Global burden of acute pancreatitis attributable to high alcohol intake from 1990 to 2021: findings and prevention potential based on GBD 2021.

This study utilized data from the GBD 2021 dataset, which is publicly available through the Global Health Data Exchange (GHDx) platform ( https://ghdx.healthdata.org/gbd-2021 ) [ 8 ].The extracted data were cross-checked against GBD source documentation to ensure completeness and consistency. The analysis covered 200 countries and territories from 1990 to 2021; Afghanistan, Kuwait, Mauritania, and Somalia were excluded due to incomplete or inconsistent data (see Table S1 for raw data). Detailed descriptions of GBD 2021 data extraction and modeling procedures have been reported elsewhere [ 9 , 10 ]. AP was defined according to the International Classification of Diseases, 10th Revision (ICD-10) code K85 [ 11 ]. High alcohol intake was defined as consumption exceeding the theoretical minimum risk exposure level (TMREL) [ 12 ], as determined by the GBD comparative risk assessment framework [ 13 ]. TMREL was calculated by integrating cause-specific relative risk curves for alcohol-related diseases, weighted according to their contribution to total alcohol-attributable DALYs. Alcohol exposure levels were estimated using the DisMod-MR 2.1 Bayesian meta-regression tool, which synthesizes survey data, sales records, and epidemiological studies, while adjusting for covariates and measurement bias [ 14 ]. The study population was restricted to individuals aged 15 years and older, consistent with the GBD comparative risk assessment methodology. In GBD analyses, alcohol-related risk estimation begins at age 15 due to negligible exposure and minimal attributable burden in younger age groups. Moreover, acute pancreatitis cases among children are rarely alcohol-related and were not explicitly modeled as such in GBD 2021. The analysis spanned the period from 1990 to 2021. Data were stratified by sex (male and female), age group (15–95 years, in 5-year intervals), and quintiles of the SDI [ 15 , 16 ]. SDI scores range from 0 to 1, with higher values indicating greater socioeconomic development. Based on 2021 SDI values, countries and territories were categorized into five development levels: low ( 0.81). Key outcome measures included DALYs, ASDRs, number of deaths, ASMRs, population attributable fractions(PAFs), summary exposure values (SEVs), and male-to-female ratios. Age standardization was performed using the GBD 2021 standard population. SEVs, ranging from 0 (no exposure) to 100 (maximum possible risk exposure), reflect the degree of population exposure to risk factors and are particularly useful for global comparisons (regional SEVs for 1990 and 2021 are presented in Table S2 ). PAFs were calculated by integrating SEVs to quantify the proportion of disease burden attributable to high alcohol consumption. Additionally, sex differences in DALYs and mortality were assessed across age groups, SDI levels, and regions. Joinpoint regression is widely used in epidemiological research to identify significant inflection points in time series data that may be missed by simple global trend analyses. In this study, Joinpoint Regression Program version 4.9.1.0 (National Cancer Institute, USA) [ 18 ]was applied, allowing between 3 and 6 Joinpoints during model fitting. This method partitions the study period into segments, each fitted with either a log-linear model. A grid search procedure systematically tests potential breakpoints, computing mean squared error, while the optimal model was selected based on the Akaike Information Criterion (AIC). Model optimization relies on Monte Carlo permutation tests, which iteratively compare hypotheses about the number of Joinpoints until the most suitable model is determined. To evaluate statistical significance, a two-sided p-value < 0.05 was used to determine whether the Annual Percent Change (APC) for a given segment or the Average Annual Percent Change (AAPC) [ 19 ] across all segments indicated a significant trend. Key output metrics therefore include both APCs for individual segments and the overall AAPC, offering a more nuanced perspective on temporal disease shifts. To further validate the robustness of the Joinpoint regression results, we conducted a sensitivity analysis using the Estimated Annual Percentage Change (EAPC). The EAPC is a widely used indicator of age-standardized rate (ASR) trends. It was calculated by fitting a log-linear regression model to the ASR over calendar years. The point estimate of the slope from this model was then converted into a percentage change per year. A 95% confidence interval (CI) was obtained from the regression fit. If both the EAPC estimate and the lower Limit of its 95% CI were greater than 0, the ASR was considered to have a significant increasing trend. Conversely, if both the EAPC estimate and the upper Limit of its 95% CI were below 0, a significant decreasing trend was inferred. Otherwise, the ASR was interpreted as stable over time. This approach provides a complementary assessment of temporal patterns and serves as a sensitivity test to support the reliability of the Joinpoint-derived AAPC results. Spearman’s rank correlation analysis was used to examine the association between SDI and key epidemiological indicators (ASMR, ASDR, and PAF) [ 20 ]. The Spearman correlation coefficient (r) was calculated as: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\:\mathrm r=1-\frac{6\sum\mathrm d_{\mathrm i}^2}{\mathrm n\left(\mathrm n^2-1\right)}$$\end{document} where d_i represents the rank difference for each pair of observations and n is the sample size. The