Abstract
Background
Esophageal cancer is a major health burden, with smoking and alcohol as preventable risk factors. Comprehensive assessments of long-term trends, disparities, and future projections remain insufficient.
Methods
Using data from the Global Burden of Disease Study 2021, this observational study evaluated the burden of esophageal cancer attributable to smoking and alcohol-attributable esophageal cancer burden across 204 countries (1990–2021). Metrics included disability-adjusted life years (DALYs), age-standardized mortality rate (ASMR), and age-standardized DALY rate (ASDR). Temporal trends were assessed using estimated annual percentage change, decomposition analysis quantified drivers of burden changes, and a Bayesian age-period-cohort model projected burden to 2040.
Results
From 1990 to 2021, global deaths and DALYs increased, while ASMR and ASDR declined. Burden was uneven ; males had higher burden than females and peak burden occurring in the 65–79 age group. Geographically, ASRs declined notably in Central Asia and Latin America but rose sharply in West Africa. Middle SDI regions had the highest absolute burden; low-middle SDI regions showed increasing alcohol-attributable ASRs. Population growth and aging were primary drivers of absolute burden increases. Projections indicate continued declines in age-standardized rates, but male burden will remain higher.
Conclusions
Despite declining age-standardized risks, smoking and alcohol remain major drivers of the global burden of esophageal cancer. Population growth and aging will further increase absolute case numbers. Projections to 2040 show continued declines in age-standardized mortality and DALY rates, yet male burden will remain substantially higher. Prevention should prioritize tobacco/alcohol control and targeted strategies for high-burden regions, males, and the elderly.
Keywords: Esophageal cancer, global burden of disease, smoking, alcohol consumption, drinking, disability-adjusted life years (DALYs), age-standardized mortality rate (ASMR)
1. Introduction
As a gastrointestinal malignancy that poses a severe threat to human health, esophageal cancer has long imposed a heavy disease burden and remains a major global public health concern [1]. Data from the Global Burden of Disease Study 2021 (GBD 2021) show that the incidence and mortality of esophageal cancer worldwide have continued to rise: in 2021, there were over 600,000 new cases and more than 500,000 deaths globally, ranking it the 8th most common cancer and the 6th leading cause of cancer death worldwide [2]. Notably, its geographic distribution exhibits marked regional disparities, with a particularly high burden in East Asia, East Africa, and South America. Beyond inflicting physical and psychological suffering on individual patients, esophageal cancer also consumes substantial healthcare resources in countries worldwide and hinders the healthy development of social economies. Therefore, in-depth exploration of the risk factors for esophageal cancer and the subsequent development of effective prevention and control strategies have become urgent priorities in the global public health field.
Smoking and alcohol consumption, as major modifiable risk factors for esophageal cancer, have attracted increasing attention in recent years [3]. A large body of epidemiological evidence confirms a clear dose-response relationship between smoking/alcohol consumption and the risk of esophageal cancer [4]. Tobacco smoke contains a variety of carcinogens that can directly damage the esophageal mucosa and induce gene mutations [5]; alcohol, particularly its metabolite acetaldehyde, acts as a solvent to enhance the absorption of carcinogens and directly causes cellular damage [6]. Of particular note, when smoking and alcohol consumption act synergistically, the risk of esophageal cancer can be several to dozens of times higher than that of exposure to either factor alone, indicating a positive interaction between smoking and alcohol consumption in the pathogenesis of esophageal cancer [7]. Despite the implementation of public health policies such as tobacco and alcohol control in many countries globally, the prevalence of smoking and alcohol consumption remains high in numerous regions, continuing to exert a profound impact on the incidence and mortality of esophageal cancer.
The GBD quantifies health losses caused by various risk factors at the global, regional, and national levels through systematic and comparable methods, providing critical evidence for public health decision-making [8]. In recent years, while some studies have used GBD data to explore the total burden of esophageal cancer or its association with individual risk factors, there remains a lack of comprehensive, dynamic analyses focusing specifically on the esophageal cancer burden jointly driven by these two key behavioral risk factors (smoking and alcohol consumption). Furthermore, projecting future trends in disease burden is crucial for evaluating the effectiveness of current prevention and control measures, optimizing the allocation of future health resources, and formulating forward-looking cancer prevention and control strategies.
Therefore, this study aims to use GBD 2021 data to comprehensively analyze the disease burden, including mortality and disability-adjusted life years (DALYs) of esophageal cancer attributable to smoking and alcohol consumption at the global, regional, and national levels from 1990 to 2021, as well as trends in this burden. On this basis, a Bayesian age-period-cohort (BAPC) model will be used to project the disease burden up to 2040. The goal is to provide a scientific basis for developing and adjusting targeted primary prevention strategies for esophageal cancer, thereby contributing to the achievement of global cancer control objectives.
2. Materials and methods
2.1. Data sources
Data for this study were obtained from the GBD 2021, a comprehensive epidemiological database that systematically evaluates health losses caused by 371 diseases and injuries, as well as 88 risk factors, across 204 countries and regions worldwide [9,10]. The specific data used in this study included the number of deaths, disability-adjusted life years (DALYs), age-standardized mortality rate (ASMR), age-standardized DALY rate (ASDR), and their corresponding 95% uncertainty intervals (95% UIs) for esophageal cancer attributable to smoking and alcohol consumption, at the global, regional, and national levels from 1990 to 2021. All data were accessed online via the official GBD data platform (http://ghdx.healthdata.org/gbd-2021).
This study does not involve individual participants, nor does it include any personally sensitive information or biological samples. After review by the Ethics Committee of Mianyang Central Hospital in accordance with the Declaration of Helsinki, it was determined that this study falls within the scope of exemption, with the exemption number being 2026-003-030.
2.2. Definition of indicators
DALYs: A composite indicator used to measure the overall health loss caused by disease, injury, or premature death. It integrates two components: years of life lost (YLLs) due to premature death and years lived with disability (YLDs) [8]. ASMR: A mortality rate adjusted to eliminate the impact of differences in population age structure, enabling fair comparisons across different populations (e.g. different countries or time periods). ASDR: Age-adjusted using the GBD World Standard Population to ensure comparability of disease burden across countries and regions. SDI (sociodemographic index): A metric that synthesizes indicators of per capita income, average years of schooling, and total fertility rate to measure the level of social development in each region. GBD classifies countries and regions into five SDI categories: high SDI, high-middle SDI, middle SDI, low-middle SDI, and low SDI.
2.3. Statistical analysis
2.3.1. Trend assessment
The estimated annual percentage change (EAPC) was used to quantify the temporal trends of attributable disease burden from 1990 to 2021. A Joinpoint regression model was fitted to calculate the EAPC of age-standardized rate (ASR) and their 95% confidence intervals(CIs). The EAPC was derived based on a linear regression model: y = α + βx + ε, where y represents ln(ASR) and x represents the year. The formula for EAPC is EAPC = 100 × (exp(β) − 1). If both the EAPC value and the upper/lower bounds of its 95% CI are positive, the indicator shows an upward trend; if both are negative, a downward trend is indicated; if the 95% CI includes zero, the trend is considered statistically non-significant.
2.3.2. Stratified analysis
Populations were stratified by 5-year age groups (ranging from 20–24 years to 95+ years) and gender (male, female). This stratification helped identify population-specific patterns of esophageal cancer burden attributable to smoking and alcohol consumption, such as pinpointing the age groups with the most concentrated burden and quantifying the magnitude of gender disparities. Additionally, stratification was performed according to GBD’s five SDI regions (high, high-middle, middle, low-middle, low) to assess the association between disease burden and socioeconomic development levels, and to examine the distribution patterns of burden across regions at different development stages.
