Global burden and trends of high alcohol use-related injuries from 1990 to 2030: a comprehensive assessment of self-harm and interpersonal violence, transport injuries, and unintentional injuries using global burden of disease 2021

Yuan, Q., Chen, Z., Sun, B., Zheng, C., Kang, Y., Lou, Y., Zheng, M.Front. Public Health 13:1675607. https://doi.org/10.3389/fpubh.2025.1675607.
Abstract
Background: High alcohol use (HAU) is a major global public health concern, contributing to injuries such as Self-harm and interpersonal violence (SIV), Transport injuries (TI), and Unintentional injuries (UII). However, comprehensive global assessments of HAU-related injury burden remain limited.
Methods: Using data from Global Burden of Disease (GBD) 2021, we estimated HAU-related mortality and disability-adjusted life years (DALYs) from 1990 to 2021 across 204 countries and regions. We analyzed trends in Age-Standardized Disability-Adjusted Life Years Rate (ASDR) and Age-Standardized Mortality Rate (ASMR), examined Socio-demographic index (SDI) disparities, and employed age-period-cohort (APC) and Bayesian APC (BAPC) models for future projections. Frontier analysis identified countries with the greatest potential for burden reduction.
Results: Despite the decline in ASDR and ASMR in overall global injuries, Low-middle SDI regions continue to experience increasing SIV and TI burdens (ASDR rose from 57.25 to 70.55; 25.08 to 30.8 respectively), while UII remains high in High-middle and High SDI countries (The ASDR were 57.94 and 59.12 respectively). Young adults and the elderly bear the greatest burden. BAPC projections indicate that China, India, and several high-burden nations will see further increases in DALYs and ASDR by 2030, highlighting the need for urgent interventions.
Conclusions: Targeted policy measures, such as raising the legal drinking age, strengthening alcohol control for young people in Low and Low-middle-SDI regions, and enhancing older adults healthcare services in High-SDI regions, are essential to mitigate HAU-related injuries. Evidence-based, SDI-adapted strategies can significantly reduce this burden.
ISFAR Summary
This paper by Yuan et al. (2025) demonstrates that high alcohol consumption is a public health issue contributing to injuries. Yuan et al. (2025) utilise the latest Global Burden of Disease (GBD 2021) data to estimate all major types of alcohol-related injuries within a single global assessment spanning 1990 to 2030. They show declining age-standardised injury rates worldwide but increasing absolute disability-adjusted life years and deaths driven by population growth and ageing. Their concluding policy recommendations, however, go beyond what the data directly show, while the projections are presented with greater confidence than the model’s uncertainty intervals suggest.
The paper assumes that injuries related to high alcohol use are indeed caused by high alcohol consumption. Furthermore, high alcohol use appears to be defined as a statistical measure that sums up all known risks, mainly ignoring health benefits and other positive effects that individuals may experience from drinking alcohol. Additionally, there is no standardised procedure for collecting injury data across different countries.
While excessive alcohol consumption is harmful and needs to be addressed, these data may not provide a reliable basis for the policy measures recommended to prevent alcohol-related injuries in all societies, where drinking cultures vary greatly.
ISFAR Critique
Background
It is well-known that alcohol abuse and misuse cause harm . Alcohol-related harms include diseases and injuries, and most come from heavy episodic or heavy continuous alcohol consumption. Injuries include ‘self-harm and interpersonal violence, ‘transport injuries’, and ‘unintentional injuries’, which are all associated with heavy or ‘high’ alcohol use, hereafter referred to as HAU-related injury burden. Global annual mortality is estimated at 60-68 million per year. Intraindividual Variability (IIV) refers to fluctuations in a person’s cognitive or motor performance over time, and increased IIV is a robust marker of an increased mortality risk.

Based on 2021 Global Burden of Disease (GBD) data, it was estimated that IIV causes approximately 5 million deaths annually, with the World Health Organization reporting 4.4 million in 2025 . This means that IIV-related mortality represents a significant portion of all deaths, accounting for nearly 8% of all deaths globally. In particular, young people (aged 5-29 years) are impacted by IIV-related mortality. Road traffic injuries, homicide, and suicide are among the leading causes of mortality in that age group. Furthermore, low- and middle-income countries account for nearly 90% of these IIV-related deaths. Strategies to reduce high alcohol use and alcohol-related harm should, therefore, be an integral part of public health strategies (Kilian et al. 2024).