r value ranges from − 1 to 1, with values closer to ± 1 indicating stronger correlations. Positive values represent a direct relationship, whereas negative values indicate an inverse association. Statistical significance was defined as a two-sided p -value < 0.05.We also calculated male-to-female ratios for DALYs and mortality across age groups and SDI quintiles to explore temporal trends in sex disparities. Decomposition analysis was conducted to quantify the contributions of population growth, aging, and epidemiological changes to changes in disease burden, measured by DALYs and mortality. Following Das Gupta’s framework, changes were partitioned into three components: population growth, reflecting changes in population size; population aging, capturing shifts in age structure; and epidemiological changes, representing variations in age-specific disease rates due to healthcare improvements or risk exposure. Age- and sex-specific population and DALY (or mortality) counts for acute pancreatitis attributable to high alcohol intake were extracted for each SDI quintile at two time points. Das Gupta’s closed-form algorithm was applied independently within each sex and SDI stratum. The algorithm first constructs three standardized series in which, successively, (i) only population size is allowed to vary while age structure and age-specific rates are held constant, (ii) only age structure is allowed to vary while population size and age-specific rates are held constant, and (iii) only age-specific rates are allowed to vary while population size and age structure are held constant. The difference between each standardized series and the reference series yields the corresponding effect; by construction these three effects are additive and sum exactly to the observed total change. The underlying assumptions are minimal and fully met in our dataset: (a) complete, non-missing age- and sex-stratified population and DALY counts for both time points, and (b) the mathematical requirement that the total observed change can be expressed as the exact sum of the three mutually exclusive components. Validation was performed by checking that the combined impact of population growth, aging, and epidemiological change precisely reproduced the observed overall shift in disease burden within each sex-SDI stratum, confirming exact internal consistency without reliance on any external model. All statistical analyses and visualizations were conducted in R software (version 4.2.2). Key packages included ggplot2 (v3.4.2), dplyr (v1.1.2), readxl (v1.4.2), and reshape2 (v1.4.4). Graphs were generated using ggplot2. Joinpoint regression analyses were performed with Joinpoint software (v4.9.1.0, National Cancer Institute) [ 20 ]. All GBD-derived estimates are reported with 95% uncertainty intervals (UIs), and derived statistics are presented with 95% CIs. This study was based entirely on publicly available, de-identified data from the GBD database. As such, it did not require approval from an institutional ethics review board. All procedures complied with ethical guidelines for secondary data analysis.

In 1990, the number of DALYs from AP attributable to high alcohol consumption was 401,671.32 (95%UI: 280,352.07–543,581.55), corresponding to an ASDR of 8.88 (95%UI: 6.25–12.04). By 2021, the DALYs had increased to 699,335.04 (95%UI: 486,293.01–924,031.99), while the ASDR slightly decreased to 8.22 (95%UI: 5.72–10.86) (Table 1 ). Mortality cases rose from 9,971.79 (95%UI: 6,888.25–13,403.82) in 1990 to 18,749.03 (95%UI: 12,763.28–24,677.82) in 2021, with the ASMR changing marginally from 0.24 (95%UI: 0.16–0.32) to 0.22 (95%UI: 0.15–0.29) (Table 2 ). Over the 32-year period, neither ASDR nor ASMR showed significant overall trends, with AAPCs of − 0.21% (95%CI: − 0.43 to 0.01) and − 0.20% (95%UI: − 0.42 to 0.01), respectively. However, joinpoint regression identified distinct inflection points, showing a decline in ASDR after 2005 and in ASMR after 2009, following initial upward trends in the earlier years. Table 1 DALYs for acute pancreatitis attributed to high alcohol intake: age-standardized rates with 95%UI and AAPC with 95% CI, 1990–2021 1990 2021 1990–2021 DALYs cases No. *10 5 (95% UI) ASDR per 100,000 No. (95% UI) DALYs cases No. *10 5 (95% UI) ASDR per 100,000 No.(95% UI) AAPC (95% CI) Global 4.02[2.80 to 5.44] 8.88 [6.25 to 12.04] 6.99 [4.86 to 9.24] 8.22 [5.72 to 10.86] −0.21 [−0.43 to 0.01] Female 0.51[0.29 to 0.76] 2.28 [1.28 to 3.40] 0.68 [0.38 to 1.03] 1.53 [0.86 to 2.32] −1.27 [−1.54 to −1.01] Male 3.51[2.49 to 4.66] 15.70 [11.20 to 20.86] 6.32 [4.50 to 8.22] 15.11 [10.76 to 19.66] −0.10 [−0.34 to 0.14] High SDI 1.05[0.76 to 1.37] 10.29 [7.45 to 13.42] 1.24 [0.86 to 1.65] 7.81 [5.51 to 10.27] −0.90 [−1.14 to −0.66] Female 0.20[0.12 to 0.30] 3.60 [2.18 to 5.28] 0.23 [0.13 to 0.34] 2.58 [1.51 to 3.83] −1.10 [−1.25 to −0.95] Male 0.85[0.63 to 1.09] 17.56 [12.99 to 22.45] 1.01 [0.71 to 1.31] 13.12 [9.53 to 16.94] −0.95 [−1.30 to −0.61] High middle SDI 1.68[1.21 to 2.24] 15.87 [11.43 to 21.08] 2.52 [1.80 to 3.36] 14.67 [10.44 to 19.43] −0.22 [−0.66 to 0.21] Female 0.24[0.14 to 0.37] 4.35 [2.49 to 6.62] 0.28 [0.15 to 0.43] 3.12 [1.68 to 4.75] −1.01 [−1.42 to −0.59] Male 1.44[1.07 to 1.90] 28.27 [21.01 to 37.00] 2.23 [1.64 to 2.93] 26.48 [19.43 to 34.79] −0.17 [−0.58 to 0.25] Middle SDI 0.76[0.49 to 1.08] 5.41 [3.47 to 7.78] 1.67 [1.15 to 2.21] 6.04 [4.19 to 7.97] 0.33 [0.20 to 0.46] Female 0.04[0.02 to 0.07] 0.60 [0.26 to 1.01] 0.09 [0.05 to 0.15] 0.67 [0.35 to 1.05] 0.31 [0.07 to 0.56] Male 0.72[0.47 to 1.01] 10.15 [6.61 to 14.40] 1.58 [1.11 to 2.08] 11.56 [8.13 to 15.22] 0.40 [0.26 to 0.53] Low