Global, regional, and national burden trends of lip and oral cavity cancer among individuals aged 60 and above from 1990 to 2021: a systematic analysis based on the 2021 global burden of disease data

Abstract

Background

Lip and oral cavity cancer (LOCC) is one of the most common malignancies among the elderly, facing challenges such as insufficient early diagnosis, difficulties in late-stage treatment, and a propensity for metastasis and recurrence, which contribute to poor prognoses. However, there remains a lack of systematic analyses regarding the global, regional, and national burden of LOCC specifically within the elderly population.

Methods

This study utilized data from the Global Burden of Disease (GBD) 2021 database, including age-standardized incidence rates (ASIR), age-standardized prevalence rates (ASPR), age-standardized mortality rates (ASMR), and age-standardized disability-adjusted life years (ASDR). A comprehensive analysis was conducted on the burden of LOCC among the elderly across the globe, five social development index (SDI) regions, 21 GBD regions, and 204 countries, examining current status, trends, future projections, and attributable risk factors.

Results

From 1990 to 2021, the burden of LOCC among the elderly significantly increased, although the growth trend has slowed in recent years. By 2021, the ASIR for LOCC was 23.13 per 100,000, the ASPR was 72.19 per 100,000, the ASMR was 12.57 per 100,000, and the ASDR was 253.10 per 100,000 years. With increasing SDI, both ASIR and ASPR continued to rise, while ASMR and ASDR exhibited a gradual decline. The absolute and relative health inequalities between different countries have gradually diminished. The burden of disease increased with age for both males and females, with males experiencing a higher burden than females. Projections indicate that from 2022 to 2050, the ASR for LOCC will continue to rise. Major risk factors identified for LOCC include chewing tobacco, high alcohol consumption, and smoking.

Conclusion

Since 1990, the burden of LOCC among the elderly has consistently increased, and this trend is expected to continue in the future, despite a slowdown in the rate of increase in recent years. Furthermore, the burden of this disease exhibits a degree of inequality based on age, gender, and region.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-025-14768-8.

Introduction

Lip and oral cavity cancer (LOCC) is a common malignancy that originates in the oral cavity and surrounding areas, including the lips, buccal mucosa, gums, floor of the mouth, hard palate, and tongue. The primary risk factors for LOCC include smoking, excessive alcohol consumption, betel quid chewing, and infection with high-risk human papillomavirus (HPV). As one of the top ten most prevalent cancers globally, LOCC poses a significant threat to public health [1, 2]. Despite notable advancements in comprehensive treatment modalities such as surgery, radiotherapy, and chemotherapy, which have partially improved patient outcomes, LOCC is often diagnosed at advanced stages due to a lack of effective early screening techniques and delays in clinical recognition, resulting in limited treatment efficacy [3–5]. Patients diagnosed at later stages frequently experience physiological impairments, including difficulties in swallowing, speech, and eating, which significantly affect their quality of life; the five-year survival rate for these patients is less than 20%, markedly lower than the 80% survival rate for those diagnosed at stage I [1]. Even after completing initial treatment, the local recurrence rate remains high [5]. Therefore, enhancing early detection capabilities and strengthening preventive measures are crucial for reducing the burden of LOCC, with epidemiological research providing an essential foundation for cancer prevention and control strategies [6].

In recent years, several studies have analyzed global trends in LOCC using data from the Global Burden of Disease (GBD) database, revealing variations in burden across different regions and populations. However, these studies predominantly focus on all age groups and lack systematic assessments of specific vulnerable populations, particularly the elderly [7]. Furthermore, most research has concentrated on specific regions (such as Europe [8], South Asia and Southeast Asia [9], China [10], India [11], and Iran [12]), or utilized earlier data, making it challenging to comprehensively reflect the current dynamics of LOCC burden [7].

As global population aging accelerates, elderly LOCC patients commonly experience decreased treatment tolerance due to physiological declines in liver and kidney function, as well as bone marrow reserves. They also face a heavy burden of comorbidities, such as cardiovascular diseases and diabetes, which affect treatment decisions and implementation while increasing the risk of complications. Additionally, age-related immune function decline may impact tumor development and treatment outcomes. These factors collectively contribute to lower survival rates, with elderly patients often having poorer prognoses than younger patients, even when staged similarly [10]. Elderly populations, particularly in low- and middle-SDI regions or among socioeconomically disadvantaged groups, encounter greater barriers to accessing healthcare services, such as financial constraints, transportation difficulties, and lack of caregiving support. This results in low screening participation, delayed diagnoses, and poor treatment adherence. Vulnerability differences within the elderly population, such as among the very old or those living alone, also warrant attention [13]. Despite the increasing number of systematic reviews related to LOCC based on GBD data, and some studies including elderly populations in their age-stratified analyses, there remains a lack of in-depth, systematic assessments of LOCC burden specifically for individuals aged 60 and older. This is particularly evident in existing research that often encompasses all age groups, making it difficult to fully reveal the unique epidemiological patterns (such as the steep increase in burden with age and persistent gender differences) and health inequalities (across Sociodemographic Index (SDI) regions, between countries, and within the elderly population). Notably, some studies have found that the disease burden observed in individuals over 60 is significantly higher than in other age groups; however, these studies have not thoroughly reported on the burden patterns specific to the elderly. Furthermore, there is currently a lack of dynamic perspectives that combine long-term historical trends (over 30 years) with future projections (up to 2050) [14, 15]. Therefore, the primary objective of this study is to quantify the long-term dynamic trends of LOCC burden in the elderly, compare health inequality differences across various SDI regions, and assess the mechanisms by which age and period effects influence changes in burden. Additionally, by analyzing the spatiotemporal evolution of gender differences in LOCC burden within the elderly population, this study aims to identify high-burden countries/regions and their core driving factors, as well as to predict the trajectory of disease burden from 2022 to 2050. This research will fill the current gap in systematic assessments of elderly LOCC, providing an evidence-based foundation for developing targeted prevention strategies for this population.