Yuan et al. (2025) assessed HAU-related injury burden across 204 countries and regions from 1990 to 2021. They also calculated age-standardised rates and estimated the annual percentage change in these rates. Additionally, they examined the relationship between HAU-related injury burden and various other parameters. These include the socio-demographic development index (SDI), age-period-cohort analysis, decomposition analysis, cross-country inequality analysis, and frontier analysis. They then predicted Disability-Adjusted Life Years (DALYs), the age-standardised disability-adjusted life years rate (ASDR), and the age-standardised mortality rate (ASMR) for HAU-related injury burden from 2022 to 2030. Their findings are, however, discussed in relation to policy measures that could be implemented to reduce the HAU-related injury burden.
This study aims to address the absence of a comprehensive, long-term global analysis that covers the entire spectrum of HAU-related injuries and their trends across diverse socio-demographic contexts.
Critique
The results of Yuan et al. (2025) show a global decline in ASDR and ASMR for HAU-related injuries, indicating that mortality and morbidity associated with HAU are declining. ASDR and ASMR are among the most important parameters to consider, as they are indicators of mortality and morbidity corrected for changes in population growth and composition, such as ageing. Additionally, the Estimated Annual Percentage Change (EAPC) in ASDR was negative for most outcomes, with few exceptions; EAPC increased in just four out of 24 outcomes listed, specifically injuries, self-harm, interpersonal violence, and transport injuries in low-middle socio-demographic index countries, as well as for self-harm and interpersonal violence in low socio-demographic index countries. This indicates that the burden of HAU-related injuries per a given population is decreasing, while the overall global HAU-related burden is rising mainly due to population growth.
Furthermore, Yuan et al. (2025) present global projections based on Bayesian statistics that suggest an increasing trend for specific injuries in only a few countries. Moreover, these trends are portrayed as more certain than they truly are, since Bayesian projections do not account for unforeseen future changes. Additionally, the global projections include wide confidence intervals, which increase their uncertainty.
Yuan et al. (2025), however, emphasise the need for urgent interventions in the results section of their paper’s abstract. While HAU-related injuries need to be minimised, their data suggest that HAU-related injuries are currently decreasing and may decrease further in the future, indicating less urgency than initially proposed. Fortunately, the authors themselves note that one limitation of their study is that the predictive methods extend historical trends into the future but cannot account for unforeseen shocks, such as sudden political changes or emerging health threats.
In addition, in the conclusion section of their paper’s abstract, the authors advocate targeted policy measures, including increasing the legal drinking age, strengthening alcohol control for young people in low- and low-middle-SDI regions, and improving healthcare services for older adults in high socio-demographic index regions. Although some of these policies may assist in preventing specific HAU-related injuries, the paper only reports on the incidences and projections of such injuries rather than the effectiveness of the recommended alcohol control measures. The data reveal correlations in burden, not the effectiveness of policy interventions. As a result, while these conclusions might seem reasonable as suggestions, they have not been tested within the study. Therefore, these statements move from descriptive epidemiology to prescriptive policy without supporting intervention data.
The authors also do not explicitly define how HAU is characterised in this paper. HAU is a term that can be understood in multiple ways. It is often regarded as comparable to heavy drinking, which the American Institute on Alcohol Abuse and Alcoholism (NIAAA) defines as consuming five or more drinks on any day or 15 or more per week for men, and four or more drinks on any day or eight or more per week for women. The Substance Abuse and Mental Health Services Administration (SAMHSA), an agency of the U.S. Department of Health and Human Services, describes heavy alcohol use as binge drinking on five or more days in the past month. Others have described HAU as regular alcohol intake of ≥3 drinks per day. A meta-analysis (Høye & Storesund Hesjevoll, 2023) demonstrated that the risk of adverse safety outcomes increases with rising blood alcohol concentration (BAC) levels, and the effects of higher BAC are significantly more severe for serious crashes than for other types. The relationship between BAC level and outcome was roughly exponential, akin to the findings of Blomberg et al. (2009).
HAU is defined in the GBD studies as alcohol consumption exceeding the Theoretical Minimum Risk Exposure Level (TMREL), which is the level of alcohol consumption where all-cause risk is minimised . Before the GBD 2020, this risk factor was just “Alcohol use” and measured the burden of alcohol consumption across the entire exposure range. Further reading of the GBD 2020 methods report indicates that the GBD study does not use a single TMREL but instead employs multiple TMRELs for alcohol consumption, which vary by age, sex, and region.