middle SDI 0.38[0.19 to 0.63] 4.46 [2.33 to 7.46] 1.12 [0.70 to 1.55] 6.29 [3.94 to 8.80] 1.10 [0.93 to 1.26] Female 0.01[0.00 to 0.03] 0.36 [0.13 to 0.69] 0.05 [0.02 to 0.08] 0.57 [0.28 to 0.96] 1.45 [1.25 to 1.66] Male 0.36[0.18 to 0.61] 8.42 [4.37 to 14.13] 1.07 [0.67 to 1.48] 12.16 [7.48 to 16.91] 1.17 [1.00 to 1.35] Low SDI 0.14[0.07 to 0.25] 4.45 [2.11 to 7.72] 0.44 [0.26 to 0.71] 5.97 [3.44 to 9.47] 0.95 [0.88 to 1.03] Female 0.01[0.00 to 0.02] 0.57 [0.22 to 1.11] 0.02 [0.01 to 0.04] 0.76 [0.36 to 1.28] 0.95 [0.74 to 1.15] Male 0.13[0.06 to 0.24] 8.27 [3.92 to 14.87] 0.42 [0.24 to 0.68] 11.33 [6.53 to 18.26] 1.02 [0.94 to 1.10] Abbreviations: DALYs Disability-adjusted life years, ASDR Age standardized disability-adjusted life years rate, AAPC Average annual percent change, SDI Socio-Demographic Index, UI Uncertainty interval, 95% CI, 95% confidential intervals DALYs for acute pancreatitis attributed to high alcohol intake: age-standardized rates with 95%UI and AAPC with 95% CI, 1990–2021 Abbreviations: DALYs Disability-adjusted life years, ASDR Age standardized disability-adjusted life years rate, AAPC Average annual percent change, SDI Socio-Demographic Index, UI Uncertainty interval, 95% CI, 95% confidential intervals Table 2 Mortality for acute pancreatitis attributed to high alcohol intake: age-standardized rates with 95% UI and AAPC with 95% CI, 1990–2021 1990 2021 1990–2021 Mortality cases No. *10 3 (95% UI) ASMR per 100,000 No. (95% UI) Mortality cases No. *10 3 (95% UI) ASMR per 100,000 No. (95% UI) AAPC (95% CI) Global 9.97[6.89 to 13.40] 0.24 [0.16 to 0.32] 18.75 [12.76 to 24.68] 0.22 [0.15 to 0.29] −0.20 [−0.42 to 0.01] Female 1.53[0.90 to 2.27] 0.07 [0.04 to 0.11] 2.19 [1.18 to 3.28] 0.05 [0.03 to 0.07] −1.36 [−1.61 to −1.10] Male 8.44[5.96 to 11.22] 0.42 [0.30 to 0.56] 16.56 [11.59 to 21.73] 0.41 [0.29 to 0.53] −0.07 [−0.31 to 0.17] High SDI 3.03[2.12 to 4.00] 0.29 [0.20 to 0.38] 4.09 [2.71 to 5.50] 0.22 [0.15 to 0.29] −0.85 [−1.05 to −0.65] Female 0.71[0.41 to 1.06] 0.11 [0.07 to 0.17] 0.89 [0.48 to 1.34] 0.08 [0.04 to 0.11] −1.23 [−1.43 to −1.04] Male 2.32[1.68 to 2.98] 0.50 [0.36 to 0.64] 3.19 [2.20 to 4.20] 0.37 [0.26 to 0.49] −0.91 [−1.12 to −0.70] High middle SDI 3.95[2.79 to 5.16] 0.39 [0.27 to 0.51] 6.51 [4.58 to 8.56] 0.36 [0.26 to 0.47] −0.20 [−0.62 to 0.22] Female 0.65[0.37 to 0.96] 0.12 [0.07 to 0.18] 0.82 [0.42 to 1.25] 0.08 [0.04 to 0.12] −1.19 [−1.71 to −0.67] Male 3.30[2.42 to 4.23] 0.71 [0.53 to 0.92] 5.69 [4.16 to 7.43] 0.67 [0.49 to 0.87] −0.15 [−0.55 to 0.25] Middle SDI 1.75[1.11 to 2.51] 0.14 [0.09 to 0.20] 4.34 [2.85 to 5.78] 0.16 [0.10 to 0.21] 0.42 [0.27 to 0.57] Female 0.10[0.04 to 0.17] 0.02 [0.01 to 0.03] 0.27 [0.13 to 0.42] 0.02 [0.01 to 0.03] 0.25 [0.05 to 0.45] Male 1.65[1.07 to 2.36] 0.27 [0.17 to 0.38] 4.08 [2.72 to 5.43] 0.31 [0.21 to 0.42] 0.50 [0.34 to 0.66] Low middle SDI 0.88[0.44 to 1.47] 0.12 [0.06 to 0.19] 2.73 [1.68 to 3.90] 0.16 [0.10 to 0.24] 1.15 [0.98 to 1.33] Female 0.04[0.01 to 0.07] 0.01 [0.00 to 0.02] 0.14 [0.06 to 0.24] 0.02 [0.01 to 0.03] 1.43 [1.25 to 1.61] Male 0.84[0.42 to 1.42] 0.22 [0.11 to 0.36] 2.59 [1.58 to 3.68] 0.32 [0.19 to 0.46] 1.26 [1.08 to 1.45] Low SDI 0.34[0.16 to 0.59] 0.12 [0.06 to 0.21] 1.06 [0.61 to 1.68] 0.16 [0.10 to 0.26] 0.96 [0.87 to 1.05] Female 0.02[0.01 to 0.04] 0.02 [0.01 to 0.04] 0.07 [0.03 to 0.12] 0.02 [0.01 to 0.04] 0.79 [0.58 to 1.00] Male 0.31[0.15 to 0.56] 0.22 [0.10 to 0.40] 0.99 [0.57 to 1.60] 0.31 [0.18 to 0.48] 1.05 [0.95 to 1.15] Abbreviations ASMR Age standardized mortality rate, AAPC Average annual percent change, SDI Socio-Demographic Index, UI Uncertainty interval, 95% CI, 95% confidential intervals Mortality for acute pancreatitis attributed to high alcohol intake: age-standardized rates with 95% UI and AAPC with 95% CI, 1990–2021 Abbreviations ASMR Age standardized mortality rate, AAPC Average annual percent change, SDI Socio-Demographic Index, UI Uncertainty interval, 95% CI, 95% confidential intervals Throughout the study period (Fig. 1 A), males consistently had markedly higher ASDR and ASMR than females. By 2021, the ASDR in females had declined to 1.53 (95%UI:0.86–2.32) with an AAPC of − 1.27% (95%CI: − 1.54 to − 1.01), while the male ASDR remained relatively stable at 15.11 (95%UI: 10.76–19.66) with an AAPC of − 0.10% (95%CI: − 0.34 to 0.14). For ASMR (Fig. 1 B), females reached 0.05 (95% UI: 0.03–0.07) in 2021 (AAPC=–1.36%, 95% CI: − 1.61 to − 1.10), compared to males at 0.41 (95% UI: 0.29–0.53) (AAPC=–0.07%, 95% CI: − 0.31 to 0.17). Both indicators in females began to decline after 2004, while in males, declines were evident after 2005, with negative APCs persisting in subsequent years. Fig. 1 Global trends in ASDR and ASMR for acute pancreatitis attributable to high alcohol intake, 1990–2021. A ASDR trends by gender. B ASMR trends by gender. ASDR Age-standardized DALY rate, ASMR Age-standardized mortality rate, AAPC Average annual percent change, APC annual percent change, SDI Socio-Demographic Index Global trends in ASDR and ASMR for acute pancreatitis attributable to high alcohol intake, 1990–2021. A ASDR trends by gender. B ASMR trends by gender. ASDR Age-standardized DALY rate, ASMR Age-standardized mortality rate, AAPC Average annual percent change, APC annual percent change, SDI Socio-Demographic Index By SDI quintile (Figures S1–S2), the highest ASDR in 2021 occurred in the high-middle SDI group at 14.67 (95% UI: 10.44–19.43), showing an overall stable pattern (AAPC=–0.22%, 95% CI: − 0.66 to 0.21). In this group, rates increased from 1998 to 2005 (APC = 3.75%, 95% CI: 2.87 to 4.64) before falling sharply from 2005 to 2021 (APC=–3.10%, 