year all-cause mortality in a population-based cohort stratified by COPD status. As a secondary aim, we examined comorbidity patterns across COPD and alcohol categories.
Methods
Study design and population
We conducted this study in a multicentre, population-based cohort from two Danish counties (Viborg and North Jutland) with baseline assessment in 2004–2006 and follow-up assessments in 2008–2009 and 2016. The inclusion process has been described previously [22]. This analysis used baseline exposure data with prospective mortality follow-up through 31 December 2014, providing up to 10 years of observation. Baseline was defined as the date of study participation (2004–2006), and all exposures and covariates were measured at this time.
Assessments
Sociodemographic variables
Self-reported variables included gender, age, years of schooling (≤7, 8–10 and >10 years) and higher education (none, <3, 3–4 and >4 years).
Alcohol consumption
Alcohol consumption was assessed at the baseline examination (2004–2006), which defined time zero for follow-up. Participants reported beverage-specific consumption across life decades; current intake was derived from reported consumption from age ≥50 years. In rare cases of internal inconsistency between frequency responses and reported unit consumption, classification was based on the quantitative intake. One unit was defined as 12 g ethanol. Primary analyses dichotomised intake as ≤10 versus >10 units·week−1. Sensitivity analyses used ≤15 versus >15 units·week−1, g·day−1 categories (0, <12, 12–24, 25–40 and >40) and drink-years ((units·week−1)×years: 0, 1–50, 51–100 and >100). Former drinkers were defined as participants reporting prior but no current intake; lifetime abstainers reported zero alcohol intake across life decades. In the primary dichotomised analysis, former drinkers were included in the ≤10 units·week−1 category. To assess potential reverse causation (“sick quitter” bias), sensitivity analyses excluded former drinkers. Although alcohol consumption was reassessed at later follow-up visits, only baseline values were used in the present analyses.
Spirometry
Spirometry followed European Respiratory Society/American Thoracic Society guidelines [23, 24]. COPD was defined as post-bronchodilator forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) <0.70.
Smoking
Participants were classified as never, former or current smokers, with pack-years calculated as (cigarettes·day−1/20)×years smoked [25].
Comorbidities
Comorbidities were ascertained from the Danish National Patient Registry using International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) codes before the individual date of baseline assessment [26]. Conditions primarily diagnosed and managed in hospital settings (e.g. myocardial infarction, heart failure and malignancy) were identified using ICD-10 codes alone.
For conditions frequently managed in primary care and therefore less consistently captured in hospital registries (e.g. hypertension, diabetes, anxiety and depression), prescription data from the Danish National Prescription Registry were incorporated (supplementary table S1) [27]. Redeemed prescriptions within the 6 months preceding individual baseline date were used. For diabetes, anxiety and depression, ICD-10 diagnoses were combined with relevant Anatomical Therapeutic Chemical (ATC) codes to improve ascertainment across healthcare sectors. Hypertension was defined using prescription data alone, requiring redemption of ≥2 antihypertensive drug classes on ≥2 occasions also within 6 months preceding the individual baseline date (e.g. diuretic and anti-adrenergic) (supplementary table S1).
The initial comorbidity set included conditions from the Charlson Comorbidity Index, Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines, COTE index and depression [3, 28, 29]. To avoid construct overlap with COPD, we excluded respiratory conditions. HIV/AIDS and certain rare malignancies were not observed in the analytic cohort (zero baseline cases) and were therefore excluded from modelling, as variables without variation cannot be included in regression analyses. For regression analyses examining associations between COPD, alcohol and specific comorbidities, we restricted the focus to conditions with more than 40 prevalent cases at baseline to ensure adequate statistical power. 13 comorbidities met this criterion and are presented in table 1; additional conditions with lower prevalence are presented in supplementary table S2. All comorbidity definitions and ICD-10/ATC codes are provided in supplementary table S1. Comorbidity ascertainment was based on nationwide registry data with high diagnostic validity, reducing the likelihood that zero prevalence reflects systematic underreporting. Asthma was defined as self-reported physician-diagnosed asthma based on questionnaire responses.
TABLE 1.
