HPV vaccination, screening disparities, and the shifting landscape of cervical cancer burden: a global analysis of trends, inequalities, and policy implications

Abstract

Synopsis

Health inequalities intensified, burden shifting to low-resource regions despite preventive advancements. HPV and screening rates diverged by SDI, highlighting coverage gaps. Screening and vaccination inversely linked to disease burden, underscoring critical efficacy. Innovative modeling exposed disparities, advocating SDI-stratified interventions.

Objective

This study analyzes global and regional cervical cancer trends (1990–2021) across different Socio-Demographic Index (SDI) levels, highlighting health inequalities, assessing the impact of HPV vaccination and screening, and modeling future trends. The findings aim to inform targeted prevention policies, reduce regional disparities, and promote global health equity.

Methods

Data were sourced from the Global Burden of Disease study 2021(GBD 2021), OECD, and WHO. The focus was on incidence and disability-adjusted life years (DALYs) of cervical cancer. Time trends were analyzed by SDI regions, alongside health inequality assessments. Correlation analyses examined links between screening rates, HPV vaccination coverage, and disease burden.

Results

From 1990 to 2021, global age-standardized incidence rate and age-standardized DALYs rates declined significantly, with estimated annual percentage changes (EAPC) of -0.54% (95% CI: -0.63 to -0.44) and − 1.27% (95% CI: -1.36 to -1.18). However, significant differences exist in specific patterns of change across SDI regions: exhibited an upward incidence trajectory. From 1990 to 2021, the burden of cervical cancer disease shifted from developed to less developed regions. Correlation analysis showed negative associations between screening rates and DALYs (r = -0.56, p < 0.01) and between vaccination coverage and incidence (r = -0.35, p < 0.01).

Conclusion

Although the global cervical cancer burden has decreased, significant regional disparities remain. Future policies should focus on tailored interventions, with low-resource regions strengthening healthcare infrastructure and implementing minimum effective preventive measures, while high-SDI regions shift to precision public health approaches. Policymakers must also incorporate culturally sensitive health education to address social barriers, challenge misconceptions, and empower communities, ultimately reducing preventable cervical cancer morbidity.

Introduction

Cervical cancer, nearly all cases of which are linked to HPV, is the fourth most common cancer among women worldwide, posing a significant threat to women’s health. Cervical cancer is both preventable and curable, with significant global reductions in its burden achieved through widespread screening and highly effective HPV vaccination [1, 2]. Previous studies indicate that cervical cancer prevention and treatment efforts have yielded relatively optimistic outcomes, with a notable global decline in cervical cancer incidence [3–5].

However, incidence and death rates remain disproportionately high in resource-limited regions, particularly in sub-Saharan Africa, South Asia, and Latin America [6, 7]. These persistent disparities underscore global health inequalities, reflecting the profound impact that inequitable resource allocation, socioeconomic factors, and individual or healthcare system barriers have on health outcomes [6]. The World Health Organization (WHO) updated its guidelines for detecting health inequalities in December 2024⁸. These guidelines outline strategies for obtaining and utilizing health inequality monitoring data, as well as techniques for analyzing, interpreting, and reporting such data. The aim is to support the expansion and strengthening of health inequality monitoring practices across various applications worldwide, ultimately contributing to the broader goal of promoting health equity [8].

This study introduces an innovative framework integrating SDI-stratified trends, health inequality metrics, and HPV prevention efficacy to analyze cervical cancer burden across 204 countries (1990–2021). Leveraging multi-source data and predictive modeling, we aim to identify region-specific disparities and inform equitable policy design for global cervical cancer elimination.

Methods

Data sources

Global burden of disease study 2021

The GBD 2021 provides comprehensive estimates on the worldwide burden of 371 diseases and injuries across 204 countries and territories, spanning from 1990 to 2021. The GBD 2021 study incorporated data from diverse sources, including peer-reviewed literature, surveys, disease registries, and hospital inpatient records, to ensure a robust and comprehensive analysis of cervical cancer. Detailed descriptions of these data sources and their validation processes are available through the Global Health Data Exchange web tool (http://ghdx.healthdata.org/). We retrieved data on cervical cancer incidence and DALYs for all age groups across 204 countries and 21 regions spanning the years 1990 to 2021.

World health organization, global health observatory

The World Health Organization, a specialized agency of the United Nations, leads global public health efforts, including the prevention of infectious diseases, management of chronic conditions, vaccination, environmental health, and the development of health policies. To support global health initiatives, WHO routinely collects health data from its member states. For our analysis, we retrieved HPV vaccination coverage data for girls aged 9–14 years from the Global Health Observatory of the World Health Organization (https://www.who.int/data/gho) for the period 2010 to 2023. HPV vaccination coverage measures the percentage of girls aged 9–14 in target groups who received all recommended doses. These data were calculated using a statistical model that combined government records, school vaccination data, and local surveys.

OECD

The Organisation for Economic Co-operation and Development (OECD) (https://data-explorer.oecd.org/) offers comprehensive data across multiple sectors, including the economy, society, and environment. This data is derived from official statistics of 38 member countries, which are standardized for consistency and comparability. Specifically, in the health sector, the OECD provides extensive data on disease burden, medical expenditures, health status, and healthcare coverage. After evaluating two data sources (programme data and survey data), we chose cervical cancer screening data from programme data. Screening coverage was defined as the percentage of people of the same sex and age group who received screening. For this study, we obtained cervical cancer screening rate data from the OECD for member countries covering the period from 2000 to 2023.