95%CI: − 3.29 to − 2.91). The high-SDI group was the only quintile with a sustained decline (AAPC=–0.90%, 95% CI: − 1.14 to − 0.66), whereas the low-middle SDI group experienced the largest increase (AAPC = 1.10%, 95% CI: 0.93 to 1.26).In the high-middle SDI group, male ASDR remained stable overall (AAPC=–0.17%, 95% CI: − 0.58 to 0.25) but decreased notably after 2005 (APC = − 2.94%, 95% CI: − 3.12 to − 2.75), while females showed a continuous decline across the entire period (AAPC=–1.01%, 95% CI: − 1.42 to − 0.59). In lower-SDI regions, both sexes showed increasing trends, with a slower rise in females, whereas in higher-SDI regions, declines were observed in both sexes, more pronounced in females. ASMR patterns across the five SDI quintiles closely mirrored those of ASDR (Tables S3–S4).Sensitivity analysis using the EAPC model yielded similar results, supporting the robustness of the ASDR and ASMR trends across the five SDI quintiles (Table S5). Across the 21 GBD regions, Eastern Europe recorded the highest burden in 2021, with an ASDR of 64.03 (95%UI:46.21–84.35) and an ASMR of 1.35 (95% UI: 0.98–1.78). This region also showed the most rapid increase over the study period, with an AAPC of 2.00% (95%CI:1.43–2.58) for ASDR and 2.47% (95% CI:1.83–3.11) for ASMR, warranting close attention. In contrast, North Africa and the Middle East reported the lowest ASDR (0.36, 95% UI: 0.21–0.55) and ASMR (0.01, 95% UI: 0.006–0.02). The steepest declines were observed in Southern Latin America for ASDR (AAPC=–2.27%, 95% CI:–2.66 to − 1.88) and in the High-income Asia Pacific for ASMR (AAPC=–2.42%, 95% CI:–2.80 to − 2.05). Detailed AAPC estimates for all 200 countries and territories were presented in Tables S6 and S7. In 2021, the highest DALY count was observed in the 45–49-year age group, with ASDR increasing up to age 45, fluctuating thereafter, and rising again after age 80 (Fig. 2 A). Mortality cases peaked in the 55–59-year group (Fig. 2 B). The ASMR increased steadily with age, accelerating markedly after age 80. Across all age groups, males consistently exhibited substantially higher DALY and mortality counts and rates compared with females. Fig. 2 Age, gender, and SDI stratification of disease burden in 2021 and trends from 1990 to 2021. A Age- and sex-specific DALY rates and counts. B. Age- and sex-specific mortality rates and counts. C. AAPC in DALYs and mortality by gender. D. AAPC in DALYs and mortality by SDI quintile. (Note: DALY = disability-adjusted life-year. Age, gender, and SDI stratification of disease burden in 2021 and trends from 1990 to 2021. A Age- and sex-specific DALY rates and counts. B. Age- and sex-specific mortality rates and counts. C. AAPC in DALYs and mortality by gender. D. AAPC in DALYs and mortality by SDI quintile. (Note: DALY = disability-adjusted life-year. Over the 32-year period, the AAPC for both DALY rates and mortality rates was negative across all age groups for females worldwide, indicating an overall downward trend. In males, significant declines were observed in the 50–54 and 75–84 age groups, while a notable increase occurred in the 40–44 age group; other age groups remained stable, with AAPC 95% CIs crossing zero. By SDI quintile, most age groups in high-SDI regions showed a consistent decline, whereas most in middle-, low-middle-, and low-SDI regions exhibited increases. In high-middle-SDI regions, rates across most age groups remained stable (Fig. 2 C and D). The analysis of gender disparities in the global burden of pancreatitis attributable to high alcohol consumption revealed a progressive increase in the male-to-female ratio for both DALY and mortality rates across most age groups (Fig. 3 A and B), indicating a widening gender gap. From 1990 to 2021, the global DALY rate ratio exhibited a steady upward trajectory, reaching 9.33 for DALY rates and 7.58 for mortality rates in 2021. This disparity was most pronounced among younger adults, particularly those aged 25–44 years, suggesting a disproportionately greater impact of alcohol-related pancreatitis on males within these age groups (Table S8).Marked regional variations were observed. High- and high-middle-SDI regions exhibited comparatively lower male-to-female ratios but experienced a steady rise over the study period. Conversely, middle-, low-middle-, and low-SDI regions displayed a gradual decline in ratios; however, the absolute values remained considerably high, underscoring the urgent need for targeted gender-specific prevention strategies in these settings (Tables S9 and S10; Figures S3A and S3B). Fig. 3 Gender disparities across age groups in DALYs and mortality. A DALY rates by age and gender. B Mortality rates by age and gender Gender disparities across age groups in DALYs and mortality. A DALY rates by age and gender. B Mortality rates by age and gender In 2021, the global PAFs for pancreatitis attributable to high alcohol consumption were 17.01% (95% CI: 11.87–22.37) for ASDR and 15.22% (95% CI: 10.37–20.24) for ASMR (Fig. 4 A and B; Table S11). Across SDI quintiles, PAFs for ASDR increased with SDI, peaking in high-SDI regions at 25.05% (95% CI: 17.92–32.34) and reaching the lowest value in low-SDI regions at 10.08% (95% CI: 6.39–14.59). Fig. 4 PAFs of ASDR for acute pancreatitis due to high alcohol intake, 1990–2021. A PAFs in 2021 by gender, global, SDI quintile, and 21 GBD regions. B Global PAF trends from 1990 to 2021. PAF Population attributable fraction, GBD Global Burden of Disease PAFs of ASDR for acute pancreatitis due to high alcohol intake, 1990–2021. A PAFs in 2021 by gender, global, SDI quintile, and 21 GBD regions. B Global PAF trends from 