Methods

Data sources and extraction

This study is based on data from the Global Burden of Disease Study 2021 (GBD 2021), one of the most authoritative databases in global health research. GBD 2021 includes data from 204 countries and territories worldwide, covering more than 370 diseases, injuries, and associated risk factors over the period from 1990 to 2021. All data were downloaded and processed through the official GBD data visualization platform (https://vizhub.healthdata.org/gbd-results/), ensuring traceability and comprehensive coverage [16]. The study population was limited to individuals aged 60 years and older. This age threshold was set according to the World Health Organization standard and aligns with age group structures used in many published studies. The elderly population was stratified by five-year age intervals and further divided by sex [17, 18]. From a geographic perspective, the data were categorized into five groups based on the SDI: low, low-middle, middle, high-middle, and high SDI regions. The analysis also included 21 global regions and all 204 individual countries and territories [19, 20].

Statistical analysis

To systematically assess the disease burden of LOCC in the elderly and its temporal and spatial trends, we applied a series of multi-level statistical methods. First, to eliminate the confounding effect of varying population age structures across countries and regions, and to ensure fair comparisons over time and space, we performed age-standardization on incidence, prevalence, mortality, and disability-adjusted life years (DALYs) using the standard population provided by GBD 2021. This yielded age-standardized rates (ASIR, ASPR, ASMR, and ASDR), which are essential for accurately describing and comparing burden levels and trends at the global, regional, and national levels. Second, to quantify the overall long-term trends in disease burden from 1990 to 2021, we calculated the estimated annual percentage change (EAPC). This method applies a linear regression to the log-transformed age-standardized rates, estimating the slope (β) as a single, robust indicator of the average annual rate of change during the study period. Third, recognizing that long-term trends may contain inflection points—due to factors such as policy shifts or advances in medical technology—that a single EAPC value may mask, we used the Joinpoint regression model. This model automatically detects statistically significant inflection points in the time series, segments the full period into distinct linear trends, and calculates the annual percentage change within each segment. This approach helps to uncover dynamic trend characteristics, such as acceleration, deceleration, or reversals, and the timing of these changes [21, 22]. Fourth, to dissect the underlying drivers of the observed temporal trends in disease burden, we constructed an age–period–cohort (APC) model. This model isolates the independent effects of age (risk changes with aging), period (factors affecting all age groups in a given year, such as healthcare improvements), and cohort (shared exposures among birth cohorts). We applied orthogonal decomposition to separate linear and nonlinear effects of the three factors, and used weighted least squares (WLS) estimation in R to derive model parameters [23]. Fifth, to objectively quantify absolute and relative inequalities in the LOCC burden (measured by DALY rates) across countries and regions with different levels of sociodemographic development, we calculated the Slope Index of Inequality (SII) and the Concentration Index. The SII was derived from regressing DALY rates against the continuous SDI values, reflecting the absolute difference between the highest and lowest SDI groups. The Concentration Index was computed by integrating the cumulative distribution of DALYs with the population ranked by SDI, capturing the extent of relative inequality and disproportionate concentration of burden [24]. Sixth, to preliminarily explore the direction and strength of associations between national SDI levels and each age-standardized burden metric, we conducted Spearman rank correlation analyses. The results were adjusted using the Benjamini–Hochberg false discovery rate method to control for multiple testing risk. Lastly, to forecast future LOCC burden trends while accounting for uncertainty inherent in APC modeling, we employed a Bayesian age–period–cohort (BAPC) model. This model integrates historical age, period, and cohort effects for probabilistic inference and utilizes the integrated nested Laplace approximation (INLA) for efficient Bayesian parameter estimation [25]. All statistical inferences regarding trend changes in this study were based on a significance level of α = 0.05.

Results

Global, regional, and national burden

Compared to 1990, the ASIR of LOCC in the elderly globally in 2021 was 23.13 per 100,000, representing an increase of 17.65%; ASPR was 72.19 per 100,000, rising by 32.68%; ASMR was 12.57 per 100,000, showing a 2.11% increase; and ASDR was 253.10 per 100,000 person-years, with a 0.64% increase. Based on these estimates, in 2021, there were 249,668.20 new cases of LOCC in the elderly, 789,771.00 prevalent cases, 134,429.65 deaths, and a loss of 2,762,875.97 health years. The EAPC results show that from 1990 to 2021, both ASIR and ASPR of elderly LOCC showed a significant upward trend, ASMR showed no significant change, and ASDR decreased significantly (see Table 1 and Supplementary Fig. 1A-D).

Table 1.