Statistical analysis

Description analysis

Cervical cancer burden data (1990–2021) were derived from the Global Burden of Disease 2021 study, analyzed using age-standardized incidence rates (ASIR) and DALY rates across 204 countries. A global map illustrates the variations in disease burden between different countries and regions [9]. Additionally, we calculated the estimated annual percentage change (EAPC) for the five SDI regions^{10 11}and 21 GBD regions. The trend chart highlights the temporal changes in the cervical cancer burden across different SDI regions. We analyzed cervical cancer screening rate data from 33 countries and HPV vaccine coverage data from 106 countries in 2021, examining their trends relative to changes in the SDI.

Joinpoint regression analysis

The joinpoint regression model, a collection of linear statistical models, was used to assess trends in disease burdens over time. This model estimates changes in disease rates through a least-squares approach, enhancing objectivity by avoiding the reliance on single linear trends typical in conventional analysis. By calculating the sum of squared residuals between estimated and actual values, the model identifies turning points where trends shift [12, 13]. This study employed joinpoint analysis to examine temporal trends in cervical cancer burden and identify significant turning points in these trends. By integrating insights from global cervical cancer prevention policies, the analysis provided a rationale for these turning points and assessed the effectiveness and accessibility of prevention measures across different regions. Based on the immediate benefits of preventive measures on incidence and the observed temporal trends in crude incidence rate changes, we conducted a joinpoint regression analysis to assess the trend of age-standardized incidence rate over time.

Measurement of health inequality

Following WHO recommendations, we applied two primary metrics, the Slope Index of Inequality (SII) and the Concentration Index (CI), to evaluate absolute and relative income-related disparities across countries. The SII assesses absolute inequality by calculating the slope of the regression line between each country’s disease burden and its weighted rank based on income. The SII represents the disparity in disease burden between countries with the lowest and highest SDI. It quantifies the absolute difference in burden across the full spectrum of SDI rankings, providing a measure of health inequality [14].

The CI, a relative measure, evaluates the distribution of disease burden by fitting a Lorenz concentration curve to the cumulative burden and population. Ranging from − 1 to 1, the CI reveals distribution patterns, with a negative value indicating a greater concentration of cervical cancer burden in countries with lower SDI scores. In this study, we calculated the SII and the CI using the number of cases and crude incidence rate to assess both the absolute and relative health inequalities related to cervical cancer.

Bayesian age-period-cohort (BAPC)

The BAPC model provides highly accurate projections by assuming that the effects of adjacent age, period, and cohort intervals remain consistent over time. It assigns appropriate prior distributions to all unknown parameters, using a second-order random walk to smooth these effects. By integrating historical data with current observations, the Bayesian approach generates estimates that align closely with chronological health trends, offering a continuous representation of gradual changes in key indicators over time [15, 16]. This study projected age-standardized incidence trends globally and across five SDI regions through 2030, aiming to assess the effectiveness and rationality of current preventive measures [10].

Correlation analysis

In this study, we conducted Pearson correlation analysis to examine the relationships between cervical cancer crude incidence/DALYs rates and screening rates/HPV vaccine coverage across countries. Screening rate data from 23 countries with consistent records over five years (2012–2016) were correlated with cervical cancer DALYs rates from 2017 to 2021. Similarly, we analyzed HPV vaccination coverage data from 30 out of 35 countries during 2012–2016, correlating with crude cervical cancer incidence rates among females aged 15–20 years (who received vaccination at ages 9–14) between 2017 and 2021.

Statistical analysis

All analysis and visualizations were conducted using R software (version 4.4.1) and Joinpoint software (version 5.2.0). Age-standardized rates, incidence counts, and DALYs were reported with 95% uncertainty intervals (UIs), while estimates for EAPC, SII, and CI were presented with 95% confidence intervals (CIs). Statistical significance between variables was assessed using two-sided tests, with a significance level of α = 0.05. P-values less than 0.05 were considered statistically significant.

Data availability

In accordance with the journal’s guidelines, we will provide our data for independent analysis by a selected team by the Editorial Team for the purposes of additional data analysis or for the reproducibility of this study in other centers if such is requested.

Results

Globally, the incidence and DALYs of cervical cancer have declined over the past few decades, but the number of cases and DALYs have been on an upward trend. According to the GBD 2021 data, the global cases of cervical cancer increased from 409,548 cases (95% UI: 383,207 to 438,506) in 1990 to 667,426 cases (95% UI: 613,030 to 726,422) in 2021. However, the age-standardized incidence rate decreased from 18.11 (95% UI: 16.94 to 19.40) to 15.32 (95% UI: 14.08 to 16.68), with an EAPC of -0.54% (95% CI: -0.63 to -0.44). Similarly, the number of DALYs associated with cervical cancer rose globally from 7,416,287 (95% UI: 6,841,378 to 8,071,400) in 1990 to 9,911,653 (95% UI: 9,053,317 to 10,798,306) in 2021. Despite this, the age-standardized DALYs rate decreased from 330.11 (95% UI: 304.67 to 359.10) to 226.28 (95% UI: 206.51 to 246.86), with an EAPC of -1.27% (95% CI: -1.36 to -1.18) (Table 1).