1990 to 2021. PAF Population attributable fraction, GBD Global Burden of Disease From 1990 to 2021, the global PAFs for ASDR generally increased across most SDI quintiles, except in the high-middle-SDI group, where a declining trend was observed after 2005. Among females, the high-SDI quintile consistently recorded the highest PAFs for ASDR throughout the study period. In contrast, among males, the highest PAFs were observed in the high-middle-SDI quintile. Regionally, North Africa and the Middle East reported the lowest ASDR PAF in 2021 at 1.88% (95% CI: 1.09–2.90), whereas Central Europe recorded the highest at 29.71% (95% CI: 21.84–37.67). Overall, European regions tended to have higher PAF values. The spatial distribution and temporal trends of ASMR PAFs closely mirrored those observed for ASDR (Figures S4A and S4B). (A) PAFs in 2021 by gender, global, SDI quintile, and 21 GBD regions. (B) Global PAF trends from 1990 to 2021. (Note: PAF = population attributable fraction; GBD = Global Burden of Disease.) Globally, both ASDR and ASMR for alcohol-attributed pancreatitis demonstrated a spoon-shaped relationship with SDI, showing an overall positive correlation. At lower SDI values, both rates remained relatively stable; however, a marked increase was observed once SDI exceeded 0.60, followed by a decline when SDI surpassed 0.75. The correlation coefficients (R) between SDI and ASDR and ASMR were 0.32 and 0.35, respectively (Fig. 5 A and B). Fig. 5 Relationship between disease burden and SDI in 2021. A ASDR vs. SDI by GBD region. B ASMR vs. SDI by GBD region. C PAFs in ASDR vs. SDI. D PAFs in ASMR vs. SDI Relationship between disease burden and SDI in 2021. A ASDR vs. SDI by GBD region. B ASMR vs. SDI by GBD region. C PAFs in ASDR vs. SDI. D PAFs in ASMR vs. SDI In contrast, PAFs for both ASDR and ASMR exhibited an S-shaped relationship with SDI, with the steepest rise occurring between SDI values of 0.40 and 0.75. Both measures were positively correlated with SDI, with correlation coefficients (R) of 0.64 for ASDR and 0.62 for ASMR (Fig. 5 C and D). (A) ASDR vs. SDI by GBD region. (B) ASMR vs. SDI by GBD region. (C) PAFs in ASDR vs. SDI. (D) PAFs in ASMR vs. SDI. From 1990 to 2021, decomposition analysis revealed that the global DALYs attributable to high alcohol consumption in pancreatitis increased substantially (Fig. 6 A; Table S12). This change was primarily driven by population growth, which accounted for an 88.85% increase, followed by population aging, which contributed an additional 25.26%. In contrast, epidemiological changes exerted a mitigating effect, reducing DALYs by 14.11%. Fig. 6 Decomposition analysis of DALYs and mortality, 1990–2021. A Changes in DALYs globally and by SDI quintile. B Changes in mortality rates globally and by SDI quintile Decomposition analysis of DALYs and mortality, 1990–2021. A Changes in DALYs globally and by SDI quintile. B Changes in mortality rates globally and by SDI quintile When stratified by sex, males experienced a smaller aging-related increase in DALYs (23.12%) compared to females (68.87%). The impact of population growth was more comparable between sexes (83.11% for males vs. 174.49% for females). Notably, epidemiological improvements contributed to a 143.36% reduction in DALYs for females, far greater than the 6.23% reduction observed in males. Across SDI quintiles, high-SDI regions exhibited the largest aging effect on DALYs (90.27% for both sexes, 92.52% for males, and 164.95% for females), but also the most pronounced reductions from epidemiological changes (− 169.14% for both sexes, − 170.8% for males, and − 303.93% for females). In high-middle-SDI regions, aging (45.61%) and population growth (76.55%) contributed relatively equally, with smaller negative effects from epidemiological changes (− 22.16% overall). In contrast, low-SDI regions showed a slight negative impact from aging (− 0.95% overall) but demonstrated positive epidemiological contributions across both sexes. Similar patterns were observed in middle- and low-middle-SDI regions, where positive epidemiological changes were seen regardless of gender. Decomposition results for mortality rates mirrored those for DALYs (Fig. 6 B; Table S13). (A) Changes in DALYs globally and by SDI quintile. (B) Changes in mortality rates globally and by SDI quintile.

Using GBD 2021 data, we provide an updated global assessment of alcohol-attributable acute pancreatitis. While age-standardized DALY and mortality rates remain stable, absolute numbers are rising due to population growth and aging. The burden is highest in Eastern Europe and among young-to-middle-aged men. These findings emphasize the need for targeted alcohol-control policies, improved surveillance in low-SDI regions, and integration of alcohol screening into clinical care to reduce the global impact of AAP.