ASR and EAPC for global, 5 SDI regions, and 21 GBD regions in 1990 and 2021

	ASIR			ASPR			ASMR			ASDR
	1990 (per 100,000 population, 95% UI)	2021 (per 100,000 population, 95% UI)	EAPCs (95% CI)	1990 (per 100,000 population, 95% UI)	2021 (per 100,000 population, 95% UI)	EAPCs (95% CI)	1990 (per 100,000 population, 95% UI)	2021 (per 100,000 population, 95% UI)	EAPCs (95% CI)	1990 (per 100,000 population, 95% UI)	2021 (per 100,000 population, 95% UI)	EAPCs (95% CI)
Global	19.66(18.42,20.69)	23.13(20.90,24.86)	0.51(0.44,0.59)	54.41(51.55,56.78)	72.19(66.03,77.03)	0.99(0.91,1.06)	12.31(11.44,13.09)	12.57(11.24,13.59)	0.01(−0.71,0.73)	251.49(235.59,266.94)	253.10(228.19,273.06)	−0.07(−0.12,−0.01)
SDI level
High SDI	24.59(22.98,25.86)	26.71(24.12,28.56)	0.35(0.26,0.43)	98.15(92.33,103.08)	118.35(108.40,125.60)	0.75(0.69,0.82)	9.90(9.22,10.37)	8.65(7.73,9.25)	−0.45(−1.30,0.41)	204.01(192.39,213.74)	172.45(158.32,182.98)	−0.52(−0.62,−0.42)
High-middle SDI	15.76(14.70,16.72)	18.17(16.10,20.09)	0.43(0.35,0.51)	42.31(39.15,45.60)	61.53(54.36,68.42)	1.29(1.22,1.36)	9.70(9.06,10.24)	8.52(7.57,9.37)	−0.55(−1.58,0.49)	197.42(185.87,207.92)	172.38(154.59,189.26)	−0.58(−0.66,−0.51)
Middle SDI	14.67(13.44,15.89)	19.78(17.39,22.11)	0.90(0.81,0.99)	27.73(25.47,29.91)	52.55(46.24,58.85)	2.04(1.91,2.17)	11.90(10.87,12.91)	12.29(10.83,13.68)	−0.05(−0.75,0.66)	232.51(213.32,251.90)	241.29(213.84,268.17)	−0.01(−0.07,0.05)
Low-middle SDI	25.08(21.57,28.70)	31.01(26.82,35.10)	0.59(0.51,0.68)	40.39(34.91,45.93)	61.50(53.62,69.33)	1.29(1.16,1.42)	22.03(18.91,25.30)	24.26(20.97,27.47)	0.28(−0.14,0.69)	450.96(389.24,517.08)	484.52(418.80,548.92)	0.15(0.08,0.21)
Low SDI	18.48(15.52,21.44)	22.42(18.74,26.38)	0.48(0.37,0.59)	28.28(23.78,32.90)	40.28(33.55,47.65)	1.01(0.86,1.16)	16.56(13.99,19.26)	18.60(15.60,21.83)	0.29(−0.09,0.68)	337.84(285.55,391.99)	367.29(307.70,431.51)	0.12(0.04,0.21)
21 regions
Andean Latin America	6.93(5.68,8.47)	8.20(6.25,10.66)	0.62(0.45,0.79)	12.07(9.82,14.84)	21.01(15.77,27.69)	2.00(1.83,2.18)	5.90(4.85,7.19)	5.46(4.18,6.96)	−0.18(−0.30,−0.06)	110.38(90.41,134.76)	100.36(76.47,128.83)	−0.27(−0.42,−0.12)
Australasia	35.37(29.58,42.08)	34.32(27.50,41.91)	−0.16(−0.43,0.11)	155.93(130.32,186.44)	179.85(144.00,221.59)	0.43(0.20,0.66)	10.51(9.07,11.96)	8.05(6.73,9.38)	−0.85(−1.61,−0.08)	219.52(189.57,251.87)	160.81(135.89,188.61)	−1.02(−1.34,−0.69)
Caribbean	21.98(19.45,24.69)	20.42(16.87,24.30)	0.02(−0.11,0.15)	51.57(45.28,58.51)	58.58(48.07,70.16)	0.78(0.62,0.94)	15.38(13.76,17.20)	12.64(10.49,14.95)	−0.41(−1.24,0.43)	292.93(261.99,328.22)	247.49(204.47,293.73)	−0.28(−0.42,−0.15)
Central Asia	11.85(10.53,13.20)	11.93(10.38,13.65)	0.16(−0.12,0.45)	26.00(23.14,28.97)	30.21(26.20,34.51)	0.69(0.29,1.10)	9.32(8.24,10.41)	8.42(7.35,9.63)	−0.27(−1.18,0.66)	194.61(175.12,215.27)	172.31(150.32,197.47)	−0.31(−0.51,−0.11)
Central Europe	17.10(15.85,18.32)	25.26(22.65,27.73)	1.31(1.24,1.38)	41.81(38.63,45.11)	86.17(76.79,95.09)	2.43(2.35,2.50)	12.41(11.49,13.26)	14.11(12.76,15.36)	0.42(−0.82,1.67)	249.54(232.21,266.38)	301.29(273.60,328.26)	0.63(0.55,0.70)
Central Latin America	10.31(9.58,10.92)	9.52(8.29,10.73)	−0.44(−0.55,−0.32)	19.73(18.35,21.03)	24.78(21.49,28.14)	0.60(0.50,0.71)	8.30(7.73,8.78)	6.30(5.49,7.07)	−1.00(−1.40,−0.60)	151.39(142.13,159.68)	116.71(102.26,131.33)	−1.01(−1.13,−0.89)
Central Sub-Saharan Africa	9.94(7.26,13.59)	11.10(7.86,15.23)	0.33(0.17,0.49)	14.65(10.79,19.94)	19.35(13.81,26.11)	0.90(0.65,1.15)	9.22(6.68,12.57)	9.70(6.74,13.47)	0.19(−0.21,0.59)	181.42(133.11,246.20)	189.95(134.52,258.87)	0.14(0.03,0.25)