Table 1.

Incidence cases, age-standardized incidence rate, estimated annual percentage change of cervical cancer, dalys, age-standardized dalys rate, estimated annual percentage change of cervical cancer (Editable table displayed after the references)

location	Incidence						DALYs (Disability-Adjusted Life Years)
	Num_1990	ASIR_1990	Num_2021	ASIR_2021	Num_change	EAPC_CI	Num_1990	Age-standardized DALY rate _1990	Num_2021	Age-standardized DALY rate _2021	Num_change	EAPC_CI
Global	409,548 (383207to438506)	18.11 (16.94to19.4)	667,426 (613030to726422)	15.32 (14.08to16.68)	0.63% (0.48to0.79)	-0.54% (-0.63to-0.44)	7,416,287 (6841378to8071400)	330.11 (304.67to359.1)	9,911,653 (9053317to10798306)	226.28 (206.51to246.86)	0.34% (0.2to0.48)	-1.27% (-1.36to-1.18)
High SDI	87,377 (84817to89249)	16.43 (16.04to16.77)	78,008 (73679to81041)	10.3 (9.91to10.66)	-0.11% (-0.14to-0.08)	-1.41% (-1.56to-1.26)	888,170 (860266to911521)	163.46 (158.83to167.8)	707,650 (664623to741914)	86.41 (82.45to90.3)	-0.2% (-0.23to-0.17)	-1.99% (-2.1to-1.88)
High-middle SDI	73,628 (68687to78720)	13.37 (12.47to14.29)	120,153 (103649to137728)	13.27 (11.44to15.16)	0.63% (0.39to0.89)	0.21% (0.12to0.3)	1,221,304 (1133350to1316829)	222.34 (206.21to239.84)	1,432,108 (1252618to1634549)	152.9 (133.88to174.32)	0.17% (0to0.35)	-1.12% (-1.17to-1.07)
Middle SDI	114,367 (105117to123880)	18.09 (16.67to19.56)	228,240 (204617to254212)	15.94 (14.3to17.75)	1% (0.76to1.27)	-0.47% (-0.54to-0.4)	2,165,690 (1982381to2359900)	342.74 (314.41to372.39)	3,167,322 (2856722to3510378)	218.95 (197.6to242.31)	0.46% (0.3to0.65)	-1.52% (-1.6to-1.45)
Low-middle SDI	84,750 (73563to97338)	22.48 (19.49to25.79)	152,720 (136721to169314)	17.79 (15.94to19.7)	0.8% (0.54to1.1)	-0.78% (-0.93to-0.64)	1,924,362 (1669283to2212532)	505.86 (438.95to583.11)	2,751,152 (2455070to3056560)	321.36 (287.11to356.77)	0.43% (0.22to0.67)	-1.48% (-1.61to-1.34)
Low SDI	48,872 (41627to59340)	34.34 (29.27to41.62)	87,633 (73984to103872)	25.47 (21.57to30.12)	0.79% (0.47to1.14)	-1.17% (-1.29to-1.05)	1,207,230 (1031665to1473493)	833.33 (711.27to1016.61)	1,843,208 (1552008to2199867)	535.11 (454.02to638.34)	0.53% (0.25to0.84)	-1.64% (-1.74to-1.54)
Andean Latin America	4154 (3640to4665)	33.19 (29.05to37.34)	9757 (7476to12359)	29.79 (22.83to37.66)	1.35% (0.81to2.04)	-0.65% (-0.81to-0.48)	83,460 (73057to94115)	667.44 (585.28to752)	140,301 (107733to176561)	431.74 (331.78to543.15)	0.68% (0.32to1.17)	-1.71% (-1.86to-1.55)
Australasia	2001 (1863to2175)	17.18 (15.98to18.67)	1752 (1576to1925)	8.51 (7.66to9.35)	-0.12% (-0.23to-0.01)	-1.92% (-2.18to-1.67)	20,693 (19368to22507)	178.43 (166.82to193.54)	13,292 (12053to14534)	61.68 (56.34to67.5)	-0.36% (-0.43to-0.28)	-3.13% (-3.39to-2.86)
Caribbean	4635 (4191to5197)	31.51 (28.52to35.21)	7382 (6192to8712)	27.58 (23.09to32.63)	0.59% (0.34to0.89)	-0.45% (-0.52to-0.37)	82,426 (72179to94674)	561.02 (491.47to642.93)	115,954 (95882to140764)	433.17 (356.8to526.53)	0.41% (0.15to0.71)	-0.8% (-0.88to-0.72)
Central Asia	5169 (4952to5420)	18.21 (17.43to19.11)	7179 (6266to8081)	14.17 (12.38to15.92)	0.39% (0.21to0.57)	-0.45% (-0.61to-0.28)	89,742 (86415to94113)	318.71 (306.47to334.38)	108,285 (94718to122647)	213.84 (187.54to241.95)	0.21% (0.06to0.36)	-1.05% (-1.2to-0.9)
Central Europe	16,320 (15591to17021)	21.66 (20.68to22.61)	14,203 (12928to15545)	15.93 (14.45to17.53)	-0.13% (-0.21to-0.04)	-1.1% (-1.31to-0.89)	263,434 (252755to272815)	344.13 (330.25to356.19)	181,664 (166208to196950)	191.83 (175.35to209.14)	-0.31% (-0.37to-0.24)	-2.05% (-2.24to-1.87)
Central Latin America	22,434 (21858to23000)	41.85 (40.6to42.84)	40,343 (34554to46252)	28.89 (24.76to33.1)	0.8% (0.54to1.06)	-1.58% (-1.76to-1.4)	330,665 (322525to338290)	627.39 (610.2to641.61)	441,930 (380717to508808)	315.97 (272.32to363.77)	0.34% (0.14to0.54)	-2.55% (-2.71to-2.38)
Central Sub-Saharan Africa	6100 (4558to7964)	39.39 (29.59to51.35)	15,328 (10594to20549)	38 (26.28to50.98)	1.51% (0.87to2.42)	-0.17% (-0.2to-0.14)	150,919 (112943to196163)	950.65 (718.75to1233.27)	330,691 (228131to448104)	813.59 (562.84to1104.29)	1.19% (0.62to2.02)	-0.54% (-0.59to-0.49)