Pancreatitis is an acute inflammatory disease caused by premature activation of pancreatic enzymes, leading to pancreatic injury and multiple organ dysfunction. Over the past three decades, its global prevalence has continued to rise [ 19 , 21 , 23 ]. A recent worldwide epidemiological study highlighted alcohol consumption, obesity, and gallstone disease as major drivers, emphasizing that reducing alcohol intake is crucial for alleviating the future burden of pancreatitis [ 5 , 24 ]. Against this background, our study utilized the Global Burden of Disease (GBD 2021) database to systematically evaluate the global, regional, and national burden of AAP from 1990 to 2021, with stratified analyses by sex, age, and SDI. Compared with previous work, our study expands upon and extends existing findings. Tang et al. (2025) [ 6 , 7 ]analyzed global trends of alcohol-related pancreatitis using GBD 2021 data, while our analysis advances in three key aspects. First, we employed the TMREL framework to specifically quantify the risk attributable to high alcohol intake, rather than overall alcohol use. Second, we incorporated PAFs and SEVs to characterize the preventable burden and exposure patterns across alcohol-dependence quintiles. Third, segmented regression was applied to identify nonlinear inflection points in temporal trends, complementing the age–period–cohort models used previously. Moreover, our findings revealed a widening sex disparity—particularly among younger populations—and quantified the strong positive correlation between alcohol dependence levels and PAFs.These methodological innovations provide important insights for formulating targeted public health strategies. Overall, the evidence demonstrates a robust positive association between per capita alcohol consumption and the incidence and mortality of pancreatitis, particularly AAP. This relationship is evident both in national-level time-series data and in individual-level dose–response analyses.Systematic reviews and meta-analyses have shown that higher total alcohol intake substantially increases the risk of both acute and chronic pancreatitis, with spirits and binge drinking being the strongest contributors [ 25 ]. Notably, women may have lower thresholds or steeper risk curves [ 26 ]. These findings are consistent with clinical observations that attacks often follow episodes of heavy drinking. In several Northern European settings, reductions in alcohol sales were accompanied by parallel declines in AAP burden, while policy interventions such as increased taxation, restricted availability, and stricter drunk-driving enforcement were shown to reduce overall consumption and related health outcomes [ 27 , 28 ]. Individual-level evidence also supports these associations: a Swedish prospective cohort study reported a dose–response relationship between spirits intake and AAP risk, whereas associations with beer or wine were weaker [ 29 ]. Similarly, in the Netherlands, epidemiological data suggest that thousands of first-time alcohol-related acute pancreatitis cases occur annually, of which approximately 20% progress to severe disease with necrosis or organ failure [ 30 ], underscoring that reducing alcohol exposure remains essential even in high-income healthcare systems.Mechanistic studies further elucidate the biological pathways linking alcohol and pancreatitis. Ethanol and its metabolites (acetaldehyde, fatty acid ethyl esters) disrupt acinar cell calcium homeostasis, trigger premature zymogen activation, induce oxidative stress, and impair mitochondrial function, thereby initiating a cascade from reversible injury to necrotizing inflammation [ 31 , 32 ]. Taken together, both macro-level (policy) and micro-level (biological) evidence converge on a consistent conclusion: reducing overall alcohol consumption and heavy episodic drinking is pivotal to lowering the incidence and mortality of alcohol-attributable pancreatitis. Our study revealed that although global ASDR and ASMR for AAP remained relatively stable over the past 32 years, absolute DALYs and deaths increased markedly. This indicates that population growth and aging are the main drivers of the rising disease burden. Similar trends have been reported in prior studies; for instance, Tang [ 6 ]and Li [ 33 ] highlighted that the global burden of pancreatitis and pancreatic cancer has increased largely due to demographic shifts. In contrast, age-standardized rates showed minimal changes or even declines, suggesting that improvements in healthcare access and early diagnosis may have offset some of the demographic pressures. However, this “overall stability” conceals substantial regional and demographic disparities, particularly in low- and middle-SDI countries where both ASDR and ASMR continue to rise, reflecting pronounced global inequities in disease burden. Regionally, Eastern Europe exhibited the highest ASDR and ASMR in 2021, with the most rapid growth. Evidence from multiple European countries suggests that Eastern and parts of Northern Europe have long maintained higher per capita alcohol consumption and a greater proportion of spirits, with heavy episodic drinking (HED) being more prevalent, consistent with the higher proportion of alcohol-related pancreatitis in these regions [ 34 ].A systematic review also demonstrated that alcohol-related acute pancreatitis accounts for a larger proportion of cases in Eastern/Northern Europe. In Malmö, Sweden, reductions in alcohol sales coincided with declines in alcohol-related pancreatitis and other alcohol-related harms, underscoring the policy sensitivity of alcohol availability.Beyond these localized findings, Russia [ 35 ]provides a compelling example of the impact of comprehensive alcohol control policies. Beginning in the mid-2000s, Russia implemented a series of interventions—including substantial excise taxation on spirits, restrictions on sales hours, and advertising bans—which were associated with a sharp reduction in per-capita alcohol consumption (by about 43% between 2003 and 2016) and concomitant improvements in life expectancy. These outcomes illustrate how sustained, multifaceted national strategies can rapidly mitigate alcohol-related harms. In contrast, North Africa and the Middle East reported the lowest ASDR and ASMR globally, reflecting low alcohol accessibility and strict cultural or religious prohibitions. High-SDI countries, however, experienced overall declines, attributable to a