East Asia	8.51(7.29,9.75)	14.28(11.44,17.49)	1.96(1.78,2.15)	16.21(13.99,18.46)	44.90(36.22,54.86)	3.70(3.54,3.86)	6.74(5.75,7.73)	7.00(5.63,8.52)	−0.11(−1.18,0.97)	129.90(110.58,149.62)	134.56(107.82,165.01)	0.26(0.12,0.40)
Eastern Europe	17.04(15.84,18.44)	23.09(20.63,25.69)	0.75(0.62,0.89)	35.16(32.16,38.83)	58.22(51.63,65.74)	1.59(1.41,1.77)	11.05(10.36,11.80)	11.69(10.48,12.89)	−0.27(−1.65,1.14)	238.48(224.39,255.13)	259.24(232.21,287.00)	−0.09(−0.28,0.09)
Eastern Sub-Saharan Africa	15.91(13.50,18.54)	16.76(13.86,19.67)	0.06(−0.02,0.13)	24.21(20.64,28.16)	30.15(24.78,35.70)	0.63(0.48,0.78)	14.43(12.27,16.86)	14.23(11.77,16.70)	−0.10(−0.66,0.48)	294.87(251.53,344.98)	284.97(234.31,336.14)	−0.21(−0.25,−0.16)
High-income Asia Pacific	14.71(13.16,16.19)	21.82(17.83,25.45)	1.19(0.87,1.51)	40.47(35.89,45.42)	68.33(55.99,80.69)	1.84(1.54,2.14)	5.72(5.22,6.04)	7.02(5.86,7.75)	0.30(−0.39,1.00)	109.89(102.03,115.77)	123.16(106.59,134.58)	0.06(−0.30,0.43)
High-income North America	34.74(32.31,36.53)	30.71(27.88,32.71)	−0.41(−0.53,−0.30)	161.54(151.35,169.57)	156.09(143.59,165.49)	−0.09(−0.17,−0.00)	11.03(10.23,11.52)	8.12(7.32,8.63)	−1.06(−1.84,−0.28)	232.31(219.05,242.21)	163.81(151.01,173.70)	−1.13(−1.33,−0.93)
North Africa and Middle East	5.35(4.40,6.34)	6.82(5.76,7.89)	0.85(0.79,0.92)	10.33(8.56,12.12)	19.11(16.11,22.35)	2.04(1.96,2.13)	4.28(3.50,5.08)	4.19(3.55,4.82)	0.03(−0.36,0.41)	80.14(66.08,94.75)	77.26(65.69,88.98)	−0.11(−0.15,−0.07)
Oceania	8.39(6.19,10.88)	9.82(7.36,12.79)	0.67(0.57,0.78)	14.52(10.79,18.71)	18.62(14.17,23.93)	0.91(0.77,1.05)	7.17(5.23,9.39)	8.16(6.01,10.75)	0.61(0.18,1.04)	134.90(97.85,177.32)	154.56(113.56,204.15)	0.64(0.55,0.73)
South Asia	36.30(31.54,41.06)	44.00(37.68,49.86)	0.43(0.30,0.57)	59.12(51.84,66.52)	91.39(78.57,103.31)	1.25(1.06,1.43)	31.65(27.43,35.94)	33.38(28.68,37.89)	0.05(−0.40,0.51)	650.62(567.94,735.86)	670.09(575.67,760.13)	−0.09(−0.19,0.02)
Southeast Asia	18.17(15.57,21.00)	21.67(17.99,25.59)	0.39(0.34,0.45)	36.68(31.31,42.61)	57.68(47.44,68.76)	1.26(1.16,1.36)	14.64(12.55,16.92)	14.69(12.22,17.18)	−0.15(−0.51,0.21)	277.78(238.98,320.72)	271.60(228.25,316.27)	−0.24(−0.30,−0.18)
Southern Latin America	13.46(11.53,15.59)	12.69(10.72,14.80)	0.12(−0.10,0.34)	31.79(26.96,37.57)	38.44(32.38,45.03)	0.91(0.66,1.16)	8.96(7.79,10.21)	7.05(6.01,8.10)	−0.26(−1.12,0.62)	181.92(158.51,208.30)	139.26(119.82,160.16)	−0.54(−0.76,−0.32)
Southern Sub-Saharan Africa	16.25(11.74,20.37)	17.57(15.34,19.94)	0.12(−0.03,0.28)	30.29(21.62,38.31)	35.95(31.19,41.11)	0.40(0.31,0.48)	13.17(9.58,16.46)	13.29(11.66,15.03)	−0.01(−0.89,0.89)	263.55(189.76,329.61)	270.28(237.46,306.21)	−0.05(−0.28,0.19)
Tropical Latin America	16.44(15.00,17.79)	17.35(15.49,18.98)	0.19(0.11,0.27)	32.00(29.27,34.70)	45.03(40.57,49.25)	1.09(1.02,1.16)	13.04(11.87,14.06)	11.54(10.27,12.56)	−0.30(−1.19,0.59)	255.29(234.82,274.90)	232.65(210.74,252.12)	−0.28(−0.37,−0.19)
Western Europe	24.14(22.07,26.20)	26.71(23.79,29.19)	0.47(0.36,0.57)	93.50(85.66,101.92)	125.66(112.63,137.38)	1.14(1.06,1.22)	10.92(10.05,11.63)	8.80(7.80,9.51)	−0.51(−1.48,0.47)	227.64(211.23,243.04)	179.39(162.59,193.56)	−0.69(−0.83,−0.55)
Western Sub-Saharan Africa	4.77(3.89,5.64)	6.23(5.15,7.44)	0.87(0.83,0.91)	7.37(5.99,8.71)	10.97(8.99,13.24)	1.29(1.19,1.39)	4.28(3.49,5.05)	5.25(4.38,6.24)	0.61(0.14,1.09)	83.51(68.03,98.33)	100.89(83.17,120.97)	0.62(0.56,0.67)

