Coronavirus disease 2019 (COVID-19) pandemic has revealed how readily viruses spread in our interconnected world. As of November 20, 2021, more than 257 million COVID-19 cases and 5.2 million deaths have been reported worldwide (https://www.worldometers.info/coronavirus/). During the pandemic, although epidemiological characteristics and risk factors (e.g., age, obesity, and lifestyle factors) of COVID-19 have been rapidly reported (1–5), the risk or protective factors for COVID-19 infection are largely unknown.

Adverse effects of alcohol consumption have been widely documented. The observed relationships between alcohol consumption and diseases are often non-linear, with low-to-moderate alcohol consumption being protective and heavy alcohol consumption being harmful (6, 7). In addition, evidence has suggested the protective effects of moderate alcohol consumption on cardiovascular events, such as coronary heart disease (8), myocardial infarction (9), and heart failure (10). Several cohort studies have pointed out that people who have light-to-moderate alcohol consumption survive longer than abstainers (11). However, the association of alcohol consumption with the risk of COVID-19 infection or pneumonia was inconsistent (12–15), and the associations of status, amount, and frequency of different alcoholic beverages with the risk of COVID-19 and associated mortality have not been systematically investigated. In the present large longitudinal observational study, we systematically investigated the associations of alcohol consumption (status, frequency, amount, and subtype of alcoholic beverages) with COVID-19 risk and associated mortality after adjusting for several confounders. We further examined the “dose” -response association between alcohol consumption and COVID-19 risk by drawing the risk trajectory of COVID-19 with different amounts of alcohol consumption.

In this prospective study of a large sized UK Biobank cohort, we documented seven novel findings. First, we identified some independent protective and risk factors for COVID-19. The protective factors included current alcohol drinker status, high frequency of alcohol consumption, red wine, white wine, and champagne, and 1–2 glasses/week fortified wine; the risk factors included alcohol consumption within and above the guidelines, beer and cider (regardless of amount), and spirits (≥5 glasses/week). Second, the protective effect of red wine for COVID-19 was significant regardless of the frequency of alcohol intake, but it only played the protective effect when subjects consumed alcohol above or double above the guidelines. Third, the consumption of beer and cider increased the risk of COVID-19, regardless of the frequency and amount (within or above the guidelines) of alcohol intake. Fourth, low frequency of consumption of fortified wine (1–2 glasses/week) had a protective effect when subjects consumed alcohol within the guidelines. Fifth, a high frequency of consumption of spirits (≥5 glasses/week) within the guidelines was associated with a higher risk of COVID-19, whereas a high frequency of consumption of white wine and champagne above the guidelines was associated with a lower risk of COVID-19. Sixth, the dose-response associations between the amount of alcohol consumption and the risk of COVID-19 showed an increased risk of COVID-19 with a greater number of alcohol consumption. Seventh, alcohol drinker status, frequency and amount of alcohol consumption, and all subtypes of alcoholic beverage were not associated with COVID-19 associated mortality.

The most striking and interesting finding for COVID-19 infection was the recommended intake of different alcoholic beverages. The protective association of low-to-moderate alcohol consumption with a lower risk of cataracts (17), a lower risk of myocardial infarction (6), and a better cognitive function (18) have been reported. In our study, the consumption of different alcoholic beverages was associated with different chances of COVID-19 infection. It is worth noting that red wine is recommended for all adults because it was associated with a lower risk of COVID-19 when subjects usually consumed alcohol above or double above the guidelines. The maximum recommended intake of fortified wine is 2 glasses/week within the guidelines, which may have chances to reduce the COVID-19 risk. The consumption of beer and cider are not recommended, regardless of frequency and amount of alcohol consumption. Furthermore, heavy drinking is not recommended for all alcoholic beverages. Therefore, the COVID-19 risk appears to vary across different alcoholic beverages, frequency and amount of alcohol consumption.

In addition, we evaluated the dose-response associations between the amount of alcohol consumption and the COVID-19 risk. We noted a non-linear J-shaped or linearity increase of the COVID-19 risk with the weekly amount of alcohol consumption across different substyles of alcoholic beverages. Studies have shown that long-term alcohol abuse and acute binge drinking are associated with immunosuppression and increased susceptibility to both bacterial and viral infections (19, 20). Therefore, heavy alcohol consumption was associated with a higher risk of COVID-19. However, in our dataset, only a few participants reported heavy alcohol consumption, which may be reflective of the healthier nature of study participants. The potential association of heavy alcohol consumption with COVID-19 and associated mortality should be further explored in a larger sample sized cohort (although ascertaining the prospective effect of heavy alcohol consumption would be challenging because heavy drinkers would be unlikely to volunteer for such studies).