2.3.3. Decomposition analysis
A factor decomposition analysis was conducted to quantify the extent to which changes in the number of deaths and DALYs from esophageal cancer (attributable to smoking and alcohol) between 1990 and 2021 were driven by three factors: population growth, population aging, and changes in age-specific risk rates. The total changes in deaths and DALYs were attributed to these three factors, which helped clarify whether future efforts to control disease burden should focus on reducing risk rates or addressing changes in population structure.
2.3.4. Projection analysis
Based on historical ASR data from 1990 to 2021, a Bayesian Age-Period-Cohort (BAPC) model was applied to project the future disease burden of esophageal cancer attributable to smoking and alcohol. The BAPC model is a classic approach for projecting cancer incidence and mortality, as it effectively distinguishes between age, period, and cohort effects while providing uncertainty intervals for projections. Modeling and projections were performed using the BAPC package in R software.
2.4. Uncertainty assessment
In line with the GBD 2021 framework, Monte Carlo simulation was used to quantify the uncertainty of the final estimates. A probability distribution (e.g. normal distribution, beta distribution) was assigned to the input values of each calculation step (e.g. number of deaths), followed by 1,000 iterations of random sampling and computation. This process generated a set of 1,000 potential outcomes for each estimate (e.g. global attributable deaths in 2021). The 2.5th and 97.5th percentiles of these 1,000 outcomes were used to construct the 95% UI for the estimate. All results are reported as point estimates (e.g. median or mean) with their corresponding UIs. The UI reflects the precision of the estimate: a wider interval indicates greater uncertainty.
The GBD estimates are derived from a modeling framework that synthesizes available data sources, including vital registration, cancer registries, and survey data. This framework assumes that, after adjusting for known biases (e.g. underreporting, cause-of-death misclassification), the available data are representative of the underlying disease burden. In regions where primary data are sparse—particularly in low-income countries—estimates rely more heavily on covariate-driven predictions, which may introduce additional uncertainty and bias. For projections, the Bayesian age-period-cohort (BAPC) model assumes that historical trends in age-standardized rates will continue into the future, and that age, period, and cohort effects will follow patterns similar to those observed in the past. This assumption does not account for unforeseen changes, such as major shifts in tobacco or alcohol control policies, breakthroughs in cancer treatment, or substantial changes in risk factor prevalence. Consequently, uncertainty surrounding long-term projections widens with the length of the projection horizon.
2.5. Ethical statement
All data used in this study were obtained from public databases and did not involve personal privacy information or interventional experiments. Therefore, ethical committee approval was not required.
3. Results
3.1. Global disease burden of esophageal cancer attributable to smoking and alcohol (1990–2021)
From 1990 to 2021, the global number of deaths from smoking-attributable esophageal cancer increased from 136,450 (95% UI: 109,956–163,495) to 205,463 (95% UI: 156,544–257,209), while deaths from alcohol-attributable esophageal cancer rose from 51,624 (95% UI: 37,454–67,613) to 81,227 (95% UI: 58,925–108,002). Correspondingly, DALYs for smoking-attributable esophageal cancer increased from 3,640,952.03 (95% UI: 2,931,006.46–4,386,514.90) to 4,765,032.00 (95% UI: 3,626,147.91–5,973,195.15), and DALYs for alcohol-attributable esophageal cancer grew from 1,526,790.76 (95% UI: 1,110,948.64–2,008,352.06) to 2,113,085.02 (95% UI: 1,522,611.07–2,792,770.00). In contrast, both the ASMR and ASDR of esophageal cancer attributable to smoking and alcohol showed downward trends. For smoking-attributable esophageal cancer: ASMR decreased from 3.46 (95% UI: 2.79–4.13) per 100,000 population to 2.38 (95% UI: 1.81–2.98) per 100,000, and ASDR declined from 88.47 (95% UI: 71.26–106.51) per 100,000 to 54.26 (95% UI: 41.29–68.02) per 100,000. For alcohol-attributable esophageal cancer: ASMR fell from 1.27 (95% UI: 0.92–1.67) per 100,000 to 0.93 (95% UI: 0.68–1.24) per 100,000, and ASDR dropped from 36.21 (95% UI: 26.35–47.62) per 100,000 to 24.02 (95% UI: 17.33–31.75) per 100,000 (Table 1 and 2).
Table 1.
Deaths, DALYs, ASMR, ASDR, and EAPC for esophageal cancer caused by smoking at the global, SDI, and GBD regional levels, 1990–2021.
| 1990 |
2021 |
1990–2021 |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Death Number | ASMR | DALYs | ASDRs | Death Number | ASMR | DALYs | ASDRs | ASMR,%, | ASDR,%, | |
| (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% CI) | (95% CI) | |
| (95% UI) | (95% UI) | (95% UI) | (95% UI) | |||||||
| Global | 136450.37(109956.41,163494.86) | 3.46(2.79,4.13) | 3640952.03(2931006.46,4386514.90) | 88.47(71.26,106.51) | 205462.95(156544.40,257209.37) | 2.38(1.81,2.98) | 4765032.00(3626147.91,5973195.15) | 54.26(41.29,68.02) | −1.38(−1.49,−1.26) | −1.76(−1.88,−1.64) |
| SDI | ||||||||||
| High | 27561.91(22425.85,32120.08) | 2.49(2.03,2.90) | 667002.28(547352.29,777431.95) | 61.87(50.84,72.11) | 35277.59(27498.42,42903.72) | 1.65(1.29,2.00) | 746184.80(590485.64,902020.92) | 37.57(29.85,45.36) | −1.48(−1.62,−1.34) | −1.74(−1.89,−1.59) |
| High-middle | 46134.04(36653.63,56475.29) | 4.61(3.67,5.64) | 1247741.58(987238.54,1526746.79) | 121.09(96.00,148.22) | 74471.39(55783.12,97500.78) | 3.70(2.77,4.83) | 1742872.02(1305979.70,2286442.01) | 86.54(64.83,113.38) | −0.85(−0.99,−0.71) | −1.24(−1.38,−1.10) |
| Middle | 54585.53(43258.95,68159.18) | 5.36(4.24,6.69) | 1500656.98(1185141.83,1878363.23) | 135.81(107.56,169.80) | 82572.24(60685.35,108091.57) | 3.14(2.31,4.11) | 1926460.10(1423912.44,2515937.36) | 69.51(51.30,90.83) | −1.91(−2.03,−1.79) | −2.37(−2.48,−2.26) |