Baseline characteristics stratified by COPD status and alcohol consumption
| Overall | No COPD/ low alcohol |
COPD/ low alcohol |
No COPD/ high alcohol |
COPD/ high alcohol |
Missing (%) | |
|---|---|---|---|---|---|---|
| Participants | 4528 | 2373 | 289 | 711 | 95 | 23.4# |
| Age (years) | 64.05±10.06 | 62.10±10.25 | 68.07±8.73 | 62.66±9.17 | 67.65±8.57 | 0.0 |
| Age categories | 0.0 | |||||
| 45–54 years | 988 (21.8) | 686 (28.9) | 29 (10) | 158 (22.2) | 7 (7.4) | |
| 55–64 years | 1187 (26.2) | 658 (27.7) | 59 (20.4) | 235 (33.1) | 21 (22.1) | |
| 65–74 years | 1618 (35.7) | 729 (30.7) | 130 (45) | 253 (35.6) | 43 (45.3) | |
| 75–84 years | 735 (16.2) | 300 (12.6) | 71 (24.6) | 65 (9.1) | 24 (25.3) | |
| Sex | 0.0 | |||||
| Female | 1562 (34.5) | 998 (42.1) | 62 (21.5) | 104 (14.6) | 8 (8.4) | |
| Male | 2966 (65.5) | 1375 (57.9) | 227 (78.5) | 607 (85.4) | 87 (91.6) | |
| BMI (kg·m−2) | 27.23±4.58 | 27.24±4.67 | 26.44±4.90 | 27.18±3.80 | 25.58±3.68 | 0.1 |
| BMI categories | 0.1 | |||||
| Underweight/normal (<25 kg·m−2) | 1443 (31.9) | 781 (32.9) | 120 (41.5) | 202 (28.4) | 44 (46.3) | |
| Overweight (25–29.9 kg·m−2) | 2075 (45.9) | 1057 (44.5) | 117 (40.5) | 364 (51.2) | 45 (47.4) | |
| Obese (≥30 kg·m−2) | 1006 (22.2) | 532 (22.4) | 52 (18.0) | 145 (20.4) | 6 (6.3) | |
| Schooling | 0.5 | |||||
| <7 years | 2245 (49.8) | 1073 (45.2) | 176 (60.9) | 253 (35.6) | 52 (54.7) | |
| 8–10 years | 1552 (34.4) | 882 (37.2) | 79 (27.3) | 293 (41.2) | 27 (28.4) | |
| >10 years | 710 (15.8) | 409 (17.2) | 34 (11.8) | 164 (23.1) | 16 (16.8) | |
| Higher education | 6.6 | |||||
| None | 1165 (27.6) | 586 (24.7) | 99 (34.3) | 96 (13.5) | 19 (20) | |
| <3 years | 865 (20.5) | 512 (21.6) | 46 (15.9) | 109 (15.3) | 17 (17.9) | |
| 3–4 years | 1573 (37.2) | 876 (36.9) | 85 (29.4) | 320 (45) | 39 (41.1) | |
| >4 years | 625 (14.8) | 290 (12.2) | 43 (14.9) | 173 (24.3) | 18 (18.9) | |
| Smoking status | 0.6 | |||||
| Never | 1520 (33.8) | 972 (41) | 35 (12.1) | 192 (27) | 5 (5.3) | |
| Former | 1949 (43.3) | 951 (40.1) | 143 (49.5) | 352 (49.5) | 43 (45.3) | |
| Current | 1034 (23) | 443 (18.7) | 111 (38.4) | 167 (23.5) | 47 (49.5) | |
| Pack-years (ever-smokers)¶ | 25.08 (12.78–41.78) | 21.00 (10.25–35.52) | 33.75 (20.21–52.50) | 26.55 (15.22–45.98) | 47.30 (32.02–61.25) | 7.4 |
| Pack-year categories | 7.4 | |||||
| Never | 1520 (36.2) | 972 (41) | <40 (<15) | 192 (27) | <10 (<10) | |
| <10 pack-years | 532 (12.7) | 310 (13.1) | 18 (6.2) | 80 (11.3) | 6 (6.3) | |
| 10–20 pack-years | 549 (13.1) | 296 (12.5) | <40 (<20) | 98 (13.8) | <5 (<5) | |
| >20 pack-years | 1594 (38) | 659 (27.8) | 184 (63.7) | 305 (42.9) | 76 (80) | |
| Age started smoking (years) | 16.0 (15.0–19.0) | 17.0 (15.0–19.0) | 16.0 (14.0–18.0) | 16.0 (14.0–18.0) | 15.0 (14.0–17.0) | 35.8 |
| Alcohol consumption (units·week−1)+ | 5.0 (2.0–10.0) | 3.0 (1.0–6.0) | 3.0 (1.0–6.0) | 16.0 (14.0–18.0) | 15.0 (14.0–17.0) | 19.3 |
| Alcohol consumption (g·day−1) | 19.3 | |||||
| 0 | 477 (13.1) | 403 (17) | 53 (18.3) | 0 (0) | 0 (0) | |
| Low (<12 g·day−1) | 1948 (53.3) | 1648 (69.4) | 196 (67.8) | 0 (0) | 0 (0) | |
| Moderate (12–24 g·day−1) | 716 (19.6) | 322 (13.6) | 40 (13.8) | 283 (39.8) | 32 (33.7) | |
| High (25–40 g·day−1) | 352 (9.6) | 0 (0) | 0 (0) | 302 (42.5) | 39 (41.1) | |
| Very high (>40 g·day−1) | 162 (4.4) | 0 (0) | 0 (0) | 126 (17.7) | 24 (25.3) | |
| Frequency of drinking alcohol | 3.9 | |||||
| Never | 236 (5.4) | 118 (5) | 20 (6.9) | <10 (<5) | <5 (<5) | |
| <1 time per month | 482 (11.1) | 277 (11.7) | <40 (<20) | <5 (<5) | 0 (0) | |
| 1–3 times per month | 799 (18.4) | 490 (20.6) | 58 (20.1) | <20 (<5) | <5 (<5) | |
| 1 time per week | 702 (16.1) | 489 (20.6) | 46 (15.9) | <25 (<5) | <5 (<5) | |
| 2–4 times per week | 1152 (26.5) | 672 (28.3) | 71 (24.6) | 218 (30.7) | 26 (27.4) | |
| 5–6 times per week | 398 (9.1) | 123 (5.2) | 25 (8.7) | 173 (24.3) | 16 (16.8) | |
| Every day | 582 (13.4) | 139 (5.9) | 23 (8) | 256 (36) | 44 (46.3) | |
| Total drink-years | 29.12 (14.15–56.57) | 20.46 (10.45–34.34) | 23.73 (12.93–38.01) | 75.84 (57.18–104.55) | 88.81 (61.28–119.88) | 29.2 |
| Total drink-years categories | 29.2 | |||||
| Never | 127 (4) | 111 (4.7) | 12 (4.2) | 0 (0) | 0 (0) | |
| Low (1–50) | 2151 (67.1) | 1723 (72.6) | 190 (65.7) | 125 (17.6) | 7 (7.4) | |
| Moderate (51–100) | 655 (20.4) | 177 (7.5) | 30 (10.4) | 365 (51.3) | 44 (46.3) | |
| High (>100) | 274 (8.5) | <40 (<5) | <5 (<5) | 183 (25.7) | 37 (38.9) | |
| Diabetes | 283 (6.2) | 149 (6.3) | <20 (<5) | 39 (5.5) | <5 (<5) | 0.0 |
| Ischaemic heart disease | 519 (11.5) | 212 (8.9) | 52 (18) | 74 (10.4) | 14 (14.7) | 0.0 |
| Hypertension | 1086 (24) | 529 (22.3) | 85 (29.4) | 147 (20.7) | 17 (17.9) | 0.0 |
| Any malignancy | 264 (5.8) | 128 (5.4) | 20 (6.9) | 45 (6.3) | 6 (6.3) | 0.0 |
| Breast cancer | 44 (1.0) | 30 (1.3) | <5 (<5) | <5 (<5) | <5 (<5) | |
