The global, regional and national burden of pancreatitis due to alcohol use: Results from the Global Burden of Disease Study 2021 and projections to 2040.

Pancreatitis is an inflammatory disease of the pancreas, characterized by the abnormal activation of enzymes within the organ, resulting in self-destructive tissue damage. This process triggers an inflammatory response that results in pancreatic dysfunction and may interfere with the normal functioning of distal organs and systems, posing a significant global public health challenge [ 1 ]. A comprehensive meta-analysis conducted by Xiao et al. demonstrated that the global annual incidence of acute pancreatitis (AP) was 33.74 per 100,000 individuals (95% CI 23.33–48.81), significantly higher than the incidence of chronic pancreatitis (CP), which was 9.62 per 100,000 (95% CI 7.86–11.78). Additionally, the mortality rate of AP (1.60 per 100,000) was considerably higher than that of CP (0.09 per 100,000), highlighting the acute form’s greater lethality [ 2 ]. Although AP occurs in nearly equal proportions among men and women, CP, particularly alcohol-related CP, is more prevalent in men [ 3 ]. Alcohol, a key modifiable risk factor [ 4 ], is the most common cause of CP and the second most common cause of AP after gallstones [ 5 ]. Notably, the epidemiologic profile of alcohol-related pancreatitis is closely linked to socioeconomic inequalities. Patients in low-income regions (e.g., Eastern Europe) experience significantly worse prognoses, with an age-standardized disability-adjusted life-years (ASDR) approximately 17 times higher than that of patients in high-income areas (e.g., Asia-Pacific high-income regions) [ 6 ]. This disparity is primarily attributed to differences in alcohol consumption patterns, unequal distribution of social resources, limited access to healthcare, and insufficient public health interventions [ 7 , 8 ]. These findings underscore the urgent need to systematically investigate the interaction between socioeconomic inequality and alcohol consumption, with the GBD database providing a valuable resource for such analysis.Despite existing studies on the morbidity burden of pancreatitis, comprehensive investigations into its mortality, disability-adjusted life years (DALYs) and future trends remain scarce, often relying on relatively outdated data [ 9 ]. Moreover, prior research has predominantly been based on single-center or regional cohorts [ 10 , 11 ], which limits sample representativeness and fails to incorporate multidimensional metrics such as DALYs and sociodemographic indices (SDIs). This shortfall hinders a thorough understanding of the global heterogeneity of the disease. The GBD 2021 study integrates national-level data from 204 countries and territories, utilizing a Bayesian hierarchical model (BHM) to address underreporting and bias. This approach enables cross-regional and SDI-stratified analyses of alcohol-related pancreatitis, providing a more robust and comprehensive perspective [ 6 ]. Employing the latest data from the GBD 2021, this study applied age-period-cohort (APC) model to quantify the age effect, period effect, and cohort effect on the epidemiological patterns of alcohol-associated pancreatitis, including death, and DALY rates over the period from 1990 to 2021. Additionally, it employed the Bayesian-APC (BAPC) model to project disease trends from 2022 to 2040. To evaluate the global burden, trends of pancreatitis due to alcohol use, the analysis encompassed 204 countries and territories across 21 regions and five sociodemographic index (SDI) subgroups. The findings of this study will offer valuable insights into the complex interplay between alcohol consumption, sociodemographic development, and pancreatitis burden. This evidence-based foundation is crucial for designing targeted public health interventions and policies aimed at mitigating the disease burden of alcohol-induced pancreatitis and enhancing global public health outcomes.