| Low-middle | 6223.47(4933.89,7593.05) | 1.06(0.84,1.30) | 171317.64(135855.82,208390.23) | 26.54(21.08,32.26) | 10174.20(7978.09,12436.32) | 0.73(0.57,0.89) | 267579.82(209088.71,327031.07) | 17.83(13.95,21.83) | −1.33(−1.41,−1.26) | −1.41(−1.48,−1.33) |
| Low | 1885.33(1436.65,2332.86) | 0.87(0.66,1.08) | 52654.61(40273.53,65297.89) | 21.82(16.63,27.06) | 2892.17(2196.50,3617.00) | 0.59(0.45,0.74) | 80075.30(60459.17,101002.93) | 14.79(11.21,18.59) | −1.47(−1.57,−1.37) | −1.53(−1.64,−1.42) |
| Regions | ||||||||||
| Andean Latin America | 57.65(42.99,74.44) | 0.30(0.22,0.39) | 1356.67(1007.80,1741.65) | 6.70(4.98,8.62) | 99.88(70.59,136.37) | 0.17(0.12,0.24) | 2180.04(1533.04,3000.62) | 3.70(2.60,5.09) | −1.76(−1.88,−1.64) | −1.89(−2.00,−1.77) |
| Australasia | 381.88(291.35,476.90) | 1.60(1.22,2.00) | 8759.57(6779.03,10799.92) | 37.17(28.81,45.80) | 456.75(319.93,606.89) | 0.83(0.58,1.09) | 9391.22(6771.03,12237.14) | 18.34(13.42,23.76) | −2.25(−2.34,−2.16) | −2.35(−2.43,−2.27) |
| Caribbean | 311.33(241.18,390.14) | 1.24(0.96,1.56) | 7405.83(5826.73,9146.16) | 28.54(22.43,35.29) | 549.40(415.03,706.25) | 1.01(0.76,1.30) | 13677.68(10369.87,17394.67) | 25.10(19.06,31.91) | −0.43(−0.52,−0.34) | −0.16(−0.26,−0.07) |
| Central Asia | 1581.88(1258.11,1919.85) | 3.37(2.66,4.10) | 43209.91(34560.88,51906.15) | 87.87(69.94,106.09) | 959.51(740.88,1191.71) | 1.21(0.93,1.51) | 25118.93(19572.21,31247.95) | 29.10(22.61,36.10) | −3.17(−3.27,−3.06) | −3.50(−3.60,−3.40) |
| Central Europe | 2064.04(1689.17,2427.74) | 1.36(1.11,1.61) | 57284.09(47399.94,66698.64) | 37.64(31.13,43.89) | 2346.85(1840.17,2859.84) | 1.09(0.86,1.33) | 59233.52(47009.41,71586.75) | 29.19(23.32,35.17) | −0.90(−1.07,−0.73) | −1.03(−1.22,−0.84) |
| Central Latin America | 472.61(368.04,580.40) | 0.62(0.48,0.77) | 11365.48(9040.49,13809.11) | 13.70(10.80,16.71) | 535.12(400.26,669.07) | 0.22(0.16,0.27) | 12412.01(9326.58,15540.98) | 4.90(3.68,6.13) | −3.60(−3.71,−3.49) | −3.53(−3.64,−3.42) |
| Central Sub-Saharan Africa | 253.45(177.08,349.37) | 1.09(0.77,1.50) | 7550.79(5256.42,10362.95) | 29.50(20.65,40.67) | 425.12(292.41,593.42) | 0.75(0.52,1.04) | 12830.91(8766.26,17838.44) | 20.08(13.81,28.04) | −1.30(−1.60,−1.00) | −1.34(−1.62,−1.06) |
| East Asia | 85523.97(66367.61,107038.31) | 10.13(7.87,12.63) | 2333079.68(1804708.83,2931960.53) | 252.13(196.07,316.03) | 141687.63(104483.64,185046.83) | 6.44(4.76,8.40) | 3270718.20(2400250.52,4302890.54) | 143.42(105.65,188.21) | −1.61(−1.78,−1.44) | −2.01(−2.16,−1.85) |
| Eastern Europe | 4868.57(4035.67,5628.30) | 1.69(1.40,1.95) | 140202.37(116782.58,161737.04) | 48.48(40.46,55.94) | 4236.68(3414.43,5053.12) | 1.20(0.97,1.43) | 115499.57(92858.19,137805.03) | 33.70(27.11,40.20) | −1.32(−1.45,−1.19) | −1.43(−1.57,−1.28) |
| Eastern Sub-Saharan Africa | 1104.64(829.37,1398.44) | 1.55(1.16,1.96) | 30928.29(23006.65,39298.64) | 38.91(29.15,49.45) | 1693.42(1275.66,2180.71) | 1.04(0.79,1.33) | 47810.31(35738.05,62052.71) | 26.19(19.66,33.83) | −1.57(−1.70,−1.44) | −1.58(−1.71,−1.45) |
| High-income Asia Pacific | 5640.17(4664.19,6519.63) | 2.79(2.31,3.23) | 138883.59(115201.98,159724.18) | 66.78(55.38,76.83) | 7005.23(5404.54,8509.79) | 1.41(1.10,1.71) | 130924.29(102462.30,158877.57) | 29.89(23.77,36.13) | −2.45(−2.60,−2.30) | −2.88(−3.08,−2.68) |
| High-income North America | 7367.01(5895.53,8711.99) | 2.12(1.70,2.50) | 178215.18(144110.91,208865.15) | 53.71(43.50,62.86) | 11025.76(8346.68,13831.45) | 1.63(1.24,2.04) | 242510.91(188030.15,298972.63) | 37.59(29.22,46.29) | −0.96(−1.14,−0.77) | −1.25(−1.43,−1.07) |
| North Africa and Middle East | 1194.98(930.40,1470.92) | 0.75(0.58,0.93) | 32015.34(25013.56,39332.88) | 18.14(14.13,22.23) | 2560.47(1955.53,3205.14) | 0.62(0.47,0.78) | 63038.14(48215.52,78830.69) | 13.63(10.40,17.04) | −0.68(−0.73,−0.64) | −1.04(−1.10,−0.99) |
| Oceania | 13.58(9.24,19.12) | 0.47(0.32,0.66) | 400.28(266.21,565.07) | 12.17(8.28,17.08) | 25.35(18.54,34.44) | 0.34(0.25,0.46) | 754.58(545.88,1037.56) | 8.85(6.45,12.03) | −1.18(−1.24,−1.12) | −1.15(−1.20,−1.09) |
| South Asia | 5168.51(4024.73,6440.29) | 0.94(0.73,1.18) | 144225.72(113131.47,178616.16) | 23.48(18.31,29.21) | 8510.91(6599.13,10653.79) | 0.60(0.46,0.75) | 220226.18(169861.94,274152.59) | 14.38(11.10,17.91) | −1.71(−1.82,−1.61) | −1.83(−1.94,−1.73) |
| Southeast Asia | 2109.81(1609.98,2742.01) | 0.86(0.65,1.10) | 58158.88(44280.30,76288.45) | 21.50(16.36,28.02) | 4283.03(3358.22,5338.32) | 0.66(0.52,0.82) | 115681.44(90160.02,143817.35) | 16.34(12.77,20.30) | −1.01(−1.07,−0.95) | −1.01(−1.06,−0.96) |
| Southern Latin America | 1066.68(825.43,1309.20) | 2.30(1.77,2.83) | 27244.17(21385.50,32955.24) | 57.98(45.55,70.24) | 868.14(640.57,1095.59) | 0.99(0.73,1.24) | 20256.20(15395.68,25375.97) | 23.65(18.08,29.54) | −2.72(−3.00,−2.44) | −2.92(−3.20,−2.65) |
| Southern Sub-Saharan Africa | 903.52(694.17,1119.61) | 3.43(2.61,4.27) | 25702.04(19964.27,31935.24) | 89.20(69.07,110.78) | 1199.48(919.09,1503.15) | 2.06(1.58,2.57) | 34444.08(26598.49,42996.17) | 54.82(42.13,68.25) | −2.23(−2.70,−1.76) | −2.16(−2.61,−1.70) |
| Tropical Latin America | 2869.99(2282.96,3441.14) | 3.25(2.58,3.93) | 77050.04(62062.06,91834.63) | 80.58(64.62,96.45) | 3476.59(2616.29,4421.40) | 1.35(1.01,1.72) | 86305.27(64685.97,108804.69) | 32.78(24.58,41.39) | −2.99(−3.09,−2.88) | −3.11(−3.25,−2.97) |
| Western Europe | 13322.66(10764.67,15590.85) | 2.32(1.87,2.71) | 313021.29(255343.48,363236.69) | 57.24(46.73,66.47) | 12965.49(9923.48,16181.97) | 1.37(1.06,1.69) | 265965.07(208553.03,327628.59) | 31.18(24.47,38.14) | −1.76(−1.85,−1.66) | −2.01(−2.12,−1.90) |
| Western Sub-Saharan Africa | 173.42(127.37,233.34) | 0.20(0.15,0.27) | 4892.80(3594.81,6616.31) | 5.21(3.82,7.02) | 552.15(396.89,730.17) | 0.28(0.20,0.37) | 16053.43(11434.79,21051.89) | 7.28(5.23,9.60) | 1.52(1.33,1.71) | 1.51(1.32,1.70) |
Abbreviations: DALYs, disability-adjusted life years; ASMR, age-standardized mortality rate; ASDR, age-standardized DALYs rate; EAPC, estimated annual percentage change; SDI, sociodemographic index; GBD, Global Burden of Disease; UI, uncertainty interval; CI, confidence interval.