East Asia	61,909 (50407to76001)	12.16 (9.94to14.88)	137,864 (101144to177755)	13.4 (9.86to17.38)	1.23% (0.45to2.08)	0.73% (0.56to0.89)	1,178,714 (958919to1452779)	231.92 (189.37to285.51)	1,616,240 (1195414to2080057)	151.15 (111.8to195.41)	0.37% (-0.1to0.89)	-1.13% (-1.27to-0.99)
Eastern Europe	23,499 (22662to24259)	14.94 (14.42to15.45)	25,339 (22883to27756)	16.49 (14.8to18.1)	0.08% (-0.04to0.2)	0.31% (0.19to0.43)	378,041 (365242to390080)	239.05 (230.85to246.89)	320,236 (289146to353815)	200.77 (180.38to221.47)	-0.15% (-0.25to-0.06)	-0.7% (-0.81to-0.59)
Eastern Sub-Saharan Africa	22,644 (18816to27483)	45.69 (37.95to55.48)	41,370 (33124to52883)	33.45 (27.17to42.12)	0.83% (0.43to1.34)	-1.34% (-1.46to-1.22)	564,509 (469048to691067)	1117.94 (929.76to1367.77)	876,849 (701728to1106312)	709.49 (572.26to890.6)	0.55% (0.21to1)	-1.81% (-1.93to-1.68)
High-income Asia Pacific	12,720 (12006to13516)	11.85 (11.18to12.58)	15,578 (14044to16882)	11.03 (10.25to11.95)	0.22% (0.13to0.33)	0.02% (-0.09to0.13)	144,582 (135083to155151)	133.71 (124.78to143.46)	132,808 (119523to144704)	87.68 (81.63to94.36)	-0.08% (-0.16to0)	-1.26% (-1.35to-1.18)
High-income North America	32,695 (31769to33480)	19.65 (19.11to20.1)	30,415 (29096to31622)	12.69 (12.19to13.22)	-0.07% (-0.1to-0.03)	-1.34% (-1.55to-1.13)	223,497 (215889to230643)	133.17 (129.02to137.29)	240,100 (229390to250584)	92.68 (88.88to96.91)	0.07% (0.04to0.11)	-1.05% (-1.18to-0.93)
North Africa and Middle East	6235 (5462to7544)	5.98 (5.25to7.24)	12,913 (11008to15182)	4.72 (4.04to5.5)	1.07% (0.7to1.43)	-0.73% (-0.78to-0.69)	137,468 (119841to167282)	131.01 (113.82to159.07)	217,737 (182979to258586)	80.04 (67.65to93.68)	0.58% (0.29to0.87)	-1.59% (-1.63to-1.56)
Oceania	668 (506to928)	34.17 (26.78to47.56)	1372 (1070to2017)	27.31 (21.47to39.68)	1.05% (0.62to1.71)	-0.8% (-0.86to-0.74)	12,844 (9709to18346)	654.65 (511.97to937.93)	25,245 (19486to37774)	499.47 (390.49to739.25)	0.97% (0.52to1.63)	-0.86% (-0.91to-0.81)
South Asia	83,507 (69577to96969)	23.7 (19.53to27.5)	132,482 (114541to151295)	15.54 (13.47to17.71)	0.59% (0.29to0.95)	-1.48% (-1.79to-1.16)	1,967,172 (1640298to2286747)	552.66 (457.21to642.7)	2,432,061 (2097850to2773134)	285.98 (247.23to325.23)	0.24% (0to0.52)	-2.23% (-2.53to-1.94)
Southeast Asia	30,365 (26327to34798)	18.06 (15.61to20.64)	58,017 (49191to67747)	15.17 (12.91to17.65)	0.91% (0.6to1.28)	-0.78% (-0.88to-0.67)	613,110 (526601to699790)	362.63 (313.26to413.29)	931,501 (800257to1086328)	241.92 (207.75to281.44)	0.52% (0.27to0.8)	-1.48% (-1.58to-1.37)
Southern Latin America	6030 (5641to6407)	24.43 (22.84to25.99)	9302 (8566to10068)	22.8 (21.07to24.73)	0.54% (0.4to0.72)	-0.22% (-0.38to-0.06)	103,034 (97006to109279)	417.99 (393.15to442.82)	123,079 (114520to133155)	296.76 (276.24to321.05)	0.19% (0.09to0.31)	-1.08% (-1.22to-0.94)
Southern Sub-Saharan Africa	5528 (4618to6832)	29.89 (25.13to37.34)	16,247 (14189to18418)	42.4 (37.16to47.85)	1.94% (1.23to2.66)	1.89% (1.41to2.38)	110,032 (92065to137136)	600.46 (503.72to750.8)	300,881 (260870to338893)	788.82 (685.47to885.31)	1.73% (1.06to2.41)	1.71% (1.19to2.23)
Tropical Latin America	13,346 (12841to13859)	23.08 (22.14to23.96)	27,823 (26275to29199)	20.27 (19.16to21.27)	1.08% (0.96to1.21)	-0.85% (-1.01to-0.68)	242,802 (233832to252337)	422.16 (406.02to438.79)	395,392 (372667to414907)	285.57 (269.19to299.68)	0.63% (0.53to0.72)	-1.64% (-1.77to-1.5)
Western Europe	35,616 (34386to36706)	14.25 (13.83to14.66)	27,922 (26003to29391)	8.71 (8.25to9.11)	-0.22% (-0.26to-0.18)	-1.27% (-1.48to-1.06)	391,715 (376291to403410)	148.36 (143.69to152.94)	257,408 (240181to272440)	72.71 (68.94to76.61)	-0.34% (-0.38to-0.31)	-2.02% (-2.22to-1.82)
Western Sub-Saharan Africa	13,972 (11425to17090)	26.2 (21.4to31.65)	34,839 (26668to42613)	24.11 (18.93to29.1)	1.49% (0.93to2.12)	-0.24% (-0.28to-0.2)	327,427 (269225to399724)	608.91 (502.55to737.46)	709,998 (548746to865456)	490.75 (383.61to591.44)	1.17% (0.69to1.73)	-0.68% (-0.74to-0.62)