Globally, between 1990 and 2021, both ASDR and ASMR of pancreatitis due to alcohol use showed a declining trend during this period, with EAPC of −0.32% (95% CI: −0.53 to −0.11) and −0.27% (95% CI: −0.45 to −0.08), respectively ( Tables 1 and 2 ). Among males, the EAPC for ASDR and ASMR were −0.18% (95% CI: −0.38 to 0.02) and −0.11% (95% CI: −0.28 to 0.07), respectively, which were significantly smaller declines compared to females, whose EAPC for ASDR and ASMR were −1.34% (95% CI: −1.66 to −1.01) and −1.41% (95% CI: −1.66 to −1.17), respectively ( Tables 3–6 ). The most substantial health burden of alcohol-induced pancreatitis was observed in middle-aged and elderly men worldwide, with number of DALY cases, number of Death cases, ASDR, and ASMR all substantially higher in males than females. Among males, the highest distribution of DALY cases occurred in those aged 35–39 years, while the highest distribution of death cases occurred in those aged 40–69 years. In contrast, in females, the highest ASDR was observed after age 85 ( Fig 1 ). Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASR, age-standardized rate; DALYs, disability-adjusted life-years;SDI, sociodemographic index; GBD, Global Burden of Diseases; EAPC, estimated annual percentage change; CI, confidence interval. Abbreviations: ASMR, age-standardized mortality rate; ASDR,age-standardized disability-adjusted life year. Regionally, Eastern Europe recorded the highest number of DALY cases, death cases, ASDR, and ASMR, whereas North Africa and Middle East reported the lowest values. The high-income Asia Pacific experienced the most pronounced annual decline in ASDR (−2.45%; 95% CI: −2.61 to −2.29) and ASMR (−2.96%; 95% CI: −3.14 to −2.78). Conversely, Southeast Asia showed the greatest increase in these rates, with EAPC for ASDR at 1.98% (95% CI: 1.87 to 2.08) and for ASMR at 2.13% (95% CI: 2.01 to 2.25) ( Tables 1 , 2 ). Nationally, the burden of pancreatitis due to alcohol use varied widely. Among combined genders (both), males, and females, the Russian Federation had the highest current ASDR at 69.92 (combined; 95% CI: 51.17 to 91.82), 133.67 (males; 95% CI: 99.29 to 174.38), and 14.13 (females; 95% CI: 7.19 to 22.20), respectively ( S1 - S3 Tables ). In terms of EAPC, the EAPC of ASDR is largest for Myanmar and smallest for Bhutan, and the EAPC of ASMR is largest for Cook Islands and smallest for Bhutan ( S4 Table ). Sociodemographically,the High-middle SDI quintile had the highest burden in terms of DALY cases, death cases, ASDR, and ASMR, while the lowest burdens were observed in the Low SDI quintile. Notably, countries in the High SDI quintile experienced the fastest declines in ASDR (−1%; 95% CI: −1.13 to −0.88) and ASMR (−0.92%; 95% CI: −1.06 to −0.78). In contrast, the Low-middle SDI quintile exhibited the greatest increases in these metrics, with EAPC for ASDR at 1.55% (95% CI: 1.33 to 1.77) and for ASMR at 1.61% (95% CI: 1.38 to 1.83) ( Tables 1 , 2 ). Age-period-cohort analyses of ASDR and ASMR for pancreatitis due to alcohol use indicated three core findings: Age effects revealed ASDR peaked in the 45–49 age group; Period effects showed continuous declines in High SDI countries but rising trends in Low/Middle SDI groups post-1997. Birth cohort effects highlighted generational ASDR reductions in High SDI nations versus upward trends in Low SDI regions. ( Fig 3 , Fig 4 and S5 Table ). Among these, high-middle SDI showed the highest ASDR, peaking at around 50 in the 45–49 age group ( S5 Table ), whereas low-middle SDI, as well as low SDI, had the lowest ASDR ( Fig 2A ). In terms of period effects, the global ASDR demonstrated a pattern of gradual decline followed by a slight increase and subsequent decrease. High SDI mirrored this trend and had the lowest ASDR among the five SDI quintiles, with a continuous downward trend observed. In contrast, middle, low-middle, and low SDI experienced an increase in ASDR starting in the 1997–2001 period, with low-middle SDI recording the highest ASDR during this period ( S6 Table ; Fig 2B ). Regarding birth cohort effects, the global trend in ASDR generally demonstrated a downward trajectory. This downward trend was mirrored in high SDI and high-middle SDI, whereas an upward trend was observed in the remaining SDI groups ( S7 Table ; Fig 2C ). For ASMR, after controlling for period and birth cohort effects, all five SDI quintile groups showed increasing trends with age. The highest ASMR was recorded in high-middle SDI, while low-middle and low SDI had the lowest values ( Fig 3 ). The period and birth cohort effects for ASMR followed patterns similar