Among the five SDI regions, in 2021, the Low-middle SDI region had the highest ASIR (31.01 per 100,000), ASMR (24.26 per 100,000), and ASDR (484.52 per 100,000 person-years), while the High SDI region had the highest ASPR (118.35 per 100,000). In contrast, the High-middle SDI region had the lowest ASIR (18.17 per 100,000) and ASDR (172.38 per 100,000 person-years), the Low SDI region had the lowest ASPR (40.28 per 100,000), and the High SDI region had the lowest ASMR (8.65 per 100,000). From 1990 to 2021, ASIR and ASPR significantly increased in all five SDI regions, with no significant changes in ASMR. For ASDR, it significantly decreased in the High SDI and High-middle SDI regions, showed no significant change in the High-middle SDI region, and increased significantly in the Low-middle SDI and Low SDI regions (see Table 1 and Supplementary Fig. 1A-D).

Among the 21 GBD regions, South Asia had the highest ASIR (44.00 per 100,000), ASMR (33.38 per 100,000), and ASDR (670.09 per 100,000 person-years), while Australasia had the highest ASPR (179.85 per 100,000). In contrast, Western Sub-Saharan Africa had the lowest ASIR (6.23 per 100,000) and ASPR (10.97 per 100,000), and North Africa and the Middle East had the lowest ASMR (4.19 per 100,000) and ASDR (77.26 per 100,000 person-years). From 1990 to 2021, ASIR significantly increased in 13 regions, with no significant changes in the remaining regions. ASPR significantly increased in 20 regions, with a significant decrease in High-income North America. ASMR showed no significant changes in 15 regions, decreased significantly in 4 regions, and increased significantly in 2 regions. ASDR significantly decreased in 12 regions, increased significantly in 5 regions, and showed no significant changes in 4 regions (see Table 1 and Supplementary Fig. 1A-D).

At the national level, in 2021, Palau had the highest ASIR (107.30 per 100,000), ASPR (254.45 per 100,000), ASMR (77.28 per 100,000), and ASDR (1363.51 per 100,000 person-years), while Sao Tome and Principe had the lowest ASIR (0.85 per 100,000), ASPR (1.84 per 100,000), ASMR (0.66 per 100,000), and ASDR (11.93 per 100,000 person-years) (Figs. 1A-D and Supplementary Table 1). From 1990 to 2021, ASIR significantly increased in 131 countries, showed no significant changes in 32 countries, and decreased significantly in the remaining countries. ASPR significantly increased in 176 countries, showed no significant changes in 14 countries, and decreased significantly in the remaining countries. ASMR significantly increased in 45 countries, showed no significant changes in 115 countries, and decreased significantly in the remaining countries. ASDR significantly increased in 79 countries, showed no significant changes in 24 countries, and decreased significantly in the remaining countries (Figs. 1E-H and Supplementary Table 1).

Fig. 1.

Global ASR and EAPC of LOCC in individuals aged ≥ 60 years across 204 countries in 2021. A Spatial distribution of ASIR; (B) Spatial distribution of ASPR; (C) Spatial distribution of ASMR; (D) Spatial distribution of ASDR; (E) EAPC trend of ASIR; (F) EAPC trend of ASPR; (G) EAPC trend of ASMR; (H) EAPC trend of ASDR

Age-sex-time variation trends

Gender-age stratified data show that both ASIR and ASMR of LOCC in elderly males and females increase with age, while ASPR initially rises and then declines with age, and ASDR first decreases and then increases. Additionally, both the ASR and absolute numbers are higher in males than in females (Figs. 2A-D). After controlling for period and birth cohort effects, age effect analysis further confirms this trend (Supplementary Fig. 2A-D). Gender-time trend analysis indicates that from 1990 to 2021, both ASIR and ASPR gradually increased over time for both males and females, while ASMR and ASDR showed no significant changes (Supplementary Fig. 3A-D). Age-time analysis reveals a slow upward trend in ASR across all age groups (Supplementary Fig. 4A-D). After excluding age and cohort effects, period effect analysis shows that ASIR and ASPR gradually increased over time, while ASMR and ASDR exhibited a trend of first decreasing and then increasing, with a turning point occurring in 2005 (Supplementary Fig. 2E-H). Birth cohort analysis indicates that later birth years correspond to higher ASR levels, suggesting that cumulative risk increases with later birth dates (Supplementary Fig. 2I-L).