| Atrial fibrillation/flutter | 250 (5.5) | 119 (5.0) | 22 (7.6) | 26 (3.7) | 9 (9.5) | 0.0 |
| Cerebrovascular disease | 265 (5.9) | 115 (4.8) | 24 (8.3) | 37 (5.2) | 10 (10.5) | 0.0 |
| Gastro-oesophageal reflux | 119 (2.6) | 59 (2.5) | 11 (3.8) | <20 (<5) | <5 (<5) | 0.0 |
| Heart failure | 137 (3.0) | 83 (3.5) | 17 (5.9) | 21 (3.0) | 5 (5.3) | 0.0 |
| Myocardial infarction | 182 (4.0) | 83 (3.5) | 16 (5.5) | 24 (3.4) | <5 (<5) | 0.0 |
| Peptic ulcer | 141 (3.1) | 59 (2.5) | 14 (4.8) | 16 (2.3) | 6 (6.3) | 0.0 |
| Peripheral vascular disease | 137 (3.0) | 50 (2.1) | 16 (5.5) | 20 (2.8) | 8 (8.4) | 0.0 |
| Rheumatic disease | 96 (1.9) | 58 (2.4) | 9 (3.1) | <10 (<5) | <5 (<5) | 0.0 |
| Asthma | 451 (10.1) | 168 (7.1) | 88 (30.9) | 39 (5.5) | 25 (27.5) | 1.4 |
| Former drinker | 126 (3.7) | 105 (4.4) | 16 (5.5) | 0 (0) | 0 (0) | 24.2 |
| Abstainer | 127 (3.7) | 111 (4.7) | 12 (4.2) | 0 (0) | 0 (0) | 24.2 |
| Pre-bronchodilator FEV1 (L) | 2.71±0.84 | 2.85±0.77 | 1.87±0.77 | 3.16±0.73 | 2.02±0.75 | 0.4 |
| FEV1 % pred | 89.83±20.70 | 94.64±16.99 | 62.61±21.82 | 96.52±15.81 | 63.24±20.29 | 0.5 |
Continuous variables are presented as n, mean±sd or median (interquartile range); categorical variables are presented as n or n (%) (calculated as n/N, where N is the number with non-missing data in that column). Participants were included if both COPD status and alcohol category were observed; other characteristics are summarised using available data within each stratum. BMI: body mass index; FEV1: forced expiratory volume in 1 s. #: n=1060 had missing on COPD or alcohol; ¶: pack-years calculated as (cigarettes·day−1/20)×years smoked; +: alcohol consumption reported in Danish units (1 unit=12 g ethanol). Additional comorbidities with ≤40 prevalent cases are presented in supplementary table S1.
Primary outcome
All-cause mortality was the primary outcome. Participants were followed from their individual date of baseline assessment (2004–2006), at which alcohol consumption and covariates were recorded, until death or 31 December 2014, whichever occurred first. Mortality data were obtained from the Danish Cause of Death Registry [30].
Secondary outcomes
Secondary outcomes were baseline comorbidities, defined as binary variables using registry diagnoses recorded prior to the individual baseline date. We examined whether the prevalence of these comorbidities differed across four COPD×alcohol groups (no COPD/low alcohol, COPD/low alcohol, no COPD/high alcohol and COPD/high alcohol). These analyses were cross-sectional at baseline.
Statistical analyses
Descriptive analyses
Baseline characteristics were summarised as mean with standard deviation, median (interquartile range (IQR)) or count (percentage). Participants with complete data on alcohol and COPD were stratified into the four groups: no COPD/low alcohol, COPD/low alcohol, no COPD/high alcohol and COPD/high alcohol.
Exposures and covariates
Primary exposures were alcohol intake (≤10 versus >10 units·week−1) and COPD (yes/no). Covariates (selected a priori) included age, sex, smoking status, educational attainment, body mass index (BMI) and 13 comorbidities. Age and BMI were modelled with natural cubic splines (knots at the 5th, 35th, 65th and 95th percentiles).
Alcohol intake was additionally modelled as a continuous variable (g·day−1). To account for potential differences between abstainers and drinkers, analyses included a binary indicator for abstinence together with continuous intake among drinkers. Extreme values were truncated at the 99th percentile to reduce undue influence of outliers.
Non-linearity was assessed using restricted cubic splines with knots placed at prespecified percentiles of intake among drinkers. Model fit was compared with the corresponding linear model using likelihood ratio tests. For visualisation, predicted hazard ratios were plotted across the observed range of intake among drinkers, using 10 g·day−1 as the reference value (supplementary figure S1).
Survival analyses
Kaplan–Meier curves illustrated unadjusted survival across COPD×alcohol groups (≤10 and ≤15 units·week−1 thresholds) compared with log-rank tests. Survival probabilities with 95% confidence intervals were reported at 9 years.
Cox regression analyses
Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals for all-cause mortality. Model 1 included alcohol consumption and COPD status. Model 2 additionally adjusted for age, sex, smoking status, educational level and BMI. Model 3 further adjusted for 13 baseline comorbidities. Proportional hazards assumptions were assessed using Schoenfeld residuals. Interactions between alcohol and COPD, sex and smoking status were evaluated using likelihood ratio tests. Because baseline comorbidities may represent shared risk factors or consequences of COPD, Model 3 should be interpreted as an analysis adjusted for measured comorbidity burden rather than a formal mediation analysis.