combination of taxation and pricing strategies, restrictions on marketing and sales, drink-driving enforcement, and mature healthcare systems [ 36 ]. At the global level, the WHO’s SAFER initiative [ 37 ]—launched in 2018—offers a scalable framework of five high-impact interventions (Strengthening restrictions on alcohol availability, Advancing drink-driving countermeasures, Facilitating screening and brief interventions, Enforcing bans on alcohol marketing, and Raising prices through excise taxes).Taken together, these findings highlight the pivotal role of public health policies in shaping the burden of alcohol-attributable pancreatitis. Embedding our epidemiological results within these policy contexts underscores the urgent need for both regionally tailored interventions to reduce the preventable morbidity and mortality associated with AAP. Sex differences emerged as another notable finding. From 1990 to 2021, males consistently had higher ASDR and ASMR for AAP compared to females, with the sex gap widening over time, particularly among individuals aged 25–44 years. Previous research consistently shows that alcohol-induced acute pancreatitis occurs more frequently in men [ 38 ]. This disparity arises from both social behaviors and biological mechanisms. Behaviorally, men have higher drinking frequency and quantity and are more prone to HED, which multiple prospective cohort and case–control studies have strongly associated with increased risk of acute pancreatitis [ 26 ]. Biologically, higher alcohol dehydrogenase activity in men accelerates ethanol metabolism but leads to rapid acetaldehyde accumulation, triggering oxidative stress, inflammatory pathway activation, and acinar cell injury [ 39 , 40 ]. Conversely, estrogen has demonstrated anti-inflammatory and antioxidant properties, which may partially explain lower risk in women. Moreover, men more frequently present with comorbid smoking, obesity, and metabolic syndrome, all of which act synergistically with alcohol to exacerbate pancreatic injury [ 41 ]. Epidemiological evidence further supports this pattern: in high-drinking regions such as Northern and Central Europe, the rising incidence among men has been a primary driver of overall AAP increases [ 42 ]. Additionally, women often demonstrate greater health-seeking behaviors, presenting earlier for medical care and thus receiving timelier intervention, which may shorten hospital stays and reduce mortality [ 43 ]. In the analysis of PAFs, we observed that global ASDR- and ASMR-related PAFs increased with rising SDI levels, peaking in high-SDI regions. This reflects the paradox that, despite advanced healthcare systems, high-SDI countries maintain elevated alcohol-attributable burdens due to persistently high consumption levels. According to WHO reports, per capita alcohol intake in many high-income countries exceeds 8–10 L annually, with strong cultural acceptance of drinking [ 44 , 45 ]. Conversely, PAFs were lower in low-SDI regions, largely due to limited alcohol availability and lower consumption. However, low PAFs do not equate to low risk, as underdiagnosis and underreporting in resource-constrained settings may obscure the true burden [ 46 ]. Importantly, the observed “S-shaped” association between PAFs and SDI suggests that during early economic development (SDI 0.4–0.75), alcohol consumption tends to rise rapidly, while public health interventions lag behind, leading to steep increases in disease attribution [ 47 ]. At higher SDI levels, taxation, sales restrictions, and public education, combined with stronger healthcare systems, moderate this trend, explaining why PAF growth slows or declines in high-SDI countries. Decomposition analysis Further showed that from 1990 to 2021, global DALY increases in AAP were primarily driven by population growth (+ 88.85%) and aging (+ 25.26%), while epidemiological improvements contributed − 14.11%, partially offsetting demographic pressures. This finding is consistent with broader GBD studies, which demonstrate that despite advances in healthcare, demographic changes remain the dominant driver of rising burdens in alcohol-related diseases [ 14 ]. Notably, the contribution of aging was higher in women than men (68.87% vs. 23.12%), likely reflecting longer female life expectancy and cumulative risk [ 48 ]. Women may also be biologically more susceptible to alcohol toxicity, with higher body fat content and slower ethanol metabolism contributing to greater long-term exposure risk.Regional differences were also pronounced. High-SDI countries exhibited the strongest aging-driven effects, yet al.so showed the greatest negative contributions from epidemiological improvements, reflecting the benefits of advanced healthcare systems, standardized management, and intensive care resources [ 34 ]. In contrast, in low-SDI countries, aging contributed less, but positive epidemiological contributions were observed, suggesting gradual improvements in diagnosis and treatment. For example, in sub-Saharan Africa, while the absolute burden of alcohol-related disease continues to rise, recent studies suggest that better emergency care and diagnostic technologies have increased reporting of alcohol-related pancreatitis [ 45 ]. Nonetheless, these improvements remain insufficient to counterbalance population growth, underscoring the persistent gaps in healthcare capacity and alcohol regulation in low-resource settings. Based on these findings, we propose several public health priorities. First, in low- and middle-SDI countries, stricter alcohol control measures—including taxation, restricted sales, and marketing bans—should be paired with