From the perspective of health inequalities, cervical cancer incidence and DALYs vary significantly across regions with different SDI scores. In high SDI regions, both the incidence and DALYs are the lowest, with the age-standardized incidence rate decreasing from 16.43 in 1990 to 10.30 in 2021 [an EAPC of -1.41% (95%CI: -1.56 to -1.26)] and the age-standardized DALYs rate dropping from 163.46 to 86.41 [an EAPC of -1.99% (95% CI: -2.10 to -1.88)]. In contrast, low SDI regions experience significantly higher incidence and DALYs, with the age-standardized incidence rate at 25.47 (95% UI: 21.57 to 30.12) and age-standardized DALYs rate reaching 535.11 (95% UI: 454.02 to 638.34) in 2021 (Table 1; Fig. 1.Part1).

Fig. 1.

Description Analysis. Part 1. Trends in disease burden over time in the five SDI regions (A) Crude DALYs rate trend by SDI regions (B) Crude incidence rate trend by SDI region (C) Age-standardized DALYs rate trend by SDI regions (D) Age-standardized incidence rate trend by SDI regions. Part 2. Global map of disease burden (A) Age-standardized incidence rate in 1990 (B) Age-standardized incidence rate in 2021 (C) Age-standardized DALYs rate in 1990 (D) Age-standardized DALYs rate in 2021

In both 1990 and 2021, the burden of cervical cancer was higher in regions such as sub-Saharan Africa and South Asia. In contrast, regions like the Middle East, North Africa, Europe, and North America generally experienced a lower disease burden (Fig. 1.Part2). In 1990, the lowest age-standardized incidence rate of cervical cancer was observed in North Africa and the Middle East [5.98, 95%UI (5.25 to 7.24)], while the highest rates were found in eastern sub-Saharan Africa [45.69, 95%UI (37.95 to 55.48)] (Difference:39.71). A similar pattern was seen in age-standardized DALYs rate, with the lowest rates in North Africa and the Middle East [131.01, 95% UI (113.82to159.07)], and the highest in eastern sub-Saharan Africa [1117.94, 95% UI (929.76to1367.77)] (Difference:986.93).

By 2021, the lowest age-standardized incidence rate remained in North Africa and the Middle East [4.72, 95%UI (4.04 to 5.50)], while the highest were found in southern sub-Saharan Africa [42.40, 95%UI (37.16 to 47.85)] (Difference: 37.68). In terms of age-standardized DALYs rate, the lowest rates were observed in Australia [61.68, 95%UI (56.34 to 67.50)], while the highest were in central sub-Saharan Africa [813.59, 95%UI (562.84 to 1104.29)] (Difference:751.91) (Fig. 1.Part1).