Sensitivity analyses
Robustness was examined by excluding former drinkers, participants with >20 pack-years smoking history and asthma without COPD, and by modelling alcohol as g·day−1 and drink-years.
In a sensitivity analysis restricted to participants with COPD, disease severity was classified according to post-bronchodilator FEV1 % predicted and categorised as GOLD 1–2 (FEV1 ≥50%) and GOLD 3–4 (FEV1 <50%) according to the GOLD criteria [28]. COPD severity was included as an additional covariate in the fully adjusted Cox regression model.
Inverse probability of treatment weighting
Inverse probability of treatment weighting (IPTW) was used to address potential confounding. Propensity scores for alcohol consumption >10 units·week−1 were estimated using COPD status, age (restricted cubic spline), sex, smoking status, BMI (restricted cubic spline), education and 13 baseline comorbidities. Participants were weighted by the inverse probability of their observed alcohol exposure, generating a pseudo-population in which measured covariates were balanced across alcohol groups. Stabilised average treatment effect weights were truncated at the 1st and 99th percentiles to reduce the influence of extreme weights. Covariate balance before and after weighting was assessed using standardised mean differences (supplementary figure S2). Weighted Cox proportional hazards models with robust standard errors were fitted, additionally adjusting for covariates to reduce residual confounding.
Secondary outcome analyses
Age- and sex-adjusted comorbidity prevalences were estimated across COPD×alcohol groups using marginal predicted probabilities from logistic models. Associations were estimated as odds ratios with 95% confidence intervals, adjusted for age, sex, smoking, BMI and education. False discovery rate (FDR) correction using the Benjamini–Hochberg procedure addressed multiple testing across 13 comorbidities.
Missing data
Alcohol consumption and spirometry data were unavailable for a subset of participants. Individuals with incomplete baseline data were older and had higher morbidity and mortality compared with those included in the analytic cohort, indicating that data were Missing Not At Random (MNAR).
As alcohol exposure variables, including former drinker status, were entirely unavailable among non-responders, multiple imputation would require unverifiable assumptions regarding the missing data mechanism. We therefore conducted complete-case analyses. Kaplan–Meier analyses included participants with complete alcohol and COPD data, and multivariable Cox models were restricted to individuals with complete covariate information.
Results
Study population
Of 4757 sampled individuals, 4528 were eligible (229 excluded due to missing civil registration number). Of these, 1060 (23.4%) had missing alcohol and/or COPD data. These individuals were older, and had higher comorbidity burden and substantially higher mortality rates compared to those with complete data (supplementary tables S3–S6). Among the participants with repeated alcohol assessments, consumption remained stable over time. Between 2004–2006 and 2008–2009, 83% (n=1791) did not change category, and between 2004–2006 and 2016 the corresponding proportion was 85% (n=412).
Baseline characteristics overall and stratified
Baseline characteristics are presented in table 1. At baseline, the cohort comprised 4528 participants with mean age 64 years; 65% were male. Of these, 3468 (76.6%) had complete data on both alcohol consumption and COPD status and were included in analytical models. These participants were stratified into four groups by COPD and alcohol status. Among the analytical cohort, COPD prevalence was 12% and 22% consumed >10 units·week−1 alcohol. Mean predicted FEV1 was 90% overall and 63% among COPD participants. Most participants were classified as non-COPD/low alcohol consumers (n=2373 (68%)). COPD participants were older, more often male and had higher smoking exposure compared to non-COPD participants. High alcohol consumers were younger and predominantly male.
Kaplan–Meier survival
Of the 4528 eligible participants, 3655 (81%) had complete alcohol data and 4291 (95%) had COPD status determined. Among 3468 participants with complete data on both exposures, 593 deaths occurred over 10 years (17% overall mortality). Survival was stratified by COPD status and alcohol consumption (figure 1). COPD was the dominant predictor of mortality throughout 10-year follow-up, while alcohol consumption showed no association with survival within either COPD stratum. Mortality was substantially higher among participants with missing alcohol and/or COPD data (supplementary table S4).
FIGURE 1.
Kaplan–Meier survival curves stratified by COPD status and alcohol use: no COPD/low alcohol, COPD/low alcohol, no COPD/high alcohol and COPD/high alcohol. Stratification using ≤10 versus >10 units·week−1 threshold. Numbers at risk shown at 0, 3, 6 and 9 years. Shaded areas represent 95% confidence intervals.
At 9 years, survival in the non-COPD group was 86.8% (95% CI 85.5–88.2%) for low alcohol consumers and 87.9% (95% CI 85.5–90.3%) for high alcohol consumers. In the COPD group, 9-year survival was 60.2% (95% CI 54.8–66.1%) for low alcohol consumers and 62.1% (95% CI 53.0–72.6%) for high alcohol consumers. The substantial mortality gap between COPD strata (∼26 percentage points) contrasted sharply with negligible differences between alcohol categories within each stratum (≤2 percentage points).
Survival curves diverged primarily by COPD status, with more rapid early mortality accumulation among COPD participants, although separation between strata persisted throughout follow-up. When examined independently, COPD was strongly associated with mortality (log-rank p<0.001), whereas alcohol consumption (≤10 versus >10 units·week−1) showed no association (log-rank p=0.61). Overall log-rank test across all four groups was highly significant (p<0.001), driven entirely by COPD status differences. Results were consistent using the 15 units·week−1 threshold (figure 2).
FIGURE 2.
Kaplan–Meier survival curves stratified by COPD status and alcohol use: no COPD/low alcohol, COPD/low alcohol, no COPD/high alcohol and COPD/high alcohol. Stratification using ≤15 versus >15 units·week−1 threshold. Numbers at risk shown at 0, 3, 6 and 9 years. Shaded areas represent 95% confidence intervals.
Cox regression models
Among 3468 participants with complete COPD and alcohol data, 3421 had complete covariate information and were included in the multivariable Cox models.