Hamer et al. found a lower risk of COVID-19 for drinkers relative to non-drinkers if the amount of alcohol consumption was below the recommended guideline (3). Our results showed that some subtypes of alcoholic beverages (e.g., wine) themselves played protective roles against the COVID-19, whereas some subtypes increased the COVID-19 risk (e.g., beer and cider, and spirits). There are several possible explanations for these findings. Polyphenols, which are present in varying degrees in alcoholic beverages, have antioxidant properties, particularly in wines among which red wine has the highest concentrations of phenolic compounds (21). Studies have shown that wines exhibit beneficial properties which are independent of the presence of alcohol, and should be attributed to their polyphenolic contents (22, 23). Red wine provides additional benefits to other alcoholic beverages probably due to its higher polyphenolic content, by decreasing blood pressure, inhibiting the oxidation of low-density lipoprotein particles and other favorable effects on the cellular redox state, improving endothelial function, inhibiting platelet aggregation, reducing inflammation and cell adhesion, and activating proteins that prevent cell death (21). Polyphenols could inhabit the effects of several types of virus, such as Epstein-Barr virus (24, 25), enterovirus (26, 27), herpes simplex virus (HSV) (28), influenza virus (29), and other virus causing respiratory tract-related infections (30, 31). These findings support the notion of the strong beneficial properties of red wine against the COVID-19 risk.

Spirits had the highest alcohol concentration and the lowest polyphenolic concentration. We discovered a higher risk of spirits for COVID-19. Sanja Radonjić et al. have reported the differences between wine and beer in the presence and the concentrations of phenolic substances (32). They found that chalcones and flavanones were found only in beer, whereas other polyphenol classes (e.g., stilbenes, proanthocyanidins, and resveratrol) are found mainly in wine and champagne. These findings may suggest that the specific class of polyphenolic constituents may be responsible for the beneficial effect of alcoholic beverages on COVID-19 events, and not the alcohol concentration. The concentration of polyphenol may greatly vary in different beer productions because it is related to the specific ingredients (33, 34). The main raw materials used in the production of beer are malt and hops, which are also the main source of polyphenols. Previous studies have suggested that moderate consumption of beer has a positive effect on some biomarkers associated with human health (35, 36), which may be associated with the nutrient effect of the polyphenols in the beer. However, the beer was identified as an independent risk factor for COVID-19 in our study. In our study, the ingredients and concentration of polyphenol of alcoholic beverage subtypes were not considered, and a study with more details about the ingredients and concentration of polyphenol is needed in the future.

Severe acute respiratory syndrome coronavirus (SARS-CoV)-2 not only attacks the respiratory system but also the cardiovascular system (37). The SARS-CoV-2-induced endothelial disruption and vascular thrombosis in the lungs may be the major pathological process of COVID-19 (38, 39). A meta-analysis involving 63 studies has indicated that moderate alcohol consumption had beneficial effects on the cardiovascular system (35). Also, the peri-infarct inflammatory infiltration following myocardial infarction was positively modulated by white wine (40). We speculated that the mechanisms responsible for the beneficial effects of moderate alcohol consumption on the cardiovascular system against the COVID-19 could be explained by increased changes in the plasma antioxidant activity (41) and reductions in the level of low-density lipoprotein (LDL) cholesterol (42–44).

The major strengths of this study are the prospective, large United Kingdom population-based cohort, dose-response associations of alcohol consumption with COVID-19 risk, and a focus on the subgroup analyses for different alcoholic beverages. However, there are several limitations that should be addressed. First, subjects in the UK Biobank have a restricted age range, and therefore our data could not represent the whole population (2, 5). Second, alcohol consumption was measured at baseline, and we did not know about potential changes during the COVID-19 pandemic. However, our study demonstrated that long-term drinking habits were associated with COVID-19 risk and associated mortality, and therefore our results will not largely be affected by the short-term changes of drinking behavior by the pandemic and associated lockdowns. Third, recruiting heavy drinkers to test different alcoholic beverages for dose–response analyses is difficult. Fourth, whether these findings are suitable for the young population and Asian are unknown. Fifth, the ingredients and the concentration of polyphenol of alcoholic beverage subtypes were not considered.

In conclusion, our study evaluated specific protective and risk factors across different alcoholic beverage subtypes for COVID-19 and found that the COVID-19 risk appears to vary across different subtypes, amounts, and frequency of alcoholic beverages. Our study suggests that subjects who usually consumed red wine and white wine and champagne above guidelines, and sometimes consumed 1–2 glasses/week fortified within the guidelines appear to have chances to reduce the risk of COVID-19. The consumption of beer and cider are not recommended regardless of frequency and amount of alcohol consumption, which increased the risk of COVID-19. Furthermore, heavy drinking is not recommended for all alcoholic beverages. Public health guidance should focus on reducing the risk of COVID-19 by advocating healthy lifestyle habits and preferential policies among consumers of beer and cider and spirits.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the North West Multi-Centre Research Ethics Committee (REC reference: 16/NW/0274). The patients/participants provided their written informed consent to participate in this study.

X-jD and YW had the idea for and designed this study, had full access to all the data in this study, take responsibility for the integrity of the data and the accuracy of the data analysis, critically revised the manuscript for important intellectual content, and gave final approval for the version to be published. X-jD and LT drafted the manuscript and did the analysis. YW, LR, YS, and WT take the responsibility for double check of the data analysis. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately resolved.

This research has been conducted using the UK Biobank Resource under Application Number 75732. This work was supported by the Guangdong Basic and Applied Basic Research Foundation (Grant No. 2020A1515011469), and the Sanming Project of Medicine in Shenzhen (Grant No. SZSM201812052).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.