Table 2.
Deaths, DALYs, ASMR, ASDR, and EAPC for esophageal cancer caused by drinking at the global, SDI, and GBD regional levels, 1990–2021.
| 1990 |
2021 |
1990–2021 |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Death Number | ASMR | DALYs | ASDRs | Death Number | ASMR | DALYs | ASDRs | ASMR,%, | ASDR,%, | |
| (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% UI) | per 100 000 | (95% UI) | (95% CI) | |
| (95% UI) | (95% UI) | (95% UI) | (95% UI) | |||||||
| Global | 51623.53(37454.34,67612.55) | 1.27(0.92,1.67) | 1526790.76(1110948.64,2008352.06) | 36.21(26.35,47.62) | 81226.79(58924.54,108002.00) | 0.93(0.68,1.24) | 2113085.02(1522611.07,2792770.00) | 24. 02(17.33,31.75) | −1.12(−1.18,−1.07) | −1.48(−1.54,−1.43) |
| SDI | ||||||||||
| High | 11343.4(8256.29,14654.36) | 1.05(0.77,1.35) | 303473.39(223732.60,386258.01) | 28.84(21.29,36.55) | 17264.65(12664.49,22087.90) | 0.84(0.62,1.07) | 398394.39(295765.60,510086.01) | 21.16(15.76,26.95) | −0.82(−0.94,−0.70) | −1.13(−1.26,−1.00) |
| High-middle | 19282.60(13816.71,25039.58) | 1.89(1.35,2.46) | 575868.56(414587.43,749177.26) | 55.33(39.81,72.05) | 28079.38(19971.28,39301.10) | 1.40(1.00,1.96) | 735978.69(515962.56,1034506.62) | 37.14(26.12,52.12) | −1.11(−1.18,−1.03) | −1.47(−1.56,−1.39) |
| Middle | 18515.33(13048.95,25763.65) | 1.69(1.19,2.35) | 570269.58(401427.42,791282.06) | 48.59(34.18,67.50) | 29411.94(20449.71,41023.23) | 1.08(0.75,1.50) | 780452.84(543434.22,1083240.33) | 27.44(19.11,38.07) | −1.59(−1.70,−1.49) | −2.03(−2.13,−1.93) |
| Low-middle | 1439.31(915.31,1978.90) | 0.22(0.14,0.30) | 45320.64(28551.84,62291.47) | 6.47(4.08,8.89) | 4170.50(2863.10,5554.33) | 0.28(0.19,0.37) | 127590.04(88295.53,170848.77) | 8.00(5.52,10.66) | 0.87(0.79,0.96) | 0.84(0.76,0.92) |
| Low | 1014.91(490.21,1405.06) | 0.43(0.21,0.60) | 31053.38(14466.28,43100.67) | 12.05(5.74,16.65) | 2258.06(1422.83,3129.87) | 0.43(0.27,0.59) | 69535.06(43583.15,96808.23) | 11.88(7.45,16.55) | −0.14(−0.39,0.11) | −0.20(−0.44,0.03) |
| Regions | ||||||||||
| Andean Latin America | 47.66(29.15,66.25) | 0.23(0.14,0.33) | 1346.06(837.00,1854.05) | 6.11(3.77,8.43) | 108.92(71.40,156.71) | 0.19(0.12,0.27) | 2810.51(1878.57,4051.47) | 4.63(3.08,6.68) | −0.55(−0.77,−0.34) | −0.76(−0.96,−0.55) |
| Australasia | 221.24(133.85,312.00) | 0.94(0.58,1.32) | 5470.96(3533.04,7417.16) | 23.75(15.46,32.04) | 478.30(343.95,617.31) | 0.89(0.65,1.13) | 10431.44(7691.40,13190.22) | 21.02(15.67,26.63) | −0.16(−0.31,−0.02) | −0.35(−0.49,−0.20) |
| Caribbean | 120.64(78.84,163.92) | 0.46(0.30,0.63) | 3476.16(2385.00,4648.64) | 13.00(8.89,17.43) | 274.50(183.76,365.05) | 0.51(0.34,0.67) | 7911.42(5394.67,10446.07) | 14.62(9.98,19.30) | 0.67(0.53,0.82) | 0.77(0.62,0.92) |
| Central Asia | 653.81(421.61,899.80) | 1.34(0.85,1.85) | 20007.09(13431.42,26918.40) | 39.36(25.78,53.57) | 395.52(249.27,548.20) | 0.46(0.28,0.65) | 11785.05(7601.57,16076.38) | 12.81(8.23,17.69) | −3.45(−3.65,−3.24) | −3.68(−3.89,−3.47) |
| Central Europe | 1042.79(727.52,1317.28) | 0.69(0.48,0.87) | 30870.12(22309.64,38771.20) | 20.52(14.86,25.69) | 1453.90(1060.79,1849.69) | 0.70(0.51,0.88) | 38742.24(28427.25,48889.53) | 19.81(14.72,24.92) | −0.19(−0.35,−0.04) | −0.38(−0.56,−0.20) |
| Central Latin America | 258.54(176.94,342.88) | 0.31(0.21,0.41) | 7663.17(5320.55,9947.26) | 8.26(5.69,10.87) | 458.42(303.11,613.05) | 0.18(0.12,0.24) | 12963.04(8730.70,17108.71) | 4.97(3.35,6.57) | −1.95(−2.12,−1.78) | −1.87(−2.04,−1.69) |
| Central Sub-Saharan Africa | 223.58(86.95,353.18) | 0.96(0.40,1.51) | 6892.27(2600.83,10927.74) | 26.36(10.18,41.73) | 463.53(228.96,707.85) | 0.80(0.40,1.22) | 14595.47(7105.30,22356.34) | 21.93(10.86,33.54) | −0.27(−0.96,0.43) | −0.30(−0.96,0.36) |
| East Asia | 30762.01(21597.24,42640.39) | 3.37(2.36,4.68) | 937653.71(657171.09,1297789.78) | 96.10(67.73,133.15) | 48350.79(32982.60,67549.80) | 2.16(1.47,3.02) | 1254375.39(846950.76,1769994.49) | 55.10(37.34,77.44) | −1.58(−1.67,−1.48) | −2.00(−2.09,−1.90) |
| Eastern Europe | 2550.52(1674.53,3297.98) | 0.88(0.59,1.14) | 78531.65(53443.75,100336.65) | 27.42(19.24,34.99) | 2131.41(1423.25,2871.02) | 0.62(0.42,0.83) | 62827.37(42526.37,83241.66) | 19.05(13.16,25.10) | −1.37(−1.50,−1.24) | −1.42(−1.55,−1.28) |
| Eastern Sub-Saharan Africa | 838.35(411.47,1175.11) | 1.08(0.54,1.51) | 25612.36(12059.11,36132.20) | 30.11(14.50,42.16) | 1841.21(1212.56,2544.64) | 1.04(0.70,1.44) | 56783.31(37142.13,79206.89) | 28.75(18.91,39.76) | −0.36(−0.48,−0.25) | −0.44(−0.55,−0.32) |
| High-income Asia Pacific | 2214.38(1669.83,2827.56) | 1.08(0.82,1.38) | 59212.53(44774.82,75159.31) | 28.28(21.39,35.90) | 2957.66(2102.82,3814.61) | 0.64(0.46,0.82) | 60735.13(44251.19,77224.51) | 15.06(11.01,19.05) | −2.02(−2.18,−1.85) | −2.40(−2.59,−2.20) |
| High-income North America | 1968.66(1221.97,2844.15) | 0.59(0.37,0.84) | 53961.10(35061.80,75672.44) | 16.88(11.03,23.35) | 4442.93(2997.65,5871.68) | 0.69(0.47,0.90) | 107665.24(73855.71,141583.49) | 17.84(12.42,23.26) | 0.61(0.47,0.74) | 0.26(0.14,0.39) |
| North Africa and Middle East | 60.19(35.23,86.09) | 0.03(0.02,0.05) | 1934.45(1172.44,2739.25) | 0.98(0.59,1.40) | 122.75(69.00,175.56) | 0.03(0.01,0.04) | 3674.26(2151.29,5187.19) | 0.71(0.41,1.01) | −0.85(−1.02,−0.68) | −1.14(−1.31,−0.97) |
| Oceania | 2.60(1.26,4.22) | 0.08(0.04,0.13) | 86.35(42.51,140.76) | 2.36(1.15,3.83) | 5.99(3.17,9.29) | 0.08(0.04,0.12) | 192.66(101.56,296.82) | 2.11(1.12,3.28) | 0.12(−0.19,0.43) | −0.03(−0.33,0.27) |