Results from Cox proportional hazards models are presented in table 2. In unadjusted models (Model 1), alcohol showed no association with mortality while COPD showed strong association. After adjusting for age, sex, smoking, education and BMI (Model 2), COPD remained strongly associated while alcohol remained unassociated. Further adjustment for 13 baseline comorbidities (Model 3) yielded similar results; COPD's mortality association attenuated but remained substantial.
TABLE 2.
Hazard ratios and 95% confidence intervals for all-cause mortality from Cox proportional hazards models: complete-case analysis (n=3421)
| Model 1 | Model 2 | Model 3 | |
|---|---|---|---|
| Alcohol >10 units·week−1 | 0.97 (0.8–1.18) | 0.87 (0.71–1.06) | 0.87 (0.71–1.07) |
| COPD (yes versus no) | 3.65 (3.04–4.39)* | 2.08 (1.72–2.52)* | 2.00 (1.64–2.43)* |
Model 1: unadjusted (alcohol and COPD only). Model 2: adjusted for age, sex, smoking status, education and body mass index (age and body mass index were modelled using restricted cubic splines with knots at the 5th, 35th, 65th and 95th percentiles). Model 3: Model 2 plus 13 baseline comorbidities. *: p<0.05.
Several covariates in the fully adjusted model were statistically associated with mortality; however, the model was specified to estimate the association between alcohol, COPD and mortality rather than to evaluate the independent prognostic impact of individual comorbidities (figure 3). Proportional hazards assumptions were evaluated using Schoenfeld residuals. Alcohol and COPD satisfied proportional hazards assumptions. Minor deviations were observed for age and BMI; however, visual inspection indicated no substantial departure from proportionality, and the global test suggested only limited deviation. No interactions were detected between alcohol and COPD, sex or smoking status (supplementary table S7).
FIGURE 3.
Hazard ratios (HRs) for all-cause mortality from fully adjusted Cox proportional hazard model. Forest plot showing HRs with 95% confidence intervals from Model 3, including alcohol consumption (>10 versus ≤10 units·week−1), COPD status, demographics, lifestyle factors and 13 baseline comorbidities. Age and body mass index were modelled with restricted cubic splines. The vertical dashed line at HR 1.0 indicates no effect. Variables whose confidence intervals exclude 1.0 show statistically significant associations with mortality (p<0.05).
Sensitivity and weighting analyses
Results remained consistent in sensitivity analyses excluding former drinkers to address potential reverse causation (“sick quitter” bias), as well as after excluding heavy smokers or participants with asthma, and when using alternative alcohol exposure metrics (supplementary table S8).
In analyses restricted to participants with COPD, greater disease severity was associated with increased mortality. Inclusion of COPD severity did not materially alter the association between alcohol consumption and mortality (supplementary table S9).
Continuous modelling of alcohol intake did not demonstrate a significant association with mortality. In the fully adjusted model, alcohol intake expressed in g·day−1 was not associated with mortality risk. Restricted cubic spline analyses did not indicate evidence of non-linearity, with hazard ratios remaining close to unity across the observed range of intake (supplementary table S10 and supplementary figure S1).
Before weighting, several covariates demonstrated imbalance between alcohol groups; after weighting, all standardised mean differences were <0.1, indicating adequate covariate balance (supplementary figure S2). Weighted Kaplan–Meier curves showed similar survival patterns to unadjusted analyses (supplementary figure S3). IPTW yielded results consistent with the primary analyses (supplementary table S11).
Secondary analyses: comorbidity patterns
Several cardiovascular comorbidities were more prevalent in participants with COPD across alcohol strata, particularly ischaemic heart disease and cerebrovascular disease, whereas patterns for hypertension varied by alcohol group (figure 4 and supplementary table S12).
FIGURE 4.
Age- and sex-adjusted prevalence of individual comorbidities stratified by COPD status and alcohol use: no COPD/low alcohol, COPD/low alcohol, no COPD/high alcohol and COPD/high alcohol. Stratification using ≤10 versus >10 units·week−1 threshold. Prevalence estimates (%) with 95% confidence intervals derived from logistic regression models adjusted for age and sex. Error bars represent 95% confidence intervals.
In multivariable logistic regression examining associations between alcohol consumption and individual comorbidities, exploratory analyses showed nominal associations with atrial fibrillation and rheumatic disease. However, neither survived FDR correction across 13 comorbidities (all FDR-adjusted p>0.05) (supplementary table S13). Stratum-specific estimates are presented in figure 5 and supplementary table S14.
FIGURE 5.
Adjusted odds ratios for individual comorbidities by COPD status and alcohol use. Forest plots showing odds ratios (ORs) with 95% confidence intervals from multivariable logistic regression models adjusted for age, sex, smoking status, body mass index and education: COPD main effect (yes versus no), alcohol main effect (>10 versus ≤10 units·week−1) and COPD×alcohol interaction. The vertical dashed line at OR 1.0 indicates no effect.
Discussion
Main findings
In this 10-year prospective cohort, moderate-to-high alcohol consumption (>10 units·week−1) was not associated with all-cause mortality. In contrast, COPD was a strong independent predictor of death, with earlier separation of survival curves that persisted throughout follow-up. This association remained after adjustment for demographic factors, smoking and baseline comorbidities. We found no evidence that alcohol consumption modified the association between COPD and mortality.
In secondary analyses, several cardiovascular comorbidities were more prevalent among participants with COPD across alcohol strata. Alcohol intake was not consistently associated with individual comorbidities after correction for multiple testing.
Alcohol and mortality
The relationship between alcohol consumption and mortality remains debated. Earlier observational studies suggested protective effects of moderate intake, whereas more recent analyses have questioned this interpretation.
Apparent benefits of light-to-moderate drinking often attenuate after accounting for abstainer bias, particularly when former drinkers who stopped due to illness are excluded [31, 32]. Mendelian randomisation studies likewise do not support a protective effect at low intake levels [33, 34].