Time trend analysis of cervical cancer disease burden data from 1990 to 2021 revealed distinct patterns across different SDI regions. In middle SDI and high-middle SDI areas, the crude incidence rate showed a clear upward trend, while in low SDI and high SDI areas, the crude incidence rate decreased significantly. It is worth noting that the crude incidence rate initially decreased before rising again, resulting in an overall upward trend in low-middle SDI areas. In contrast, the crude DALYs rate in low-middle SDI, middle SDI, and high-middle SDI areas did not show significant changes. Regarding age-standardized DALYs rate, all SDI regions exhibited a downward trend. For age-standardized incidence rate, low and high SDI regions experienced significant declines, whereas middle-high SDI regions exhibited a notable increase, with stable trends observed in low-middle and middle SDI regions (Fig. 1.Part1).

We divided the global data into five groups based on SDI. The high-middle SDI region exhibited distinctly different trends compared to the other regions. During the study period, there were five joinpoints, indicating changes in the trend line at five distinct points. The periods of increasing trends were from 1998 to 2002(APC: 0.33, 95%CI: -0.30 to 0.96), 2002 to 2005(APC: 1.99, 95%CI:0.55 to 3.46) and 2005 to 2018(APC:0.16, 95%CI:0.05 to 0.27). In contrast, the periods of decreasing trends were from 1994 to 1998(APC: -1.27, 95%CI: -1.91 to -0.62) and from 2018 to 2021(APC: -1.12, 95%CI: -2.74 to 0.51) (Fig. 2).

Fig. 2.

Joinpoint analysis of Age-standardized incidence rate (A) Global (B) High SDI (C) High-middle SDI (D) Middle SDI (E) Low-middle SDI (F) Low SDI

Both the high SDI and low SDI regions show similar trends, with four joinpoints in each. In both regions, all time periods indicate a downward trend in the age-standardized incidence rate of cervical cancer. The middle-low SDI region has five joinpoints, with an upward trend observed only between 2011 and 2017(APC: 0.87, 95%CI: 0.62 to 1.12). In all other periods, the trend is downward. The middle SDI region also has five joinpoints, showing an upward trend between 2001 to 2004(APC: 0.16, 95%CI: -0.42 to 0.74) and 2013 to 2018(APC: 0.7, 95%CI: 0.42 to 0.97), with a downward trend in the other time periods (Fig. 3).

Fig. 3.

BAPC analysis of Age-standardized incidence rate (A) Global (B) High SDI (C) High-middle SDI (D) Middle SDI (E) Low-middle SDI (F) Low SDI

The Bayesian Age-Period-Cohort (BAPC) model was employed to predict the global incidence of cervical cancer from 2022 to 2030. The results indicate that the global incidence will decrease from 15.34 (95%CI: 15.30 to 15.37) in 2021 to 14.82 (95%CI: 12.74 to 16.90) in 2030. Under current preventive measures, the incidence rates in high-SDI and high-middle SDI regions are expected to decline at a faster pace. In contrast, the decline in the other three SDI regions will be slower, with the middle-SDI region even projected to experience a slight increase during this period. (Fig. 3)

In cervical cancer incidence, we observed significant absolute and relative inequalities associated with SDI. The distribution of incidence rates across different SDI regions has shifted notably over time. In 1990, the SII was 1.13(95%CI: -3.02 to 5.29), indicating that the country with the highest SDI had a cervical cancer incidence rate 1.13 higher than that of the country with the lowest SDI. By 2021, this difference had reversed, with the SII reaching − 4.74(95%CI: -9.61 to 0.12). Additionally, the CI showed an upward trend from 1990 to 2019, reflecting increasing disparities in cervical cancer burden between countries with differing SDI levels. The CI was 0.01(95%CI: -0.07 to 0.09) in 1990 and 0.06(95%CI: -0.03 to 0.15) in 2021. (Fig. 4)

Fig. 4.

Analysis of health inequalities based on global Age-standardized incidence rate (A) Slope Index of Inequality (B) Concentration Index

We plotted trends in screening rates and HPV vaccination coverage as a function of SDI. Among the 33 countries included in 2021, screening rates increased with increasing SDI. Among the 106 countries analyzed in 2021, countries with middle SDI and high-middle SDI exhibited unusually lower HPV vaccination coverage compared to those with high and low-middle SDI. (Fig. 5.Part1)

Fig. 5.

Analysis of screening rate and HPV vaccination coverage. Part 1. Screening rate and HPV vaccination coverage rate change with SDI (A) Screening rate (B) HPV vaccination coverage. Part 2. Correlation analysis between preventive measures and disease burden (A) Screening rate and Crude DALYs rate (B) HPV vaccination coverage and Crude incidence rate

Pearson correlation analysis illustrates a significant negative correlation between the screening rate and cervical cancer DALYs rates (correlation coefficient r = -0.56, p < 0.01). The data show that as screening rates increase, cervical cancer DALYs rates decrease, particularly in areas with lower coverage. However, the decline in DALYs rates begins to slow as screening rates increase, indicating a diminishing marginal effect of screening in areas with very high coverage. Pearson correlation analysis revealed a significant negative correlation between HPV vaccination coverage and cervical cancer incidence rates (correlation coefficient r = -0.35, p < 0.01). This indicates that higher HPV vaccination coverage is associated with a lower incidence of cervical cancer, supporting the effectiveness of the HPV vaccine in reducing cervical cancer risk [2]. (Fig. 5.Part2)

Discussion

Over the past three decades, the global ASIR and age-standardized DALYs rate of cervical cancer have significantly declined. This reflects advancements in preventive measures, such as improved screening programs and expanded HPV vaccination coverage, particularly in high SDI regions [5]. However, the absolute number of cases and DALYs has increased, primarily due to population growth and aging. This discrepancy highlights the urgent need to scale up prevention strategies globally, with a focus on regions that continue to bear a disproportionately high burden of cervical cancer [17].