| South Asia | 906.21(371.87,1372.18) | 0.14(0.06,0.22) | 29686.83(11951.37,45062.27) | 4.32(1.78,6.52) | 3103.77(1848.45,4324.73) | 0.20(0.12,0.28) | 94735.76(56034.01,130271.24) | 5.82(3.46,8.04) | 1.13(0.94,1.31) | 0.99(0.82,1.15) |
| Southeast Asia | 407.52(262.32,606.89) | 0.15(0.10,0.22) | 13120.02(8420.59,19519.61) | 4.42(2.85,6.59) | 1834.89(1292.05,2476.04) | 0.26(0.18,0.35) | 56126.51(39469.74,75823.41) | 7.56(5.31,10.22) | 2.00(1.86,2.13) | 1.88(1.74,2.02) |
| Southern Latin America | 785.55(576.88,991.76) | 1.71(1.25,2.16) | 20037.43(14968.38,24942.55) | 42.80(31.93,53.34) | 642.51(452.54,839.28) | 0.73(0.52,0.95) | 14924.37(10691.57,19498.55) | 17.50(12.54,22.89) | −2.61(−2.79,−2.43) | −2.78(−2.95,−2.61) |
| Southern Sub-Saharan Africa | 531.01(332.23,703.59) | 1.88(1.18,2.52) | 17098.13(10548.38,22445.33) | 55.77(34.33,73.47) | 979.15(638.37,1282.93) | 1.62(1.08,2.14) | 30006.88(18809.88,39323.47) | 46.16(29.01,60.49) | −1.19(−1.69,−0.69) | −1.36(−1.85,−0.87) |
| Tropical Latin America | 877.10(620.37,1146.16) | 0.91(0.64,1.20) | 27682.82(19890.63,35886.36) | 26.84(19.07,34.68) | 1966.64(1401.96,2595.85) | 0.75(0.53,0.99) | 58259.12(41868.30,75967.53) | 21.89(15.74,28.55) | −0.67(−0.94,−0.39) | −0.74(−1.03,−0.45) |
| Western Europe | 6910.73(4840.30,8819.96) | 1.25(0.87,1.58) | 179531.43(128247.87,225483.52) | 34.25(24.60,42.83) | 8118.33(5945.16,10301.24) | 0.90(0.65,1.14) | 181375.65(132364.41,228819.66) | 22.58(16.73,28.38) | −1.12(−1.22,−1.01) | −1.44(−1.57,−1.32) |
| Western Sub-Saharan Africa | 240.43(144.83,357.22) | 0.27(0.16,0.40) | 6916.10(4154.48,10354.96) | 7.22(4.35,10.78) | 1095.67(720.54,1515.18) | 0.56(0.37,0.76) | 32164.20(20909.40,44971.85) | 14.37(9.44,19.92) | 2.88(2.64,3.12) | 2.76(2.53,2.98) |
Abbreviations: DALYs, disability-adjusted life years; ASMR, age-standardized mortality rate; ASDR, age-standardized DALYs rate; EAPC, estimated annual percentage change; SDI, sociodemographic index; GBD, Global Burden of Disease; UI, uncertainty interval; CI, confidence interval.
3.2. Regional and national disease burden of esophageal cancer attributable to smoking and alcohol
Between 1990 and 2021, the Central Latin America region saw the most substantial declines in the ASMR and ASDR of smoking-attributable esophageal cancer, with both metrics dropping by over 350%; for alcohol-attributable esophageal cancer, the Central Asia region recorded the largest decreases in ASMR and ASDR, exceeding 345% for both. In contrast, the ASMR and ASDR of esophageal cancer attributable to both smoking and alcohol rose significantly in Western Sub-Saharan Africa (Table 1 and 2). At the national level over the same period, West African countries (e.g. São Tomé and Príncipe, Guinea-Bissau) experienced the most notable increases in the ASMR and ASDR of smoking-attributable esophageal cancer, while Southern Latin American countries (e.g. Colombia) and Central Asian countries (e.g. Turkmenistan, Kazakhstan) had the most prominent decreases in these two rates; for alcohol-attributable esophageal cancer, West Asian countries (e.g. Afghanistan, Islamic Republic of Iran, Kuwait) showed obvious rises in ASMR and ASDR, whereas African countries (e.g. Mauritania, Sudan) had the sharpest declines in these rates (Figure 1A–D). In 2021, China reported the highest ASMR and ASDR for smoking-attributable esophageal cancer, while Cabo Verde had the highest ASMR and ASDR for alcohol-attributable esophageal cancer (Figure 2A–D).
Figure 1.
Global maps of the EAPC of ASMR and ASDR for esophageal cancer due to smoking and drinking in 204 countries and regions from 1990 to 2021.
(A) The EAPC in the ASDR of esophageal cancer related to smoking;
(B) The EAPC in the ASMR of esophageal cancer related to smoking;
(C) EAPC in the ASDR of alcohol-related esophageal cancer;
(D) EAPC in the ASMR of alcohol-related esophageal cancer.
Abbreviations: ASDR,age-standardized disability-adjusted life years rate; ASMR, age-standardized mortality rate; EAPC, estimated annual percentage change.
Figure 2.
Global maps of the ASDR and ASMR for esophageal cancer due to smoking and drinking in 204 countries and regions from 1990 to 2021.
(A) The ASDR of esophageal cancer related to smoking;
(B) (B) The ASMR of esophageal cancer related to smoking;
(C) The ASDR of esophageal cancer related to drinking;
(D) The ASMR of esophageal cancer related to drinking.
Abbreviations: ASDR, age-standardized disability-adjusted life years rate; ASMR, age-standardized mortality rate.
3.3. Gender and age-specific disease burden of esophageal cancer attributable to smoking and alcohol
Gender-specific analysis showed that in 2021, the ASMR and ASDR of esophageal cancer attributable to smoking and alcohol were far higher in males than in females, with the largest gender gap observed in high-middle SDI regions. Age-specific patterns indicated that the ASMR and ASDR of smoking- and alcohol-attributable esophageal cancer were relatively high in the 65–79 years age group (Figure 3A–D).
Figure 3.
The gender and age-related disease burden of esophageal cancer caused by smoking and drinking in 2021.
(A) The number of deaths due to smoking-related esophageal cancer by gender and age;
(B) Gender and age-specific DALYs for esophageal cancer related to smoking;
(C) The number of deaths due to esophageal cancer related to drinking by gender and age;
(D) Gender and age-specific DALYs related to esophageal cancer caused by drinking.
Abbreviations: ASDR,age-standardized disability-adjusted life years rate; ASMR, age-standardized mortality rate.