In our cohort, alcohol consumption was not associated with mortality across multiple modelling strategies. Exclusion of former drinkers to address potential reverse causation and use of alternative exposure definitions, including g·day−1, drink-years and spline-based modelling, yielded consistent results. Although some higher intake categories showed point estimates below unity, confidence intervals were wide and patterns were inconsistent across exposure definitions. These downward estimates may reflect residual confounding, healthy drinker selection or limited precision in smaller exposure strata rather than a true protective effect. Accordingly, they should be interpreted cautiously and not as evidence of benefit.
Evidence in COPD populations remains heterogeneous. European population-based cohorts have reported lowest mortality among light-to-moderate drinkers [19, 20], whereas hospital-based studies show clear harm, where alcohol use disorder is present in ∼10% of COPD admissions and associated with increased intensive care unit admissions and higher comorbidity burden [16]. In contrast, prospective COPD cohorts examining exacerbations have not identified consistent associations between baseline alcohol consumption and clinical outcomes, aligning with our findings [17, 18]. These differences likely reflect variation in exposure severity and population selection. Our cohort was predominantly community dwelling and relatively healthy (88% without COPD, 4.4% consuming >40 g·day−1), thus capturing primarily moderate consumption. At these levels, we observed no association with mortality. In contrast, hospital-based studies likely capture severe alcohol use with immunological impairment, malnutrition and reduced treatment adherence. Conversely, reports of apparent protection may reflect residual confounding by lifestyle factors such as diet, healthcare engagement or socioeconomic position. Cross-sectional findings of lower COPD prevalence among binge drinkers [35] may also reflect survival bias, as individuals with severe COPD are unlikely to tolerate heavy episodic drinking.
Importantly, the absence of an association in this cohort should not be interpreted as evidence that alcohol consumption is safe. Heavy alcohol use is well established as a contributor to morbidity and premature mortality. The limited representation of very high intake in this population restricts inference regarding harmful drinking patterns.
COPD, comorbidity and mortality
The strong association between COPD and mortality confirms COPD as a major determinant of long-term survival [28]. Survival curves diverged early and remained separated throughout follow-up. Although adjustment for demographic factors, smoking and comorbidities attenuated the hazard ratio, COPD remained independently associated with mortality, consistent with previous studies demonstrating that COPD confers excess risk beyond measured comorbidity burden [36].
In analyses restricted to participants with COPD, greater airflow limitation was associated with higher mortality, as expected. Adjustment for spirometric severity did not materially alter the alcohol estimate, suggesting that disease severity did not confound the alcohol–mortality association.
Several cardiovascular comorbidities were more prevalent among participants with COPD across alcohol strata, consistent with shared risk factors such as smoking and ageing and with systemic mechanisms linking COPD and cardiovascular disease [36–38]. However, adjustment for comorbidities only partially attenuated the association between COPD and mortality, indicating that additional disease-specific mechanisms including respiratory failure, infections and systemic consequences of chronic inflammation likely contribute to excess risk.
Alcohol consumption showed no consistent associations with individual comorbidities after correction for multiple testing. This suggests that comorbidity patterns in COPD are primarily driven by COPD itself and shared risk factors rather than alcohol intake at the levels observed in this cohort. Several comorbidities with established prognostic importance (e.g. ischaemic heart disease, heart failure and myocardial infarction) did not reach statistical significance in multivariable mortality models. This likely reflects relatively low baseline prevalence and overlap between cardiovascular conditions, leading to reduced precision in fully adjusted models rather than absence of clinical relevance.
Strengths and limitations
Strengths include the prospective design with 10-year follow-up, population-based sampling, comprehensive comorbidity ascertainment through validated national registries, post-bronchodilator spirometry performed according to international guidelines, and multiple analytical approaches including spline modelling and IPTW. Among participants with repeated assessments, alcohol consumption categories remained stable over time (83–85% unchanged over 4–12 years), supporting the use of baseline exposure for long-term risk estimation.
Several limitations merit consideration. First, alcohol consumption and COPD status were assessed at baseline only. Changes in drinking behaviour or incident COPD during follow-up were not captured, potentially leading to regression dilution and attenuation of associations. Although alcohol was reassessed at the 2008–2009 follow-up visit, a time-varying exposure analysis was not pursued: only 1791 participants had alcohol data at both time-points with 164 observed deaths, providing insufficient power for a fully adjusted model, and alcohol consumption was highly stable across assessments (83% unchanged), suggesting reclassification would be unlikely to materially alter the findings.
Alcohol intake was self-reported and may be affected by underreporting. The primary exposure was derived from quantitative units per week rather than drinking-frequency responses; a small number of participants (n=9) showed discordant responses between these items, explaining the presence of a few “never” responses within higher intake strata. Given the very low frequency, any impact on effect estimates is likely negligible. Dichotomisation at 10 units·week−1 may oversimplify dose–response relationships; however, continuous modelling and spline analyses yielded consistent results.
Second, although spirometric severity was available and included in sensitivity analyses, we lacked longitudinal measures of disease progression and exacerbation burden, which may further influence prognosis.
Third, asthma was defined using self-reported lifetime physician diagnosis rather than registry-confirmed or clinically validated asthma. This may capture historical or childhood asthma and likely contributes to the relatively high asthma prevalence observed among participants meeting spirometric COPD criteria, consistent with asthma–COPD overlap and potential misclassification related to applying a fixed FEV1/FVC threshold in older adults. Importantly, exclusion of participants reporting asthma did not materially alter the mortality estimates.
Fourth, residual confounding from lifestyle factors such as diet, physical activity and social engagement cannot be excluded.