to those observed for ASDR. Abbreviations: APC, Age-period-cohort; ASDR, age-standardized disability-adjusted life year. Abbreviations: APC, Age-period-cohort; ASMR, age-standardized mortality rate. Abbreviations: ASDR,age-standardized disability-adjusted life year;SDI,sociodemographic index. The analysis reveals a non-linear trend: ASDR initially decreases with increasing SDI, then rises gradually, followed by a sharper increase, peaking at approximately 27 when SDI reaches around 0.74. After this peak, ASDR decreases rapidly. From 1990 to 2021, certain regions—such as Western Sub-Saharan Africa, Tropical Latin America, Central Europe, and Western Europe—exhibited ASDR values higher than expected based on their respective SDI. Conversely, Oceania and North Africa and Middle East showed ASDR values lower than expected for their SDI. A positive correlation between ASDR and SDI was observed across regions (p = 0.30, p < 0.001) ( Fig 4 ). Similarly, Fig 5 demonstrates the relationship between ASMR and SDI across regions, which closely parallels the relationship observed between SDI and ASDR. ASMR was also found to exceed expectations based on SDI in Western Sub-Saharan Africa, Tropical Latin America, Central Europe, Western Europe, and Central Latin America, while it was lower than expected in Oceania and North Africa and Middle East. A positive correlation between ASMR and SDI was observed across regions (p = 0.32, p < 0.001) ( Fig 5 ). Abbreviations: ASMR, age-standardized mortality rate;SDI,sociodemographic index. Between 1990 and 2021, global deaths and DALYs attributable to pancreatitis due to alcohol use increased significantly, with the most substantial rises observed in middle-SDI countries ( Fig 6 ). In Fig 6 , the black dots represent the overall contributions of three key influencing factors: epidemiologic change, population growth, and aging. Globally, epidemiologic change, population growth, and aging contributed −12.62%, 77.95%, and 34.66%, respectively, to the increase in deaths ( S8 Table ). In detail, −12.62% Epidemiologic change refers to negative contribution: This indicates that due to improvements in the epidemiological environment (such as reduced alcohol consumption, increased early diagnosis rates, improved treatment options, and the implementation of public health intervention measures), the ASMR of pancreatitis has decreased, thereby partially offsetting the increased burden brought about by population growth and aging. The black dot denotes the overall value of the change resulting from all three components. For each component, the magnitude of a positive value suggests a corresponding increase in pancreatitis due to alcohol use DALYs attributed to the component; the magnitude of a negative value suggests a corresponding decrease in pancreatitis due to alcohol use DALYs attributed to the component. Abbreviations: DALYs, disability-adjusted life-years; SDI, sociodemographic index. The contributions of these factors varied substantially across SDI quintile groups. High SDI experienced the largest impact of epidemiologic change (−87.41%), middle SDI showed the greatest influence of population growth (61.82%), and high-middle SDI demonstrated the most pronounced effect of aging (55.48%) ( S8 Table ). Similarly, for global DALYs, epidemiologic change, population growth, and aging accounted for −14.11%, 88.85%, and 25.26% of the total increase, respectively ( S8 Table ). The largest contribution of epidemiologic change was again observed in high SDI (−169.14%). Population growth had the greatest impact in high-middle SDI(76.55%), while aging contributed most significantly in high-middle SDI as well (45.61%) ( S8 Table ). We projected the number of death cases, ASMR, number of DALY cases and ASDR for pancreatitis due to alcohol use to 2040. Globally, the number of death cases is expected to decline from 2021 until 2025, after which it is projected to stabilize at approximately 1,800 cases per year ( Fig 7 ). Similarly, the ASMR is projected to decline from 2021, reaching a minimum in 2034, followed by a gradual increase thereafter ( S9 Table ). The number of DALY cases is projected to rise gradually from 2021, stabilize temporarily, and then decline slightly ( Fig 7 ). Meanwhile, the ASDR is expected to decrease gradually until 2035, after which it will begin to rise gradually ( S9 Table ). The observed decline in ASR alongside rising case numbers may be attributed to global population growth. Abbreviations: ASMR, age-standardized mortality rate; ASDR,age-standardized disability-adjusted life year;DALY, disability-adjusted life-year.

