The Dilemma of Moderate Alcohol Consumption and Type 2 Diabetes Risk: Drinking Frequency Versus Quantity Free

Diabetes is a major global health challenge, and type 2 diabetes (T2D) accounts for the highest proportion of cases. T2D is intricately linked to lifestyle factors, including diet, alcohol consumption, and physical activity; there is a critical need for understanding of these relationships to develop effective prevention strategies. In this issue of Diabetes Care, Li et al. (1) examine the association of alcohol drinking patterns and risk of T2D in three prospective cohorts of both women and men.

Diabetes significantly contributes to morbidity and mortality, especially as a key risk factor for ischemic heart disease and stroke (2,3). In 2021, ∼529 million people worldwide were living with diabetes, with T2D accounting for 96.0% of these cases. Projections indicate that its prevalence will increase by 61.2%–9.5% by 2050, affecting >1.27 billion people (4).

Lifestyle factors, including diet, physical activity, and alcohol consumption, have been extensively studied in relation to T2D risk, with unhealthy dietary patterns and reduced physical activity levels identified as key contributors to its development (5,6). Although excessive alcohol intake is clearly harmful, findings on associations between light-to-moderate alcohol consumption and T2D risk remain inconclusive and findings are controversial (7–9). For instance, a meta-analysis of 26 studies predominantly from Western countries indicated the lowest T2D risk to be among light and moderate drinkers (10). In contrast, a meta-analysis among Asian men suggested that light-to-moderate drinking was not uniformly related to lower T2D risk (11). In 2023, an updated meta-analysis of 55 cohort studies further refined these findings, showing that low-to-moderate alcohol use (<50 g per day) is related to lower T2D risk among women in particular, especially those with BMI ≥25 kg/m² (12). These discrepancies may also stem from variations in drinking patterns and frequency across different study populations, underscoring the need for a deeper exploration of drinking patterns to fully understand their impact on T2D risk.

However, the association between drinking patterns and risk of T2D has not been adequately investigated, with inconsistent findings from the only three published studies. In a study among male health professionals, drinking frequency was inversely associated with T2D risk in the case of moderate drinking patterns in men, even if the level of consumption per drinking day was low (13). Similarly, investigators in two studies of Japanese cohorts examined the impact of drinking frequency and quantity but reached conflicting conclusions. The Kansai Healthcare Study suggests that more frequent alcohol consumption the lower the risk of T2D among Japanese men (14), and the other study indicates that the usual alcohol amount consumed per drinking occasion is a more critical determinant than weekly drinking frequency in relation to risk of T2D among Japanese men (15). A notable gap in these studies is their exclusive focus on men, which may limit the generalizability of findings to broader demographic groups.

In looking at three well-designed prospective cohort studies, Li et al. (1) explore how alcohol drinking patterns are related to T2D risk in both women and men. The study included analysis of data from 200,969 participants across three cohorts, the Nurses’ Health Study (NHS), Nurses’ Health Study II (NHSII), and Health Professionals Follow-up Study (HPFS), with a follow-up period spanning up to three decades. A total of 20,551 incident T2D cases were identified. Cox proportional hazards models were used to estimate hazard ratios (HRs) and CIs, with adjustment for numerous potential confounders, including socioeconomic status, dietary habits, and physical activity. The main findings of total alcohol intake analyses revealed that alcohol consumption was associated with a lower risk of T2D in comparisons with risk for nondrinkers or those in the lowest intake category (0.1–4.9 g per day), for both men and women. This inverse association remained consistent across various sensitivity analyses, performed to address reverse causality and residual confounding, including extended latency periods, alternative modeling of exposure, and restrictions on symptomatic cases. Notably, the frequency of drinking was significantly related to T2D risk after additional adjustment for the total amount of alcohol consumed per drinking day (Fig. 1). The lower risk of T2D in association with alcohol consumption began at 1–2 drinking days per week for women and 3–4 days per week for men, with the strongest associations observed among individuals who drank >5 days per week, regardless of whether intake was <10 or ≥30 g per drinking day (HR ∼0.73). These findings underscore the complex interplay of drinking frequency, drinking quantity, and risk of developing T2D, emphasizing the need to consider both factors in evaluating the effects of alcohol consumption on metabolic health.

The study boasts several strengths. By leveraging data from three well-established cohorts including both men and women, this is the first study on both drinking frequency and drinking quantity in association with T2D among women. The long follow-up period and longitudinal data on alcohol consumption allow for a more comprehensive evaluation of associations of long-term alcohol consumption with T2D risk. Additionally, the authors adjusted for multiple lifestyle and socioeconomic confounders, thereby strengthening the validity of their conclusions. Furthermore, the study explores alcohol consumption considering both quantity and frequency in association with T2D risk along with an assessment of their joint association with T2D risk, confirming that drinking frequency might be the primary factor that account for the lower risk of T2D in association with moderate alcohol consumption.

The study faces several limitations. Due to the observational nature of the study, potential residual confounding from unmeasured lifestyle and genetic factors cannot be excluded despite extensive adjustments. Future research should include well-designed randomized controlled trials to confirm the causal relationships suggested by observational data and to precisely evaluate the impact of specific drinking patterns on T2D risk. The ongoing clinical trial (Advice of Moderate Drinking Pattern Versus Advice on Abstention on Major Disease and Mortality [UNATI] [University of Navarra Alumni Trialist Initiative]) on a moderate drinking pattern in Spain may provide timely insightful data (clinical trial reg. no. NCT06338215, ClinicalTrials.gov) (16). Moreover, the predominantly White study population limits the generalizability of the findings to other racial and ethnic groups. Notably, individuals of Asian descent frequently possess a genetic variant of the aldehyde dehydrogenase 2 (ALDH2) enzyme, leading to impaired alcohol metabolism and increased alcohol sensitivity (17). Future studies are warranted with inclusion of more ethnically diverse populations and with the examination of interactions of genetic variants and drinking pattern and quantity to improve external validity and ensure broader applicability of the conclusions.

In summary, the study published in this issue of Diabetes Care, among both U.S. men and women, demonstrates the importance of examining both frequency and quantity in their associations with T2D risk. Although findings from the study suggest that regular and moderate alcohol consumption was related to lower risk of T2D, caution is warranted due to the well-documented adverse effects of heavy alcohol consumption on liver function, cardiovascular health, and the risk of cancer.