The TMREL is calculated based on modelling that includes various corrections such as tourist consumption in a region, unrecorded alcohol consumption and estimates for current drinkers in specific groups. This results in a table that specifies TMRELs by age group, sex, and world region. For example, the TMREL for women aged 20-24 years was 0 standard drinks (10 g of pure alcohol) per day in most of Europe, whereas it was 0.4 standard drinks per day for women in South-East Asia. Other notable differences include that the global TMREL was 0.1 standard drinks per day among women aged 15-19 years, whereas it was 0 among men of that age. In Western Europe, the TMREL for women was 0 up to age 24 years, and for men, up to age 30 years. Any consumption above these levels was regarded as unsafe.
This study is innovative in scope, but not entirely the ‘first of its kind’, when judged against prior GBD-based alcohol-attributable injury analyses. The original paper by Bryazka et al. (2020) indicated that while the TMREL was about 0.5 standard drinks per day, the non-drinker equivalence, defined as the consumption level at which health risk is equivalent to that of a non-drinker, was nearly two standard drinks per day. The very low TMREL figures used in the GBD studies are only slightly different from the previous risky level of alcohol consumption, which was simply described as “alcohol use”. TMRELs are also significantly lower than the levels suggested by the J-shaped curves that illustrate the relationship between alcohol consumption and mortality or morbidity, as reported by traditional epidemiology. TMREL may be a statistical construct based on summing all known risks, mainly overlooking health benefits and other positive effects that people may experience from consuming alcohol. This could partially explain why TMRELs do not align with the J-shaped association observed in classical epidemiological studies between alcohol consumption and all-cause mortality or morbidity.
This also suggests that by using these very low TMRELs instead of the non-drinker baseline or the nadirs from J-shaped associations, Yuan et al. (2025) assume that any alcohol consumption is harmful and that their HAU term used in this paper essentially means “Any Alcohol Use”.
Specific comments
Forum member Ellison agrees with both the positive and negative comments by other Forum members on this massive analytic paper. He did notice that the authors’ opening statement in the text mentions only the adverse effects of alcohol, or the ‘evils of alcohol’. “There is essentially no reference in the paper to the potential favourable effects of light to moderate consumption. Alcohol itself (while still in a jar, bottle, or other container) is not the problem; it is the excessive or inappropriate consumption of alcohol by people that causes problems.
This brings to mind the extensive attacks on ‘drugs’ and ‘drug importation’ by the current US administration. I am not aware that the current US government has stated that drugs are harmless until people use them, nor has there been the same effort to stop people from buying or using them. As for alcohol, we all agree that its misuse has many serious consequences and strongly support any proven measures to prevent such use. Still, we know that human consumption of some beverage containing alcohol has existed for perhaps 8,000 years or longer. If there were only adverse effects, why has it remained so common? Early on, humans recognised that certain substances had harmful effects and quickly learned to avoid plants and other materials that were solely toxic. Such knowledge has been passed down through generations to prevent the ongoing use of poisonous plants in our culture.
Therefore, there must be some positive effects of alcohol consumption, at least in moderation, such as pleasure, increased self-esteem, or greater sociability, and as we now understand, possibly longer life, which have contributed to its enduring presence in societies around the world. The present paper aims to help scientists focus on culture-specific factors related to alcohol abuse.”
Forum member Harding states that “Despite the obvious danger of drowning in a sea of acronyms, I struggle to see the point of a paper like this. As far as I can tell, the paper relies on the assumption that high alcohol use (HAU) related injuries are caused by HAU. Second, I fail to see how statistics collected across countries, which are inevitably compiled in different ways, can be combined and presented meaningfully. Third, statistics collected in this manner are no solid basis for recommending policy measures to address these perceived problems in individual societies, where the culture of alcohol consumption varies significantly.”
Forum member de Gaetano entirely agrees with Forum member Ellison, especially regarding the 8,000 years of alcohol use among all peoples. “I would also include the famous episode recounted in the Odyssey, 3,000 years ago, the encounter between Odysseus and Polyphemus: two contrasting ways of drinking wine; one is moderate and part of a broader culture, and the other excessive and associated with violence and arrogance. Since then, we have known that Odysseus managed to save himself and his companions, while Polyphemus became a victim of his own cultureless binge drinking.”