Fifth, missing baseline data (23.4%) represents an important limitation. Participants with missing alcohol or COPD data were older and had substantially higher morbidity and mortality, indicating that missingness was unlikely to satisfy the Missing At Random (MAR) assumption. Alcohol exposure variables, including former drinker status, were entirely unavailable among non-responders. Under these conditions, multiple imputation would require strong, unverifiable assumptions regarding the joint distribution of exposure and outcome. We therefore performed complete-case analyses to avoid introducing model-based assumptions; however, the findings may not generalise to frailer populations in whom alcohol effects could differ.
Finally, although the cohort included 593 deaths, providing adequate power to detect moderate associations, smaller effect sizes cannot be excluded. The observed hazard ratio for alcohol consumption was close to unity, suggesting that any undetected effect, if present, is likely to be small. Also, generalisability may be limited to populations with similar demographic characteristics and predominantly moderate alcohol consumption.
Conclusions
Moderate alcohol consumption was not associated with mortality in this 10-year cohort. COPD conferred substantial excess mortality independent of comorbidities and lifestyle factors. These findings do not support claims that moderate alcohol consumption influences survival and emphasise the importance of early COPD diagnosis and comprehensive disease management.
Acknowledgement
The authors used ChatGPT (OpenAI; https://openai.com) for assistance with language editing in the original manuscript. The manuscript was critically reviewed and revised by the authors prior to submission.
Footnotes
Provenance: Submitted article, peer reviewed.
Ethics statement: The study was approved by the Regional Scientific Ethics Committee for Viborg and North Jutland Counties (N-20130063) and the Danish Data Protection Agency (2012-41-0775). Written informed consent was obtained from all participants.
Conflict of interest: L.B. Nielsen reports support from AstraZeneca to attend the ERS Congress in Amsterdam in 2025. S.J. Riddersholm, P.A. Jacobsen and J.G. Hansen have nothing to disclose. U.M. Weinreich reports grants or contracts from Fisher & Paykel Healthcare; payment or honoraria for lectures, presentations, speakers' bureaus, manuscript writing or educational events from AstraZeneca, Fisher & Paykel Healthcare, ResMed, Sanofi, Orion Pharma, GSK, Pfizer, Boehringer Ingelheim and Chiesi; support for attending meetings and/or travel from Sanofi; participation on a data safety monitoring or advisory board for AstraZeneca, GSK, Sanofi, O2Matic and Fisher & Paykel Healthcare; and leadership or fiduciary roles for the Danish Respiratory Society and the Danish Lung Association Research Fund, all in the past 36 months.
Support statement: This research received no specific grant from any funding agency.
Supplementary material
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
00335-2026.SUPPLEMENT
References
- 1.Global Initiative for Chronic Obstructive Lung Disease (GOLD) . Global Strategy for the Diagnosis, Management and Prevention of COPD. 2025. Available from: http://goldcopd.org/
- 2.Celli BR, Cote CG, Marin JM, et al. The body-mass index, airflow obstruction, dyspnea, and exercise capacity index in chronic obstructive pulmonary disease. N Engl J Med 2004; 350: 1005–1012. doi: 10.1056/NEJMoa021322 [DOI] [PubMed] [Google Scholar]
- 3.Divo M, Cote C, De Torres JP, et al. Comorbidities and risk of mortality in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2012; 186: 155–161. doi: 10.1164/rccm.201201-0034OC [DOI] [PubMed] [Google Scholar]
- 4.Calverley PMA, Anderson JA, Celli B, et al. Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease. N Engl J Med 2007; 356: 775–789. doi: 10.1056/NEJMoa063070 [DOI] [PubMed] [Google Scholar]
- 5.Rehm J, Mathers C, Popova S, et al. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. Lancet 2009; 373: 2223–2233. doi: 10.1016/S0140-6736(09)60746-7 [DOI] [PubMed] [Google Scholar]
- 6.World Health Organization . Global status report on alcohol and health and treatment of substance use disorders. 2024. www.who.int/publications/i/item/9789240096745 Date last accessed: 15 May 2026.
- 7.Wood AM, Kaptoge S, Butterworth AS, et al. Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599 912 current drinkers in 83 prospective studies. Lancet 2018; 391: 1513–1523. doi: 10.1016/S0140-6736(18)30134-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shield KD, Parry C, Rehm J. Chronic diseases and conditions related to alcohol use. Alcohol Res 2014; 35: 155–171. doi: 10.35946/arcr.v35.2.06 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Briasoulis A, Agarwal V, Messerli FH. Alcohol consumption and the risk of hypertension in men and women: a systematic review and meta-analysis. J Clin Hypertens 2012; 14: 792–798. doi: 10.1111/jch.12008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bagnardi V, Rota M, Botteri E, et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br J Cancer 2015; 112: 580–593. doi: 10.1038/bjc.2014.579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Boden JM, Fergusson DM. Alcohol and depression. Addiction 2011; 106: 906–914. doi: 10.1111/j.1360-0443.2010.03351.x [DOI] [PubMed] [Google Scholar]
- 12.Voskoboinik A, Prabhu S, Ling LH, et al. Alcohol and atrial fibrillation. J Am Coll Cardiol 2016; 68: 2567–2576. doi: 10.1016/j.jacc.2016.08.074 [DOI] [PubMed] [Google Scholar]