In summary, pancreatitis caused by alcohol use is a major global public health problem, with a significant increase in disease burden between 1990 and 2021, especially among middle-aged and older men. Analysis based on the GBD database shows that changes in the disease burden of alcoholic pancreatitis are primarily driven by two major demographic factors, population growth and population aging, and this trend and our findings predict a likely dramatic increase in the number of cases of alcoholic pancreatitis globally in the coming decades. The findings highlight the dual challenges facing the prevention and control of alcoholic pancreatitis: the continuing increase in the burden of disease and the significant heterogeneity of the burden distribution across different regions of the world. These findings provide an important evidence-based basis for optimizing public health policy making and medical resource allocation. It is recommended that global health policymakers adopt more targeted intervention strategies, implement flexible prevention and control measures, and focus on building personalized health care systems to address the differentiated health needs of different countries.

The primary data source for this study was the Global Burden of Disease 2021 (GBD2021), accessed through the Global Health Data Exchange GBD Outcomes Tool ( https://ghdx.healthdata.org/gbd-2021 ). GBD2021 offers a comprehensive assessment of mortality patterns, stratified by age and gender, encompassing 288 causes across 204 countries and 811 localities from 1990 to 2021, along with comparative risks for 88 risk factors [ 12 ]. It also evaluates disability-adjusted life expectancy for 371 diseases and injuries. Our analysis specifically focused on extracting GBD2021 datasets pertaining to alcohol-induced pancreatitis during the period from 1990 to 2021. Data included the number of Death cases, age-standardized mortality rate (ASMR), the number of DALY cases, and ASDR. DALYs, a key metric for measuring the public health impact of a disease, combine two components, years of life lost due to premature death (YLL) and years lived with disability (YLD). All rates are reported per 100,000 population.Additionally, the Sociodemographic Index (SDI), which quantifies the economic and sociodemographic status of a country or region, was used to classify the 204 national and regional entities into five categories according to SDI levels: High SDI (0.80–1.0), High-middle SDI (0.70–0.80), Middle SDI (0.60–0.70), Low-middle SDI (0.45–0.60), and Low SDI (0–0.45) [ 13 ]. Details regarding the modeling approaches and methodologies employed in GBD2021 are available in related publications [ 14 , 15 ]. This study adheres to the definition and classification criteria of alcoholic pancreatitis as stipulated by the Global Burden of Disease (GBD) study, which refers to acute or chronic pancreatitis caused by long-term alcohol consumption (≥ 4 standard drinks per day for at least 5 years), corresponding to ICD-10 codes K86.0 and K85.2. In the model construction, we integrated the GBD 2019 attribution framework and, through stratified analysis, isolated the association between alcohol exposure and the risk of pancreatitis from other known confounding factors such as cholelithiasis, smoking, hyperlipidemia, and hypercalcemia. To comprehensively evaluate the global health impact of pancreatitis due to alcohol consumption during the three-decade period spanning 1990–2021, we conducted descriptive analyses at the global, regional, national, and five SDI quintile levels. The primary indicators analyzed included the number of DALY cases, ASDR, the number of Death cases, and ASMR for pancreatitis due to alcohol use. To assess trends in these rates over time, we calculated the estimated annual percentage change (EAPC). EAPC is calculated through a linear regression model, taking into account the variation of the ratio by year,A linear regression model was constructed as y =  α  +  β x, in which y = ln (ASR) and x = calendar year. The EAPC was then calculated by (exp( β )-1) * 100%, with corresponding 95% confidence interval (CI) obtained through the analytical model [ 16 ].An ASR was considered to be in a downward trend if both the EAPC value and