3.4. Socioeconomic disparities in disease burden of esophageal cancer attributable to smoking and alcohol
In 2021, middle SDI regions had the highest number of deaths and DALYs from esophageal cancer attributable to smoking and alcohol, while high-middle SDI regions recorded the highest ASMR (smoking: 3.70 [95% UI: 2.77–4.83] per 100,000; alcohol: 1.40 [95% UI: 1.00–1.96] per 100,000) and age-standardized DALY rate (ASDR) (smoking: 86.54 [95% UI: 64.83–113.38] per 100,000; alcohol: 37.14 [95% UI: 26.12–52.12] per 100,000). Notably, from 1990 to 2021, the ASMR and ASDR of smoking-attributable esophageal cancer showed a downward trend across all SDI regions, whereas the ASMR and ASDR of alcohol-attributable esophageal cancer exhibited an upward trend in low-middle SDI regions and either declined or remained stable in other SDI regions (Figure 4A–D).
Figure 4.
The disease burden of esophageal cancer related to smoking and drinking in different SDI regions.
(A) ASMR of esophageal cancer related to smoking;
(B) ASDR of esophageal cancer related to smoking;
(C) ASMR of esophageal cancer related to alcohol consumption;
(D) ASDR of esophageal cancer related to alcohol consumption.
Abbreviations: ASDR,age-standardized disability-adjusted life years rate; ASMR, age-standardized mortality rate;SDI, sociodemographic index.
Analysis of the disease burden of smoking- and alcohol-attributable esophageal cancer across 21 global regions revealed a complex association with sociodemographic development: ASMR showed a weak positive correlation with SDI (smoking: ρ = 0.40, p < 0.001; alcohol: ρ = 0.30, p < 0.001), and ASDR also exhibited a slight positive correlation with SDI (smoking: ρ = 0.38, p < 0.001; alcohol: ρ = 0.26, p < 0.001). Both correlations followed a non-linear pattern, initially increasing in regions with lower SDI and decreasing in regions with higher SDI (Figure 5A–D).
Figure 5.
Analysis of the Association between the Burden of Esophageal Cancer Related to Smoking and Drinking and SDI.
(A-B) Analysis of the association between the burden of smoking-related esophageal cancer and the social disease index;
(C-D) Analysis of the association between the burden of alcohol-related esophageal cancer and the social disease index.
Abbreviations: SDI, sociodemographic index.
Results of decomposition analysis indicated that between 1990 and 2021, high-middle SDI regions had the largest increase in deaths from smoking-attributable esophageal cancer, while middle SDI regions had the biggest rise in DALYs from this type of cancer—with population growth and aging contributing significantly to these increases and epidemiological changes exerting a negative contribution. For alcohol-attributable esophageal cancer, middle SDI regions had the largest increases in both deaths and DALYs, followed by high-middle SDI regions. Regionally, population growth and aging made substantial contributions to deaths and DALYs in East Asia. Overall, the increases in deaths and DALYs from smoking-attributable esophageal cancer across regions were mainly driven by population growth; for alcohol-attributable esophageal cancer, the rise in deaths across regions was primarily fueled by population aging, while the increase in DALYs was mainly driven by population growth (Figure 6A–D).
Figure 6.
Decomposition analysis of the disease burden of esophageal cancer related to smoking and drinking in each SDI region.
(A-B) Decomposition analysis of the disease burden related to smoking-induced esophageal cancer;
(C-D) Decomposition analysis of the burden of esophageal cancer related to alcohol consumption.
Abbreviations: SDI, sociodemographic index; DALYs.
3.5. Projection of disease burden of esophageal cancer attributable to smoking and alcohol (2022–2040)
Results from the BAPC model projection showed that between 2022 and 2040, the ASMR and ASDR of esophageal cancer attributable to smoking and alcohol would continue to decline. The decline was particularly prominent in males; however, males would consistently contribute more to the ASMR and ASDR of smoking- and alcohol-attributable esophageal cancer than females (Figure 7A–D).
Figure 7.
Prediction and analysis of the disease burden of esophageal cancer related to smoking and drinking from 2022 to 2040.
(A-B) Prediction analysis of the disease burden of smoking-related esophageal cancer from 2022 to 2040.
(C-D) Prediction analysis of the disease burden of alcohol-related esophageal cancer from 2022 to 2040.
4. Discussion
Based on data from the GBD 2021, this study comprehensively analyzed the global disease burden of esophageal cancer attributable to smoking and alcohol, its changing trends from 1990 to 2021, and projected these trends up to 2040. Several key findings were revealed, which hold important implications for the development of global esophageal cancer prevention and control strategies. These findings include that smoking and alcohol together impose a substantial burden of esophageal cancer, with significant disparities across regions, age groups, and genders; population growth and aging are key drivers of the increase in absolute burden; projections indicate that by 2040, although ASRs may stabilize or decline, the absolute number of cases and deaths will continue to rise due to population aging, and the burden on males will remain far higher than that on females.
This study found that between 1990 and 2021, the global number of deaths and DALYs from esophageal cancer attributable to smoking and alcohol both showed an upward trend, which is consistent with the overall global trends in esophageal cancer incidence and mortality [1]. However, the ASMR and ASDR exhibited a significant downward trend at the global level, reflecting the positive impacts of advancing tobacco and alcohol control policies, the popularization of endoscopic early diagnosis technologies, and improvements in treatment methods over the past few decades [11,12]. This divergent phenomenon that the absolute numbers rose but age-standardized rates fell highlights the profound impact of global population growth and aging on cancer burden. Decomposition analysis showed that population growth was the primary driver of the increased burden of smoking-attributable esophageal cancer, while population aging played a key role in the growth of alcohol-attributable burden. This indicates that even if the relative risk of esophageal cancer is controlled, the expanding and aging global population base will continue to place severe challenges on healthcare systems due to the sustained increase in the absolute burden of esophageal cancer in the future.
Notably, global efforts to control tobacco have begun to yield results. According to the 2025 Global Tobacco Epidemic Report, the number of countries implementing MPOWER tobacco control measures increased from 44 in 2008 to 155 in 2024, and the coverage of health warnings on tobacco packaging expanded from 9 countries in 2007 to 110 countries. Between 2007 and 2023, the global smoking rate decreased from 22.3% to 16.4%, and China’s community-based smoking cessation model was also recognized as a successful case, providing important references for primary smoking cessation services [13]. However, significant regional disparities in burden persist, indicating obvious inequalities in policy implementation, accessibility of health resources, and population risk awareness. To systematically address this public health challenge, it is recommended that countries continue to strengthen implementation of the World Health Organization’s Framework Convention on Tobacco Control, building on the evidence from regions where substantial declines in burden have been observed, as well as the Global Strategy to Reduce the Harmful Use of Alcohol, and promote comprehensive measures such as tax regulation, plain packaging, advertising restrictions, and control of alcohol availability. At the same time, in light of population aging trends, esophageal cancer prevention and treatment should be integrated into chronic disease management systems, with enhanced screening of high-risk groups and early intervention, thus controlling age- and gender-specific risks while alleviating the sustained growth of absolute burden.
The study found significant heterogeneity in the disease burden of smoking- and alcohol-attributable esophageal cancer across regions and countries. This heterogeneity arises from the complex interaction of multi-level factors, including accessibility of early screening, exposure to risk behaviors, standardization of diagnosis and treatment, distribution of medical resources, and the macro policy environment. Taking China as an example, esophageal cancer early screening faces multiple barriers: in many regions, the cost of gastroscopy and colonoscopy remains out-of-pocket for individuals, while high-quality screening relies on advanced equipment and professional medical personnel—creating both economic and technical barriers [14]. Additionally, risk behaviors such as smoking and alcohol consumption are more prevalent in regions and populations with lower socioeconomic status [15]; the complex cancer diagnosis and treatment process poses challenges for groups with limited health literacy; and underdeveloped regions lack support resources such as patient navigation systems [16]. At the policy level, economically developed regions typically have more sufficient financial resources to invest in health education, popular science promotion, and environmental governance, forming a positive cycle [17].