Forum member Mattivi considers that “methodologically, it seems difficult to find a common interpretation of data collected using non-rigorously standardised protocols and in countries where conditions are completely unequal. A sufficient level of standardisation or control over the main factors is an essential methodological prerequisite for making robust and accurate interpolations (and even more so, predictive extrapolations). Given that excessive alcohol consumption is highly negative and should be addressed, seeking single solutions that apply to all different situations (one size fits all assumption) may be an unlikely aspiration.”
Forum member Skovenborg remarks that “in the introduction it states “Alcohol is a toxic psychoactive substance that harms the body, damages the liver and brain, induces addiction, increases accident risk, and disrupts family and social relationships” which clearly demonstrates that the study is built on an anti-alcohol bias as a foundation. Also, I have not found a clear definition of “high alcohol use” anywhere in the study and the cause is probably that a common definition across the multitude of studies does not exist.”
Forum member Waterhouse is, however, “confused as to how the authors could link these adverse health events, Self-Harm and Interpersonal Violence (SIV), Transport Injuries (TI), and Unintentional Injuries (UII) to high alcohol use without measuring alcohol consumption in the studied populations. Then, if somehow achievable using techniques unfamiliar to me, how could they justify the need to intervene against high alcohol consumption, when it seems that the incidence of these adverse events is decreasing in nearly all of their datasets?”
References
Bryazka, D., Reitsma, M. B., Griswold, M. G., Abate, K. H., Abbafati, C., Abbasi-Kangevari, M., Abbasi-Kangevari, Z., Abdoli, A., Abdollahi, M., Abdullah, A. Y. M., Abhilash, E. S., Abu-Gharbieh, E., Acuna, J. M., Addolorato, G., Adebayo, O. M., Adekanmbi, V., Adhikari, K., Adhikari, S., Adnani, Q. E. S., … & Gakidou, E. (2022). Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020. Lancet (London, England), 400(10347), 185–235. https://doi.org/10.1016/S0140-6736(22)00847-9
Hindsholm, M. F., Das, A. S., Gokcal, E., Morotti, A., Rotschild, O., Simonsen, C. Z., Dipucchio, Z., Viswanathan, A., Greenberg, S. M., Anderson, C. D., Rosand, J., Goldstein, J. N., & Gurol, M. E. (2025). Effects of heavy alcohol use on acute intracerebral hemorrhage and cerebral small vessel disease. Neurology, 105(11), e214348. https://doi.org/10.1212/WNL.0000000000214348
Kilian, C., Klinger, S., Manthey, J., Rehm, J., Huckle, T., & Probst, C. (2024) National and regional prevalence of interpersonal violence from others’ alcohol use: a systematic review and modelling study. Lancet Regional Health Europe, 17, 40:100905. doi.org/10.1016/j.lanepe.2024.100905.
Yuan, Q., Chen, Z., Sun, B., Zheng, C., Kang, Y., Lou, Y., & Zheng, M. (2025). Global burden and trends of high alcohol use-related injuries from 1990 to 2030: a comprehensive assessment of self-harm and interpersonal violence, transport injuries, and unintentional injuries using global burden of disease 2021. Frontiers in Public Health, 13, 1675607. https://doi.org/10.3389/fpubh.2025.1675607
Comments on this critique by the International Scientific Forum on Alcohol Research were provided by the following members:
Henk Hendriks, PhD, Independent consultant and partner of the Nutrition Consultants Cooperative, Netherlands
Creina Stockley, PhD, MBA, Independent consultant and Adjunct Senior Lecturer in the School of Agriculture, Food and Wine at the University of Adelaide, Australia
Luc Djoussé, MD, DSc, Dept. of Medicine, Division of Aging, Brigham & Women’s Hospital and Harvard Medical School, Boston, MA, USA
R. Curtis Ellison, MD, Section of Preventive Medicine/Epidemiology, Boston University School of Medicine, Boston, MA, USA
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Richard Harding, PhD, Formerly Head of Consumer Choice, Food Standards and Special Projects Division, Food Standards Agency, UK
Giovanni de Gaetano, MD, PhD, Department of Epidemiology and Prevention, IRCCS Istituto Neurologico Mediterraneo NEUROMED, Pozzilli, Italy
Fulvio Mattivi, MSc, Professor and Scientific Advisor, Research and Innovation Centre, Fondazione Edmund Mach, San Michele all’Adige, Italy
Erik Skovenborg, MD, specialized in family medicine, member of the Scandinavian Medical Alcohol Board, Aarhus, Denmark
Andrew L. Waterhouse, PhD, Professor Emeritus of Enology, Department of Viticulture and Enology, University of California, Davis, CA, USA