the upper limit of its 95% CI were  0. The ASR was classified as stable if the absolute value of the EAPC was close to 0. Additionally, we analyzed variations in the disease burden of pancreatitis due to alcohol use between male and female across different age groups to further understand gender-based differences. Our investigation additionally employed the age-period-cohort model (APC), applied to different SDI levels, which quantifies the independent contributions of age, period, and birth cohort to the ASMR and ASDR in pancreatitis due to alcohol use [ 17 ].Age effects are defined as biological changes in disease risk in a given age group, period effects reflect the impact of environmental or policy interventions across years, and cohort effects indicate cumulative exposure to health risks in the same birth cohort population [ 18 ].The APC model was obtained through the National Cancer Institute ( https://analysistools.nci.nih.gov/apc ), and the APC model could be written as a log-linear Poissonmodel: lnErij = lnθijNij = μ + αi + βj + γk, where Erij represents the expectation for rate, μ the mean effect, and αi, βj and γk the age, period, and cohort effects, respectively [ 19 ]. To evaluate the associations between SDI and both ASMR and ASDR, Spearman correlation coefficients (r) were computed. The formula for Spearman’s r is: r = 1 – (6Σd_i^2)/ [n(n^2–1)], where d_i represents the difference in ranks of corresponding variables and n is the number of observations [ 20 ]. The r value ranges from −1–1, with values closer to the extremes indicating stronger correlations. Positive r values signify positive relationships, whereas negative values reflect inverse relationships. Statistical significance was defined as a p-values less than 0.05 as the critical value. This study utilized the Das Gupta decomposition analysis method to determine the drivers of time-series changes in the number of Death cases, ASMR, the number of DALY cases, and ASDR for pancreatitis due to alcohol use over the period 1990–2021 [ 21 , 22 ]. This approach attributes time-series changes in the burden of death and disability-adjusted life years (DALYs) to three drivers: epidemiological variability, which refers to age-specific changes in disease-specific morbidity or mortality rates, mirroring the effects of medical technology innovations and interventions stemming from public health policies; population size expansion, which refers to the amplifying effect of growth in the total population size on the absolute value of the burden of disease even if age-specific rates of morbidityor mortality rates remain stable; and aging, which reflects the cumulative effect of a rising proportion of older age groups in the population on the burden of chronic and degenerative diseases, the risk of which typically rises significantly with age. In order to better formulate the allocation of health care resources and develop health care policies, we made further projections of the global burden of pancreatitis due to alcohol use through 2040. The projections were made using Bayesian Age-Period-Cohort Analysis (BAPC), which integrates integrated nested Laplacian approximation (INLA) for complete Bayesian inference [ 23 ]. By introducing the Prior Distribution and Markov Chain Monte Carlo (MCMC) algorithms, the Identifiability Problem, i.e., the covariance problem between age, period, and cohort effects, is solved in the traditional APC model. The covariance problem between age, period and cohort effects is solved in the traditional APC model. The BAPC and INLA packages within R software (v4.4.0) were utilized to conduct these projections. Our implementation enhanced traditional APC modeling through two integrated components: (1) Gaussian process priors with regularization for cohort effects, effectively balancing overdispersion mitigation and temporal pattern preservation; (2) The Knuiman-Cholesterol constraint system anchored by WHO surveillance trends, imposing biologically plausible boundaries on period slopes via INLA’s linear combination sampling. The GBD 2021 study is an open database, and all data is anonymous. No additional ethical approval is required for this study.

(XLSX) (XLSX) (XLSX) (XLSX) (XLSX) (XLSX) (XLSX) (XLSX) (XLSX)