This imbalance in resources and policies is also reflected in global regional disparities. The substantial decline in age-standardized rates in regions such as Central Asia and Latin America suggests the effectiveness of their tobacco and alcohol control policies (e.g. tax increases, smoke-free legislation) [18]. However, age-standardized rates have risen sharply in some low-income regions such as West Africa, and the ASMR and ASDR of alcohol-attributable esophageal cancer have shown an opposing upward trend in low-middle SDI regions, warning of insufficient control of risk factors and weak early screening and diagnosis systems [19]. As a populous country, China bears the highest absolute burden and ASMR of esophageal cancer, which is closely linked to its high smoking and alcohol consumption rates, the predominance of squamous cell carcinoma as the pathological type, and traditional dietary culture [20,21]. Meanwhile, the extremely high age-standardized rates in countries such as Cabo Verde remind us to pay attention to the unique combinations of risk factors or data reporting differences that may exist within specific countries. To address these challenges, differentiated strategies are essential. In East Asia, strengthening endoscopic screening for high-risk populations (e.g. males aged 50 years or older with a history of smoking or alcohol use) and establishing tiered diagnosis and treatment networks linking county-level hospitals with tertiary centers should be prioritized. In regions with rising burden such as West Africa and low-middle SDI countries, integrating tobacco and alcohol control into essential public health service packages and building primary healthcare capacity for early symptom recognition are critical. Regions that have achieved substantial declines, such as Central Asia and parts of Latin America, can serve as models for sharing best practices, including sustained tax increases and comprehensive alcohol control policies.
Consistent with previous studies, this study reconfirms that the burden of esophageal cancer attributable to smoking and alcohol is far higher in males than in females [22]. This is mainly due to the fact that males have much higher prevalence and intensity of smoking and alcohol consumption than females. Furthermore, compared with females, males are more exposed to other esophageal cancer risk factors such as hot diets and insufficient intake of vegetables and fruits, and they also have lower cancer prevention literacy and lack awareness of self-health management [13,23]. On the other hand, relevant studies have also pointed out that estrogen receptors exist in esophageal tissue, which can inhibit the growth and malignant progression of esophageal cancer cells [10]. The largest gender gap in disease burden is observed in high-middle SDI regions, which may be related to the longer duration and greater intensity of smoking and alcohol consumption among males in these regions, while the smoking rate among females remains generally low. Given this persistent disparity, gender-sensitive interventions are urgently needed. Workplace-based programs for smoking cessation and alcohol reduction in settings such as factories and offices can reach large numbers of men. Health systems should integrate cancer prevention into occupational health services and train primary care providers to proactively discuss risk behaviors with male patients. Policy measures such as enforcing advertising restrictions on products marketed predominantly to men can further support behavior change.
In terms of age distribution, the peak disease burden occurs in the 65–79 years age group—consistent with esophageal cancer’s characteristic as a typical ‘elderly cancer’ associated with long-term cumulative exposure. This emphasizes the importance of health screening and intervention for middle-aged and elderly populations [24]. For elderly individuals, particularly those aged 65–79 years where the burden peaks, countries should prioritize this group as key monitoring targets and establish a long-term mechanism for cancer screening. Integrating opportunistic screening into routine chronic disease management—such as follow-up visits for hypertension or diabetes—is essential. Risk-based screening protocols should be implemented for elderly individuals with a long-term history of smoking or alcohol use, and age-appropriate guidelines that balance benefits and harms need to be developed. Establishing ‘cancer screening and early diagnosis centers’ at county-level hospitals can facilitate early intervention for precancerous lesions or early-stage esophageal cancer, thereby curbing disease progression and improving five-year survival rates. Comprehensive geriatric care packages should include nutritional support, smoking cessation, and alcohol reduction interventions, complemented by family-based education to encourage behavior change. Health systems should also strengthen diagnostic and treatment capacities in advance to meet the growing demand for esophageal cancer services driven by population aging.
Notably, the relationship between disease burden and socioeconomic development level (SDI) exhibits a complex non-linear pattern. In low SDI regions, the burden increases with rising SDI—possibly reflecting a ‘risk transition’ during the early stages of economic development, where the accessibility of tobacco and alcohol increases while effective control measures lag behind [15,25,26]. In contrast, in high SDI regions, the burden decreases with rising SDI, reflecting the positive role of robust public health systems, effective health education, and advanced medical technologies in disease control. Middle SDI regions bear the highest absolute burden; this ‘middle-income trap’ phenomenon indicates that these regions are simultaneously facing the dual challenges of traditional risk factors (smoking and alcohol consumption) and modern lifestyle risk factors, placing enormous pressure on health systems [27]. Targeted and differentiated prevention and control strategies are recommended: high-burden regions (e.g. East Asia) should strengthen endoscopic screening and early diagnosis for high-risk groups; regions with rising disease burden (e.g. West Africa, low-middle SDI countries) should integrate tobacco and alcohol control into basic public health service packages.
Finally, the projection study based on the Bayesian Age-Period-Cohort (BAPC) model provides cautiously optimistic results. Projections indicate that if current trends continue, the age-standardized disease burden of global smoking- and alcohol-attributable esophageal cancer will continue to decline by 2040, with a particularly notable decline among males—instilling confidence in esophageal cancer prevention and control. This optimistic projection is based on the assumption that global tobacco and alcohol control efforts and advances in medical technology will continue. However, the projection that the burden on males will always be higher than that on females indicates that targeted interventions for males will be a top priority in future work. To achieve this goal, strengthening the implementation of cost-effective policies recommended in WHO’s Framework Convention on Tobacco Control and Global Strategy to Reduce the Harmful Use of Alcohol—such as tax increases, comprehensive advertising bans, and the creation of smoke-free environments—may help sustain the downward trends observed in this study, particularly in regions where such policies have been effectively implemented [28]. Furthermore, the projected increase in the absolute number of cases serves as a reminder to remain vigilant: the inertial force of population growth means that any slack in prevention and control efforts may lead to unavoidable deaths and disabilities.
It is important to note the limitations of this study. First, the study is based on estimated data from the GBD model, whose accuracy depends on the quality and completeness of data from source countries. Data quality and availability vary considerably across regions, with greater uncertainty in low-income countries where vital registration systems and cancer registries may be incomplete or absent. In these settings, estimates rely more heavily on modeling assumptions and covariate data, which may introduce additional bias. Second, although the GBD study quantifies the independent attributable burden of smoking and alcohol, these two risk factors often co-occur and are known to have a synergistic biological interaction in esophageal carcinogenesis. The statistical decomposition of their independent effects cannot fully account for this interaction, leading to potential overlap in the attribution of disease burden and uncertainty in the estimated individual contributions. Third, the BAPC projections are based on historical trends and assume that past age, period, and cohort effects will continue into the future. Long-term projections are inherently uncertain, as they cannot anticipate unforeseen breakthroughs in prevention or treatment, major shifts in global tobacco and alcohol control policies, or changes in risk factor prevalence that deviate from historical patterns. The further the projection extends, the wider the uncertainty intervals and the lower the precision of the estimates. Therefore, the projections to 2040 should be interpreted as a continuation of observed trends under current conditions, rather than a definitive forecast.
5. Conclusions
In summary, despite the decline in age-standardized rates, smoking and alcohol consumption remain the primary modifiable drivers of the substantial global disease burden of esophageal cancers. This burden is highly unevenly distributed, concentrating in males, elderly populations, and middle SDI regions. Population growth and aging will continue to drive an increase in the absolute number of cases. Projections of the global esophageal cancer burden to 2040 demonstrate that while age-standardized rates will continue to decline, the burden on males will remain markedly higher than on females. The projected persistent gender gap calls for the design and implementation of gender-sensitive prevention strategies to mitigate this ongoing disparity. Future prevention and control efforts should, at the global level, further strengthen tobacco and alcohol control policies, while implementing targeted prevention, early diagnosis, and treatment strategies tailored to high-burden regions, genders, and age groups, so as to address this ongoing public health challenge.