Fig. 2.

Age and sex differences in ASR of LOCC among individuals aged ≥ 60 years from 1990 to 2021. A ASIR; (B) ASPR; (C) ASMR; (D) ASDR

Fig. 3.

Joinpoint analysis of ASR trends for LOCC in individuals aged ≥ 60 years from 1990 to 2021. A ASIR; (B) ASPR; (C) ASMR; (D) ASDR

Fig. 4.

Correlation between ASR of LOCC and SDI in individuals aged ≥ 60 years across 21 GBD regions. A ASIR; (B) ASPR; (C) ASMR; (D) ASDR

Joinpoint regression further revealed the complex trends of LOCC in the elderly. From 1990 to 2021, ASIR and ASPR showed a clear upward trend, while ASMR and ASDR exhibited significant fluctuations over time. Specifically, ASIR experienced inflection points in 1995, 2004, and 2018, with no significant changes since 2018. ASPR had inflection points in 1995, 2005, and 2017, and the upward trend since 2017 has slowed, although it remains significant. ASMR showed inflection points in 1995, 2006, and 2019, with a notable downward trend since 2019. ASDR had inflection points in 1995, 2009, and 2019, and has shown a downward trend since 2019, but this change is not significant (Fig. 3A-D).

Correlation between ASR and SDI

The correlation between ASIR of LOCC in the elderly and SDI is strong, while the correlations of ASMR and ASDR with SDI are weaker, showing similar patterns across 21 regions and 204 countries (Figs. 4A-D and Supplementary Figs. 5 A-D). Specifically, ASIR and ASPR increase with rising SDI, particularly when SDI exceeds approximately 0.65, where the upward trends become more pronounced. ASMR and ASDR exhibit a slow downward trend as SDI increases. Detailed correlation coefficients and p-values are provided in Supplementary Table 2.

Fig. 5.

Inequality assessment and improvement potential of LOCC burden in individuals aged ≥ 60 years from 1990 to 2021. A SII reflecting absolute inequality; (B) Concentration index reflecting relative inequality; (C) DALY trend in 204 countries; (D) Avoidable DALY burden under the theoretical minimum risk level (improvement potential)

Health inequality analysis based on ASDR indicates no significant absolute or relative inequality between high-SDI and low-SDI countries. Specifically, the SII shows that from 1990 to 2021, the gap in ASDR between the highest and lowest SDI countries gradually narrowed, from 26.56 (95% UI: −19.76, 72.88) per 100,000 person-years to −9.46 (95% UI: −49.72, 30.79) per 100,000 person-years (Fig. 5A). This suggests that there is no significant absolute inequality between high-SDI and low-SDI countries, and the disease burden is gradually transitioning from high-SDI to low-SDI countries. Meanwhile, the concentration index indicates that the disease burden changed from 0.02 (95% CI: −0.04, 0.08) in 1990 to −0.02 (95% CI: −0.08, 0.04) in 2021 (Fig. 5B). This change reflects the absence of significant relative inequality between high-SDI and low-SDI countries, with the disease burden increasingly concentrating in low-SDI countries.

Frontier analysis results show that the disease burden (measured in DALYs) has significantly decreased in most countries (Fig. 5C). However, compared to the global best disease burden, there are still significant gaps in some countries and regions. Specifically, countries and regions such as Pakistan, Palau, Northern Mariana Islands, Seychelles, India, Namibia, Nepal, Bangladesh, Bhutan, Sri Lanka, Taiwan (Province of China), Zambia, Hungary, Kiribati, and Kenya have substantial potential for improvement, with improvement potentials ranging from 368.48 to 1502.16 (Fig. 5D). In low-SDI countries (SDI < 0.50), the countries with the smallest frontier differences include Niger, Chad, Yemen, Mali, and Somalia; while in high-SDI countries/regions (SDI > 0.85), the countries/regions with the largest frontier differences are Taiwan (Province of China), Lithuania, Belgium, Austria, and Germany.

Future projections and attributable risk factors

From 2022 to 2050, the ASIR and ASPR of LOCC in the elderly are expected to further increase, while ASMR and ASDR are projected to remain stable (Figs. 6A-D). By 2050, ASIR is expected to reach 26.67 (95% UI: 15.65, 37.68) per 100,000, ASPR is projected to be 95.42 (95% UI: 37.73, 153.12) per 100,000, ASMR is expected to be 12.58 (95% UI: 7.81, 17.34) per 100,000, and ASDR is projected to be 259.67 (95% UI: 146.16, 373.19) per 100,000 person-years (Supplementary Table 3). This implies that by 2050, there will be an additional 573,512.79 (95% UI: 336,653.26, 810,372.31) new cases of LOCC among the elderly, with a total of 2,052,079.49 (95% UI: 811,305.96, 3,292,853.02) individuals living with LOCC, and 270,438.89 (95% UI: 167,936.47, 372,941.31) deaths due to LOCC, resulting in a loss of 5,584,430.29 (95% UI: 3,143,217.44, 8,025,643.14) years of healthy life.

Fig. 6.