While the global ASIR and DALYs rates of cervical cancer have shown an overall downward trend, significant differences exist in specific patterns of change across SDI regions. In middle-high SDI regions, the incidence rate has generally increased. Middle-high SDI regions may be in a “transition period” of vaccine promotion. Governments may prioritize the procurement of bivalent/quadrivalent vaccines, but the public delays vaccination due to preference for nine-valent vaccines, resulting in lower than expected vaccination coverage. This finding is consistent with the unexpectedly low vaccination coverage in middle-high SDI regions observed in this study. Additionally, factors such as inadequate follow-up management after screening and economic development priorities that favor high-return sectors may contribute to the rising incidence.

The growth rates in middle-high SDI regions varied significantly across periods. From 1998 to 2018, the age-standardized incidence rate showed an upward trend. Notably, the growth rate was particularly high between 2002 and 2005, which may be attributed to the introduction of high-risk human papillomavirus (HR-HPV) testing as a cervical cancer screening method around 2000. Although advances in detection technology improved sensitivity, they also brought challenges associated with the detection paradox. Higher sensitivity may identify non-progressive or low-risk lesions that are unlikely to pose a threat during a patient’s lifetime, thereby increasing the likelihood of false positives. In other periods, slower growth rates may be linked to increased disease exposure risks driven by rapid economic development, such as delayed childbirth and increased contraceptive use. Furthermore, this slowdown in growth may reflect the inability to expand public health resources during this period to adequately meet the growing demand for cervical cancer prevention and treatment services.

From a health economics perspective, the single-test cost of HPV screening has been shown to be significantly higher than traditional cytological methods. The transition in screening strategies, while improving detection sensitivity, creates dual economic constraints: it involves both initial investments in equipment acquisition and technician training, and requires bearing additional diagnostic and treatment costs triggered by elevated false-positive rates. This economic impact exerts crucial influence on public health policy formulation - during the initial implementation of HR-HPV primary screening protocols, the lack of region-specific empirical support for the cost-effectiveness threshold recommended by the World Health Organization has led to differentiated decision-making mechanisms based on financial risk considerations among nations. These divergent policy approaches may potentially explain the varying trends in cervical cancer incidence rates across regions with different SDI levels.

In other SDI regions, cervical cancer incidence rates have generally declined. High-SDI regions have established mature screening-vaccination-treatment systems, and age-standardized incidence rates have decreased significantly due to early screening and treatment despite population aging [18–21]. Middle-SDI regions may be in a “catch-up phase of preventive measures,” where rapid improvements in vaccine and screening coverage offset the impact of rising risk factors, leading to stabilized or slightly reduced incidence rates [22, 23]. In low SDI regions, reported incidence rates may be severely distorted due to insufficient medical resources and awareness.

Although the global burden of cervical cancer has declined overall, health inequality analysis reveals that the SII for crude incidence shifted from positive to negative between 1990 and 2021, and its absolute value increased, indicating a concentration of cervical cancer burden in low SDI regions and a widening gap between high and low SDI regions. This suggests that while existing preventive measures and treatments are generally effective, their outcomes are disproportionately influenced by inequalities in healthcare access and resource allocation. Addressing these gaps is critical to achieving global cervical cancer control and equity.

The BAPC prediction results across SDI regions highlight challenges in resource allocation and effectiveness. Generally, the rate of decline in cervical cancer incidence correlates with the level of resource allocation. Middle, low-middle, and low SDI regions face higher disease burdens but receive fewer resources, exacerbating health inequalities. Additionally, the issue of high investment with low returns is evident. For example, in middle-high SDI regions, although incidence rates have declined significantly, the corresponding changes in DALYs are minimal. This indicates inefficient resource utilization and raises concerns about potential resource waste.

SDI primarily reflects healthcare accessibility and resource availability. However, when discussing health inequalities, we must also emphasize inequalities caused by individual or healthcare system barriers. Overreliance on medical technology and cognitive biases may lead to increased disease risks. Some populations develop false reassurance from frequent high-sensitivity screening, neglecting health behavior management and increasing the likelihood of high-risk sexual behaviors [24]. The failure to translate advanced technologies into universal services (e.g., stratification between public and private healthcare systems) may result in unequal distribution of health benefits. These issues may prevent the complete elimination of the disease burden in high SDI regions [25, 26].

Although the SDI provides a proxy for healthcare accessibility and resource availability, its capacity to quantify inequities in medical resource distribution remains inherently constrained. To strengthen causal inference validity, future investigations should systematically integrate multidimensional healthcare accessibility metrics (e.g., geographic reachability, economic affordability, cultural acceptability) and structural inequality proxies (e.g., racial disparity indices, health insurance coverage gradients). By developing hierarchical multivariable models incorporating these covariates, researchers may establish context-specific SDI policy simulation systems to refine the design and evaluation of targeted social development interventions.