Funding Statement
This study was supported by the China Postdoctoral Science Foundation (Grant number 2025M772814); the Medical Youth Innovation Research Project of Sichuan Province(Grant number Q23004).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Global Burden of Disease study 2021 data resources are available online from the Global Health Data Exchange (GHDx) query tool (https://vizhub.healthdata.org/gbd-results/).
References
- 1.Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–249. doi: 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
- 2.Jin W, Huang K, Ding Z, et al. Global, regional, and national burden of esophageal cancer: a systematic analysis of the Global Burden of Disease Study 2021. Biomark Res. 2025;13(1):3. doi: 10.1186/s40364-024-00718-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cai Y, Lin J, Wei W, et al. Burden of esophageal cancer and its attributable risk factors in 204 countries and territories from 1990 to 2019. Front Public Health. 2022;10:952087. doi: 10.3389/fpubh.2022.952087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kamangar F, Nasrollahzadeh D, Safiri S, et al. The global, regional, and national burden of oesophageal cancer and its attributable risk factors in 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet Gastroenterology & Hepatology. 2020;5(6):582–597. doi: 10.1016/S2468-1253(20)30007-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cook MB, Kamangar F, Whiteman DC, et al. Cigarette Smoking and Adenocarcinomas of the Esophagus and Esophagogastric Junction: a Pooled Analysis From the International BEACON Consortium. J Natl Cancer Inst. 2010;102(17):1344–1353. doi: 10.1093/jnci/djq289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chandramouleeswaran PM, Guha M, Shimonosono M, et al. Autophagy mitigates ethanol-induced mitochondrial dysfunction and oxidative stress in esophageal keratinocytes. PLoS One. 2020;15(9):e0239625. doi: 10.1371/journal.pone.0239625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kwon MJ, Kang HS, Choi HG, et al. Risk for Esophageal Cancer Based on Lifestyle Factors–Smoking, Alcohol Consumption, and Body Mass Index: insight from a South Korean Population Study in a Low-Incidence Area. JCM. 2023;12:7086. doi: 10.3390/jcm12227086. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Vos T, Lim SS, Abbafati C, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet. 2020;396(10258):1204–1222. doi: 10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ferrari AJ, Santomauro DF, Aali A, et al. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021. The Lancet. 2024;403(10440):2133–2161. doi: 10.1016/S0140-6736(24)00757-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brauer M, Roth GA, Aravkin AY, et al. Characterization of serum estradiol level and tissue estrogen receptor immunostaining with clinical response and reproductive factor changes in Chinese female patients with esophageal squamous cell carcinoma. The Lancet. 2024;403:2162–2203. doi: 10.1016/S0140-6736(24)00933-4. [DOI] [PubMed] [Google Scholar]
- 11.Danpanichkul P, Pang Y, Mahendru T, et al. Sex disparities in alcohol-associated liver disease and subtype differences in alcohol-attributable cancers in the United States. Clin Mol Hepatol. 2025;31(3):1058–1070. doi: 10.3350/cmh.2025.0169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Arnold M, Rutherford MJ, Bardot A, et al. Progress in cancer survival, mortality, and incidence in seven high-income countries 1995–2014 (ICBP SURVMARK-2): a population-based study. Lancet Oncol. 2019;20(11):1493–1505. 05. doi: 10.1016/S1470-2045(19)30456-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.WHO report on the global tobacco epidemic. ; 2025. World Health Organization. [Google Scholar]
- 14.Chen W, Zheng R, Zeng H, et al. The incidence and mortality of major cancers in China, 2012. Chin J Cancer. 2016;35(1):73. doi: 10.1186/s40880-016-0137-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tomás CC, Oliveira E, Sousa D, et al. Proceedings of the 3rd IPLeiria’s International Health Congress: leiria, Portugal. 6-7 May 2016. BMC Health Serv Res. 2016;16 Suppl 3(Suppl 3):200. s12913-016-1423–5. doi: 10.1186/s12913-016-1423-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mosquera I, Todd A, Balaj M, et al. Components and effectiveness of patient navigation programmes to increase participation to breast, cervical and colorectal cancer screening: a systematic review. Cancer Med. 2023;12(13):14584–14611. doi: 10.1002/cam4.6050. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Scheffers FR, Boer JMA, Gehring U, et al. The association of pure fruit juice, sugar-sweetened beverages and fruit consumption with asthma prevalence in adolescents growing up from 11 to 20 years: the PIAMA birth cohort study. Prev Med Rep. 2022;28:101877. doi: 10.1016/j.pmedr.2022.101877. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Boffetta P, Tubiana M, Hill C, et al. The causes of cancer in France. Ann Oncol. 2009;20(3):550–555. doi: 10.1093/annonc/mdn597. [DOI] [PubMed] [Google Scholar]
- 19.Fitzmaurice C, Global Burden of Disease Cancer Collaboration . Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 2006 to 2016: a systematic analysis for the Global Burden of Disease study. J Clin Oncol. 2018;36(15_suppl):1568–1568. doi: 10.1200/JCO.2018.36.15_suppl.1568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhu H, Wang Z, Deng B, et al. Epidemiological landscape of esophageal cancer in Asia: results from GLOBOCAN 2020. Thorac Cancer. 2023;14(11):992–1003. doi: 10.1111/1759-7714.14835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chen W, Xia C, Zheng R, et al. Disparities by province, age, and sex in site-specific cancer burden attributable to 23 potentially modifiable risk factors in China: a comparative risk assessment. Lancet Glob Health. 2019;7(2):e257–69–e269. doi: 10.1016/S2214-109X(18)30488-1. [DOI] [PubMed] [Google Scholar]
- 22.Rubenstein AP, Kenneth OO, Joel H.. The Synergistic Effects of Alcohol and Tobacco Consumption on the Risk of Esophageal Squamous Cell Carcinoma: a Meta-Analysis. Am J Gastroenterol. 2014;109:822–827. doi: 10.1038/ajg.2014.71. [DOI] [PubMed] [Google Scholar]
- 23.Roy A, Roe MT, Neely ML, et al. Impact of Human Development Index on the profile and outcomes of patients with acute coronary syndrome. Heart. 2015;101(4):279–286. doi: 10.1136/heartjnl-2014-306389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Li S, Chen H, Man J, et al. Changing trends in the disease burden of esophageal cancer in China from 1990 to 2017 and its predicted level in 25 years. Cancer Med. 2021;10(5):1889–1899. doi: 10.1002/cam4.3775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Allen LN, Feigl AB.. Reframing non-communicable diseases as socially transmitted conditions. Lancet Glob Health. 2017;5(7):e644–6–e646. doi: 10.1016/S2214-109X(17)30200-0. [DOI] [PubMed] [Google Scholar]
- 26.Miranda JJ, Barrientos-Gutiérrez T, Corvalan C, et al. Understanding the rise of cardiometabolic diseases in low- and middle-income countries. Nat Med. 2019;25(11):1667–1679. doi: 10.1038/s41591-019-0644-7. [DOI] [PubMed] [Google Scholar]
- 27.Varghese JS, Stein AD.. Malnutrition among women and children in India: limited evidence of clustering of underweight, anemia, overweight, and stunting within individuals and households at both state and district levels. Am J Clin Nutr. 2019;109(4):1207–1215. doi: 10.1093/ajcn/nqy374. [DOI] [PubMed] [Google Scholar]
- 28.Wolfenden L, McCrabb S, Barnes C, et al. Strategies for enhancing the implementation of school-based policies or practices targeting diet, physical activity, obesity, tobacco or alcohol use. Cochrane Database Syst Rev. 2022. Aug 29;8:CD011677. doi: 10.1002/14651858.CD011677.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Global Burden of Disease study 2021 data resources are available online from the Global Health Data Exchange (GHDx) query tool (https://vizhub.healthdata.org/gbd-results/).