Projected ASR trends of LOCC in individuals aged ≥ 60 years from 2022 to 2050. A ASIR; (B) ASPR; (C) ASMR; (D) ASDR

Currently, chewing tobacco, high alcohol use, and smoking have been identified as attributable risk factors for LOCC. Globally, as well as in high-middle and middle-SDI regions, smoking accounts for the most DALYs, while high alcohol use is the leading cause of DALYs in high-SDI regions. In low-middle and low-SDI regions, chewing tobacco is the primary cause of DALYs (Supplementary Fig. 6).

Discussion

This study systematically evaluates the evolution of the disease burden of LOCC in individuals aged 60 and older globally from 1990 to 2021, highlighting the increasing public health challenge posed by LOCC among the aging population. We found that the ASIR and ASPR for LOCC in the elderly increased by 17.65% and 32.68%, respectively, during this period. This upward trend is corroborated by the dynamic analyses of EAPC and AAPC, and predictive models further suggest that ASIR and ASPR will continue to rise in the coming decades. The persistence of this trend presents substantial challenges for global elderly health management and healthcare systems. In contrast, while ASMR has shown a slight increase, the EAPC indicates that this change is not significant, whereas ASDR exhibits a clear downward trend. The divergent trends of these two metrics reflect the complex dynamic relationships among disease burden indicators. Notably, although ASMR and ASDR have not consistently increased overall, they have demonstrated significant fluctuations during the observation period, suggesting the influence of multiple factors. Particularly after 2019, both ASMR and ASDR have tended to decline, likely due to the widespread adoption of early screening methods, enhanced health education for high-risk populations, and ongoing advancements in treatment options. This phenomenon indicates that while early screening has improved detection rates and increased ASIR, standardized treatment has extended patient survival, thereby raising ASPR [26]. Nevertheless, the growth rates of various ASRs have significantly slowed in recent years, with some even showing a downward trend, suggesting that current interventions have achieved partial success. This change may stem from the implementation of preventive policies, such as smoking cessation and alcohol limitation, which have reduced exposure to high-risk behaviors globally [27]. Increased public health awareness has led to higher rates of early medical consultations, improving treatment outcomes. Additionally, the allocation of medical resources to middle- and low-income countries has narrowed the gap in treatment levels, allowing more patients to receive standardized care. Overall, while the disease burden of LOCC continues to increase, there are signs of alleviation, which carries important implications for public health policy in the context of global aging.

Significant disparities in the burden of LOCC persist across different regions, with developing countries, particularly low- and middle-income nations, exhibiting higher ASMR and ASDR compared to high-income countries. This disparity is closely linked to uneven distribution of healthcare resources, prevalent unhealthy lifestyles, and weak public health awareness. In contrast, high SDI regions demonstrate superior performance in controlling LOCC mortality rates due to timely access to healthcare services and widespread health education. Correlation analyses further support this conclusion, showing that ASMR and ASDR decline as SDI increases, while ASIR and ASPR rise. This trend may stem from the robust disease monitoring systems in high SDI countries, which facilitate higher rates of early diagnosis and improve the accuracy of incidence and prevalence data recording. Additionally, rapidly developing economies often see an increase in unhealthy behaviors, such as smoking and excessive alcohol consumption, which may contribute to regional disparities in LOCC incidence. Regional risk factors also play a crucial role in the global distribution of LOCC. For instance, smoking and alcohol consumption remain major risk factors in Europe, North America, and Latin America [28]; in Melanesia, South Asia, and Southeast Asia, the widespread practice of betel quid chewing has become a significant driver of increased LOCC burden among the elderly; while in Australia and New Zealand, prolonged exposure to intense ultraviolet radiation is one of the primary causative factors [28, 29]. Particularly in the Asia-Pacific region, the economic development associated with betel quid cultivation and sales has led to a notable increase in related cases, reflecting a direct correlation between economic activities and disease prevalence trends. Health inequality analyses indicate that despite the existing disparities in disease burden across regions, the level of diagnosis and treatment for LOCC is becoming more homogeneous globally, with no significant absolute or relative inequalities observed between high SDI and low SDI countries. However, the 15 countries identified for targeted intervention through frontier analysis, including five high-SDI countries, require significant attention. Notably, in 2021, Palau recorded the highest values across all four age-standardized rates, while Sao Tome and Principe had the lowest. The extremely high burden in Palau may be related to higher exposure to risk factors and/or comprehensive reporting due to its advanced diagnostic system, despite its high SDI level. Additionally, its small population base may amplify fluctuations in rate values. Conversely, the low burden in Sao Tome and Principe may be associated with lower diagnostic capabilities under its low SDI level, potentially leading to underreporting and reduced risk exposure. These differences highlight significant regional variations in risk factors and data reporting quality, warranting further investigation in future studies. Moreover, previous research has cautioned against interpreting these data too broadly, as estimates for smaller countries like Palau may rely on approximations rather than specific studies conducted within those nations [30].