For low and middle SDI regions, the following potential risk factors may exist. Under free screening policies, impoverished populations may abandon services due to unaffordable hidden costs (e.g., transportation and lost wages) [27]. Extreme poverty and educational deprivation suppress individual health needs. Even with international aid, survival pressures permanently postpone health demands. Additionally, cultural beliefs, religious practices, and misinformation contribute to disease burdens in both developed and underdeveloped regions [28, 29].

The significant negative correlation between screening rates and cervical cancer DALY rates (r = -0.56, p < 0.01) underscores the effectiveness of early detection programs in reducing disease burden. As screening rates increase, the overall burden of cervical cancer decreases, further highlighting the importance of widespread screening initiatives. Similarly, HPV vaccination coverage negatively correlates with cervical cancer incidence (r = -0.35, p < 0.01), emphasizing the critical role of vaccination in mitigating cancer risk. However, data suggest that in high SDI regions, the effectiveness of screening and HPV vaccination in reducing cervical cancer burden has slightly declined. This phenomenon may be linked to the aforementioned individual or healthcare system barriers in high SDI regions. It also reminds us that policy measures should not blindly increase resource inputs.

Based on the above findings, we offer the following policy recommendations. First, all regions should strengthen education on cervical cancer knowledge to dismantle cultural and ideological barriers [30]. For high SDI regions, prevention strategies should shift from large-scale interventions to precision governance, reducing costs while ensuring equitable access to advanced technologies [31, 32]. For low SDI regions, strategies should prioritize expert assistance and resource allocation, focusing on basic healthcare system development alongside minimum effective interventions [33, 34]. For middle-development regions, which are likely in a phase of adapting prevention systems to local contexts, it is essential to address “center-periphery” resource disparities and hidden burdens imposed on low-income groups by prevention strategies.

Our study has several limitations. First, the reliability of cervical cancer burden estimates is closely tied to the accessibility and completeness of data from the GBD 2021. This limitation is particularly pronounced in regions with underdeveloped cancer registry systems, such as sub-Saharan Africa and South Asia, where estimates rely heavily on model extrapolation applied to countries with incomplete primary data, thereby introducing substantial uncertainties. While GBD 2021 attempts to correct for underreporting through modeling, residual biases could inflate measured inequalities. Second, this study synthesizes data from multiple international databases for statistical analysis. It is important to note that cervical cancer screening rates were derived exclusively from OECD member countries, which predominantly represent high- and middle-high SDI regions. Similarly, HPV vaccination coverage data obtained from WHO reports may not fully encompass realities in low-income settings. These geographic and socioeconomic disparities in data availability may constrain the global generalizability of our conclusions. Despite its limitations, this research is pioneering in its statistical analysis of the relationship between preventive measures and disease burden, as well as its exploration of cervical cancer health inequalities. These findings provide valuable insights to inform current policy-making efforts aimed at reducing the global burden of cervical cancer.

In conclusion, this study analyzed trends in the cervical cancer disease burden across different SDI regions, assessed health inequalities among 204 countries, and evaluated the effectiveness of existing preventive measures. While the global cervical cancer burden has significantly declined in recent decades, substantial regional disparities persist. Future policies should focus on targeted interventions tailored to regional needs. Low-resource settings should prioritize strengthening foundational healthcare infrastructure to mitigate their disproportionately high disease burden while simultaneously implementing minimum effective preventive interventions. In contrast, high-SDI regions should transition prevention strategies from broad population-level interventions to precision public health approaches, optimizing resource efficiency to ensure equitable access to technological advancements in medicine. Furthermore, policymakers globally must integrate culturally sensitive health education campaigns into cervical cancer control frameworks. These initiatives should aim to address sociocultural barriers, counteract entrenched misconceptions, and empower communities with evidence-based knowledge, ultimately reducing preventable cervical cancer morbidity through sustained behavioral and systemic changes.

Abbreviations

ASIR

Age-Standardized Incidence Rate

BAPC

Bayesian Age-Period-Cohort

CI

Concentration Index (health inequality) or Confidence Interval (statistics)

DALYs

Disability-Adjusted Life Years

EAPC

Estimated Annual Percentage Change

GBD

Global Burden of Disease

OECD

Organisation for Economic Co-operation and Development

SDI

Socio-Demographic Index

SII

Slope Index of Inequality

UI

Uncertainty Interval

Author contributions

Aimin Huang and Wentao Wu conceptualized and designed the study. Yingxin Zhang and Zhe Fan performed data collection and statistical analyses. Jiankang Wang and Fengyi Zhou interpreted the results and drafted the manuscript. Bing Guan and Zihan Tang provided critical revisions and supervised the study. All authors contributed to the discussion of the results, approved the final manuscript, and agreed to be accountable for the work presented.

Funding

This study was supported by Jiyuan Charity Federation Project (Grant No. 2024-KY-06-001). The funding agency had no role in the study design, data collection, analysis, or interpretation, the writing of the manuscript, or the decision to submit the manuscript for publication.

Data availability

The datasets used and analysed during the current study available from the corresponding author on reasonable request.

Declarations

Ethical approval

All data used in this study are publicly available on the respective websites. Since no personal or sensitive information was involved, ethical approval was not required for the implementation of this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.