health education targeting young adults to curb high-risk drinking [ 36 ].Second, greater investment in primary healthcare is also needed to enhance early detection and management of acute pancreatitis and to strengthen surveillance systems [ 44 ].Third, in high-SDI countries, targeted screening and interventions for individuals with alcohol dependence or chronic pancreatitis remain essential to prevent rebounds in disease burden.Clinically, our results underscore the importance of alcohol cessation counseling and relapse prevention for patients with a history of AP, given the high risk of recurrence and progression to chronic pancreatitis. The disproportionate burden in younger males highlights the need for early identification and brief interventions during gastroenterology visits. Moreover, frequent comorbidities such as diabetes, liver disease, and malnutrition warrant integrated screening and management to improve long-term outcomes.Finally, international collaboration remains vital. By combining evidence-based alcohol policies with clinical prevention and comprehensive care, the global burden of alcohol-attributable pancreatitis can be substantially reduced [ 47 ]. This study has several Limitations. It relies on GBD 2021 estimates, which vary in quality across regions; underreporting, data sparsity, and ICD coding variability, especially in low- and middle-income countries, may bias results. GBD does not distinguish acute from chronic pancreatitis, limiting subtype-specific analysis. The ecological design precludes causal inference, and unmeasured confounders (e.g., obesity, smoking, viral hepatitis) may influence outcomes. Alcohol attribution using the theoretical minimum risk exposure level may misrepresent heterogeneous drinking patterns, while decomposition analysis assumes independent additive effects despite potential interactions. Additionally, the lack of additional external databases restricted our ability to further validate the findings, underscoring the need for future research with broader data sources.Finally, projections reflect past trends and may not capture future policies or emerging risk factors.Nonetheless, the study highlights important global patterns and disparities in AAP.

Acute pancreatitis(AP) is a common yet potentially Life-threatening inflammatory disorder of the pancreas, with a global annual incidence of approximately 30–40 cases per 100,000 population [ 1 , 2 ]. Alcohol consumption is one of its major modifiable risk factors, particularly prevalent among men and younger populations, and accounts for a substantial number of hospitalizations and deaths worldwide [ 3 ]. Both acute heavy drinking and long-term excessive alcohol intake can precipitate the onset and recurrence of AP.Despite advances in healthcare, alcohol-attributable acute pancreatitis(AAP) remains a pressing public health concern due to its high recurrence rate, severe complications, and considerable burden on healthcare systems [ 4 ]. In recent years, the understanding of AP epidemiology has been considerably advanced through data from the Global Burden of Disease (GBD) study. Li et al. [ 5 ] utilized the GBD 2021 dataset to systematically assess global and regional trends in incidence, mortality, and disability-adjusted life years (DALYs) for AP from 1990 to 2021, and projected trends to 2050. Tang et al. [ 6 ] conducted the first comprehensive assessment of AAP burden at global and regional levels, identifying epidemiological and demographic drivers across different populations. Another recent study [ 7 ] analyzed changes in mortality, DALYs, age-standardized mortality rate (ASMR), and age-standardized disability rate (ASDR) for alcohol-related pancreatitis at global, regional, and national levels from 1990 to 2021, and projected patterns to 2040, providing more granular insights into disparities across socio-demographic index (SDI) strata. Despite providing a solid epidemiological foundation, these studies have several limitations: (1) the relative contributions of population growth, population aging, and epidemiological change to AAP burden trends have not been systematically quantified; (2)few have applied a combination of joinpoint regression and decomposition analysis to identify trend inflection points and elucidate driving mechanisms; and (3) refined trend analyses stratified by SDI level, age, and sex remain limited, and discussions contextualizing findings within public health policy frameworks are insufficient. A comprehensive understanding of global distribution patterns and temporal trends in alcohol-related AP is therefore essential for developing effective prevention strategies and optimizing healthcare resource allocation. To address these gaps, the present study integrates joinpoint regression with decomposition analysis to comprehensively evaluate the global, regional, and national burden of AAP from 1990 to 2021, stratified by sex, age group, and SDI quintile. We quantify the relative contributions of population growth, aging, and epidemiological changes, and identify key temporal inflection points. Furthermore, we incorporate case studies such as the WHO SAFER initiative and Russia’s alcohol taxation reform to explore the public health implications of our findings across countries at different development levels. In doing so, this study not only validates and extends the conclusions of Tang et al. but also provides more targeted and actionable evidence to inform prevention strategies.

Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3. Supplementary Material 4. Supplementary Material 4. Supplementary Material 5. Supplementary Material 5.