In the age-gender-time association analysis, this study found that both ASIR and ASMR of LOCC significantly increased with age in elderly males and females, with the burden in males being significantly higher than in females. This trend is consistent with findings from analyses across all age groups. The observed difference may stem from more frequent exposure to high-risk behaviors, such as smoking and alcohol consumption, among males, as well as occupational exposure factors [31]. Data from 2019 indicated that there were approximately 1.14 billion smokers globally, with a smoking rate of 32.7% among males aged 15 and older, compared to only 6.6% among females [32], supporting the validity of this gender difference. Chen et al. also found that the disease burden of LOCC primarily affects the elderly population in their study across all age groups; however, they did not further analyze specific trends within the elderly cohort [40347073]. Our age-stratified analysis revealed that ASPR initially increases and then decreases with age, while ASDR first decreases and then increases. This trend suggests that the mortality rate of the disease remains high among older individuals, likely related to late-stage diagnosis, limited treatment options, and the decline in physiological functions in the elderly. From 1990 to 2021, both ASIR and ASPR exhibited a stable upward trend, while ASMR and ASDR showed no significant changes, reflecting gradual improvements in treatment and quality of life, particularly in early diagnosis and intervention. Further age-time analysis indicates a slow upward trend in ASR across all age groups, suggesting that the persistent burden of LOCC is closely related to the global aging trend. Notably, in high-income countries, the expanding elderly population base may be a significant driving factor for the increases in ASIR and ASPR. However, this trend is more pronounced in low-income countries, indicating that, in addition to aging, systemic issues such as inadequate healthcare resources and weak health education are key contributors to the escalating burden.

This study also assessed the global distribution of attributable risk factors for LOCC. Smoking remains the primary risk factor contributing to DALYs, as carcinogens in tobacco smoke can induce genetic mutations and disrupt cellular homeostasis, promoting malignant transformation [33]. In high-income countries, excessive alcohol consumption has become a central factor in the increase of DALYs. The metabolite acetaldehyde can cause DNA damage, inhibit repair mechanisms, and further exacerbate its carcinogenic effects by enhancing the absorption of tobacco carcinogens and generating reactive oxygen species [34, 35]. In contrast, in low- and middle-income countries, chewing tobacco is the leading risk factor, with the World Health Organization estimating that approximately 80% of the world’s smokers reside in these regions. This reflects the significant impact of regional culture and behavioral patterns on disease burden, suggesting that future public health interventions should be tailored to local cultural contexts to develop more targeted prevention strategies [36].

Based on the global distribution characteristics and trend evolution of the disease burden of LOCC, we propose the following targeted policy recommendations. At the global level, it is essential to strengthen the enforcement of tobacco and alcohol control policies, particularly in the Asia-Pacific region where chewing tobacco culture is prevalent, by promoting legislative regulation of betel nut products. In high-SDI countries, efforts should focus on mitigating the risks associated with alcohol-related LOCC by increasing alcohol taxes, restricting advertising, and promoting early screening programs (such as oral mucosal exfoliative cytology) to reduce disease burden. In low- and middle-SDI regions, priority should be given to enhancing the diagnostic capabilities of primary healthcare institutions for oral cancer, promoting low-cost screening tools (such as acetic acid/Lugol’s iodine staining for visual inspection), and reducing regional risk behaviors like chewing tobacco through community health education. For countries with unusually high burdens, such as Palau, it is recommended to conduct specialized epidemiological surveys to clarify whether the increased burden is due to genuine risk exposure or data reporting biases, providing a basis for precise interventions.

This study, based on the GBD 2021 database, integrates data from 204 countries/regions to achieve a comprehensive assessment of LOCC burden in individuals aged 60 and older. By using age-standardized rates to eliminate population structure bias and cross-validating trends with EAPC and Joinpoint regression models, we reveal the complex dynamic evolution of the disease burden. However, this study has certain limitations. First, differences in tumor anatomical classification may lead to significant bias: lip cancer (ICD-O-3 C00) may be confused with skin cancer of the lip (C44.0) in some national cancer registries. This classification heterogeneity could result in inaccurate estimates of the LOCC burden. Second, inherent biases in the GBD dataset may affect the accuracy of the results, particularly the underreporting issues commonly found in low-income countries, which may lead to an underestimation of ASR. Additionally, differences in LOCC definitions (ICD-9/ICD-10 coding conversions, varying clinical diagnostic standards) may weaken comparability across regions. Furthermore, the quality of raw data in some areas may be limited, potentially introducing estimation errors. Although the study incorporates multiple dimensions, there remains insufficient mechanistic exploration for specific populations. Future research should combine field epidemiological studies to enhance the robustness of the conclusions.

Conclusion

From 1990 to 2021, the ASIR and ASPR of LOCC in the elderly significantly increased, while changes in ASMR and ASDR were minimal, with the burden in males being higher than in females. The study highlights the impact of SDI on disease patterns, noting that while the burden remains heavy in low- and middle-income regions, mortality has decreased in high-income areas. Given that the primary risk factors are smoking, alcohol consumption, and chewing tobacco, this study suggests the need for strengthened global public health interventions and early prevention strategies to address the expected rise in disease burden due to population aging by 2050, particularly in regions with relatively limited medical resources. To effectively reduce the LOCC burden in the elderly, it is crucial to implement a comprehensive public health strategy focused on controlling tobacco/alcohol use, enhancing screening and education, and improving healthcare accessibility.

Supplementary Informatio

Supplementary Material 1. ^{(11.2MB, docx)}

Authors’ contributions

BH undertook the design, guidance and modification of the project and paper, YL completed the implementation of project and writing of the paper. All authors reviewed the manuscript.

Funding

Not applicable.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All data used in this study were obtained from the publicly available GBD 2021 database. The database contains fully de-identified data at the country/region level and does not include any individual-level or personally identifiable information. As this study used publicly accessible, anonymized aggregate data, no ethical approval or informed consent was required.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.