- 13.Simou E, Britton J, Leonardi-Bee J. Alcohol and the risk of pneumonia: a systematic review and meta-analysis. BMJ Open 2018; 8: e022344. doi: 10.1136/bmjopen-2018-022344 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Simet SM, Sisson JH. Alcohol's effects on lung health and immunity. Alcohol Res 2015; 37: 199–208. doi: 10.35946/arcr.v37.2.05 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mehta AJ, Guidot DM. Alcohol and the lung. Alcohol Res 2017; 38: 243–254. doi: 10.35946/arcr.v38.2.07 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Gómez-Garcia T, Jiménez-Garcia R, Hernández-Barrera V, et al. Nationwide trends and outcomes of alcohol use disorders in COPD hospitalizations in Spain, 2016–2023. J Clin Med 2025; 14: 6045. doi: 10.3390/jcm14176045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Greene CC, Bradley KA, Bryson CL, et al. The association between alcohol consumption and risk of COPD exacerbation in a veteran population. Chest 2008; 134: 761–767. doi: 10.1378/chest.07-3081 [DOI] [PubMed] [Google Scholar]
- 18.Wetherbee E, Niewoehner D, Sisson J, et al. Self-reported alcohol intake and risk of acute exacerbations of chronic obstructive pulmonary disease: a prospective cohort study. COPD 2015; 20: 1363–1370. doi: 10.2147/COPD.S86572 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tabak C, Smit HA, Räsänen L, et al. Alcohol consumption in relation to 20-year COPD mortality and pulmonary function in middle-aged men from three European countries. Epidemiology 2001; 12: 239–245. doi: 10.1097/00001648-200103000-00018 [DOI] [PubMed] [Google Scholar]
- 20.Kaluza J, Harris HR, Linden A, et al. Alcohol consumption and risk of chronic obstructive pulmonary disease: a prospective cohort study of men. Am J Epidemiol 2019; 188: 907–916. doi: 10.1093/aje/kwz020 [DOI] [PubMed] [Google Scholar]
- 21.Li Y, Yang H, Cao J. Association between alcohol consumption and cancers in the Chinese population – a systematic review and meta-analysis. PLoS One 2011; 6: e18776. doi: 10.1371/journal.pone.0018776 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hansen JG, Pedersen L, Overvad K, et al. The prevalence of chronic obstructive pulmonary disease among Danes aged 45–84 years: population-based study. COPD 2008; 5: 347–352. doi: 10.1080/15412550802522635 [DOI] [PubMed] [Google Scholar]
- 23.Stanojevic S, Kaminsky DA, Miller MR, et al. ERS/ATS technical standard on interpretive strategies for routine lung function tests. Eur Respir J 2022; 60: 2101499. doi: 10.1183/13993003.01499-2021 [DOI] [PubMed] [Google Scholar]
- 24.American Thoracic Society . Standardization of Spirometry, 1994 Update. Am J Respir Crit Care Med 1995; 152: 1107–1136. doi: 10.1164/ajrccm.152.3.7663792 [DOI] [PubMed] [Google Scholar]
- 25.Pleasants RA, Rivera MP, Tilley SL, et al. Both duration and pack-years of tobacco smoking should be used for clinical practice and research. Ann Am Thorac Soc 2020; 17: 804–806. doi: 10.1513/AnnalsATS.202002-133VP [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Schmidt M, Schmidt SAJ, Sandegaard JL, et al. The Danish National Patient Registry: a review of content, data quality, and research potential. Clin Epidemiol 2015; 7: 449–490. doi: 10.2147/CLEP.S91125 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Wallach Kildemoes H, Toft Sørensen H, Hallas J. The Danish National Prescription Registry. Scand J Public Health 2011; 39: 7 Suppl., 38–41. doi: 10.1177/1403494810394717 [DOI] [PubMed] [Google Scholar]
- 28.Singh D, Agusti A, Anzueto A, et al. Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease: the GOLD Science Committee Report 2019. Eur Respir J 2019; 53: 1900164. doi: 10.1183/13993003.00164-2019 [DOI] [PubMed] [Google Scholar]
- 29.Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987; 40: 373–383. doi: 10.1016/0021-9681(87)90171-8 [DOI] [PubMed] [Google Scholar]
- 30.Helweg-Larsen K. The Danish Register of Causes of Death. Scand J Public Health 2011; 39: 26–29. doi: 10.1177/1403494811399958 [DOI] [PubMed] [Google Scholar]
- 31.Zhao J, Stockwell T, Naimi T, et al. Association between daily alcohol intake and risk of all-cause mortality: a systematic review and meta-analyses. JAMA Netw Open 2023; 6: e236185. doi: 10.1001/jamanetworkopen.2023.6185 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Stockwell T, Zhao J, Panwar S, et al. Do “moderate” drinkers have reduced mortality risk? a systematic review and meta-analysis of alcohol consumption and all-cause mortality. J Stud Alcohol Drugs 2016; 77: 185–198. doi: 10.15288/jsad.2016.77.185 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Millwood IY, Walters RG, Mei XW, et al. Conventional and genetic evidence on alcohol and vascular disease aetiology: a prospective study of 500 000 men and women in China. Lancet 2019; 393: 1831–1842. doi: 10.1016/S0140-6736(18)31772-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Larsson SC, Burgess S, Mason AM, et al. Alcohol consumption and cardiovascular disease: a Mendelian randomization study. Circ Genom Precis Med 2020; 13: e002814. doi: 10.1161/CIRCGEN.119.002814 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ragireddy HR, Lama E, Adsure MR, et al. Alcohol intake and binge drinking and their association with chronic obstructive pulmonary disease, emphysema, and chronic bronchitis: a retrospective study using the Behavioral Risk Factor Surveillance System (BRFSS) database. Cureus 2025; 17: e90371. doi: 10.7759/cureus.90371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Barnes PJ, Celli BR. Systemic manifestations and comorbidities of COPD. Eur Respir J 2009; 33: 1165–1185. doi: 10.1183/09031936.00128008 [DOI] [PubMed] [Google Scholar]
- 37.Gan WQ. Association between chronic obstructive pulmonary disease and systemic inflammation: a systematic review and a meta-analysis. Thorax 2004; 59: 574–580. doi: 10.1136/thx.2003.019588 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Agustí AGN, Noguera A, Sauleda J, et al. Systemic effects of chronic obstructive pulmonary disease. Eur Respir J 2003; 21: 347–360. doi: 10.1183/09031936.03.00405703 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.
Supplementary material
00335-2026.SUPPLEMENT