Unequal patterns of tuberculosis in women of reproductive age: Insights from the Global burden of disease study 2021

Highlights

• •Persistent global inequalities, with TB DALYs up to 100-fold higher in sub-Saharan African countries compared to high-income regions.
• •Age-shifting disease dynamics, where TB incidence peaks in women aged 15–24, while mortality and drug-resistant TB burden are highest among those aged 45–49.
• •A growing contribution of metabolic risk factors (e.g., high BMI) to TB mortality in both high- and low-resource settings, particularly for drug-resistant TB.

Abstract

Background

Tuberculosis (TB) remains a major global cause of mortality and is a leading cause of death among women of reproductive age (WRA). Untreated TB in this population endangers both maternal and neonatal health, yet the age-stratified burden and the shifting distribution of drug-resistant tuberculosis remain poorly defined in global health agendas. This study aims to employ data from the 2021 Global Burden of Disease Study to systematically evaluate the global, regional, and age-specific burden of TB among WRA, to delineate temporal trends in drug-resistant TB, and to assess associated risk factors using standardized estimation methods.

Methods

Leveraging data from the Global Burden of Disease Study 2021 covering 204 countries and territories, we systematically assessed TB burden in WRA aged 15–49 years from 1990 to 2021. We analyzed age-standardized incidence, prevalence, mortality, and disability-adjusted life years across socio-demographic index (SDI) regions, TB subtypes, and risk factors. Projections to 2030 were generated using autoregressive integrated moving average models.

Results

Between 1990 and 2021, the global prevalence and incidence of TB among WRA declined significantly. However, this decline was uneven across age groups, disease types, and regions. TB incidence peaked in women aged 15–24 years, while mortality and disease burden were highest among those aged 45–49 years. Drug-resistant tuberculosis was most prominent in the 25–29 age group. Geographically, sub-Saharan Africa remained the core of the global TB burden, with BRICS (Brazil, Russia, India, China, South Africa) countries carrying significantly higher burdens than G7 countries. Risk attribution patterns revealed a shift from behavioral to metabolic drivers, with high body mass index emerging as the leading risk factor in high-SDI regions. Middle-SDI regions showed a combination of metabolic and behavioral risks, while metabolic factors remained dominant in low-SDI settings. Drug-susceptible tuberculosis was primarily linked to smoking and malnutrition, whereas drug-resistant tuberculosis was increasingly associated with metabolic dysfunction.

Conclusions

Despite global declines in TB prevalence and incidence among WRA, stark disparities persist. The age-stratified rise in drug-resistant TB, particularly in resource-constrained regions, underscores the need for targeted, gender-responsive interventions that integrate infectious disease control with metabolic health strategies.

1. Introduction

Tuberculosis (TB) remains a major global cause of mortality.¹ Particularly in the context of maternal health, untreated TB during reproductive years endangers both maternal and neonatal health, compounding intergenerational consequences. It is associated with a significantly increased risk of adverse pregnancy outcomes, including spontaneous abortion, preterm labor, and low birth weight.^2,3 According to the World Health Organization’s Global Tuberculosis Report 2023, an estimated 10.6 million new TB cases occurred worldwide in 2022, corresponding to an incidence rate of 133 per 100,000 population—reflecting a 3.9% increase from 2020 to 2022.^4,5 Although decades of investment in vaccines and diagnostics contributed to substantial declines in TB incidence, recent trends indicate a worrying resurgence.^6,7

Among women of reproductive age (WRA), TB is also a leading cause of death, yet remains unrecognized in global health agendas.8, 9, 10 The risk of TB in this population is shaped not only by chronological age, but also by evolving screening strategies, diagnostic criteria, and access to healthcare services. Behavioral, environmental, and individual factors further influence longitudinal patterns across birth cohorts. These include risks like smoking, environmental factors such as healthcare access, and individual comorbidities.11, 12, 13

Despite these vulnerabilities, most current evidence focuses on the general population or narrowly defined subgroups, with limited attention to the age-stratified burden in WRA. Moreover, the distribution of drug-resistant TB, the socioeconomic gradient in disease outcomes, and pregnancy-specific risks remain poorly defined. The Global Burden of Disease Study 2021 (GBD 2021) provides a unique opportunity to systematically quantify TB burden by age, region, and risk factor profile. Here, we present a comprehensive analysis of TB trends among WRA across 204 countries and territories from 1990 to 2021, stratified by socio-demographic index (SDI) quintiles. We further explore the relationship between healthcare resource distribution and disease burden, with the aim of informing age- and sex-responsive TB control strategies.

2. Materials and methods

2.1. Data resources and classification

This study utilized data from the GBD 2021, encompassing 204 countries and territories.¹⁴ Core GBD methodologies—including standardized data modeling, uncertainty quantification, and cross-regional comparability validation—have been extensively documented by the GBD Collaborator Network.^15,16 TB cases were classified according to GBD criteria into four categories: latent TB infection, drug-susceptible TB, multidrug-resistant TB without extensive drug resistance and extensively drug-resistant TB. To comprehensively reflect the clinical complexity and epidemiological burden of drug-resistant TB, the latter two categories were consolidated as “drug-resistant TB”, encompassing the aggregate burden of both multidrug-resistant and extensively drug-resistant TB.

The SDI—a composite metric integrating per capita income, educational attainment, and total fertility rate—stratified 204 countries and territories into five developmental quintiles: high, high-middle, middle, low-middle, and low SDI.¹⁷ To elucidate TB burden-socioeconomic relationships, SDI quantified national/regional development levels. The G7 countries (Canada, France, Germany, Italy, Japan, the United Kingdom, the United States of America) have a high SDI background and represent developed countries. The BRICS countries (Brazil, Russia, India, China, South Africa) have a middle-high SDI and a middle-low SDI, corresponding to developing countries. The East African Community (Kenya, Uganda, Tanzania, Rwanda, Burundi, South Sudan, the Democratic Republic of the Congo, Somalia) is in a low SDI environment and corresponds to developing countries. Women aged 15–49 years were defined as the reproductive-age population per World Health Organization standards.¹⁸ Analyses were stratified across seven age subgroups: 15–19, 20–24, 25–29, 30–34, 35–39, 40–44, and 45–49 years.

2.2. Data standardization

To account for heterogeneity in population age structures across compared regions, all rates underwent age standardization. The standardization procedure utilized the Global Burden of Disease Study 2019 (GBD 2019) reference population weights.¹⁴ The age-standardized rate (ASR) was calculated as:

$$\text{ASR}=\frac{{\sum }_{i=1}^A{\alpha }_i{\omega }_i}{{\sum }_{i=1}^A{\omega }_i}$$ (1)

Wherein, ${\alpha }_i$ denotes the crude rate for each age group, ${\omega }_i$ represents the standard population weight, and A refers to the seven reproductive age groups (15–19 years to 45–49 years).

2.3. Data visualization

Leveraging GBD data spanning 1990–2021, this study employed autoregressive integrated moving average (ARIMA) models to project mortality, prevalence, and disability-adjusted life years (DALYs) rates for TB among WRA from 2021 to 2030.

The optimal model parameters (p, d, q), were automatically selected using the auto.arima() function from the forecast package in R (version 4.4.2), with the optimization objective of minimizing the Akaike Information Criterion. During the data preprocessing stage, ASR indicators were extracted from the GBD database, aggregated by year to generate annual time series, and independent time series objects were constructed for mortality rates, prevalence rates, and DALYs rates respectively. After model fitting, point estimates for the next decade were generated. The forecast results were merged with historical data and visualized using the ggplot2 and gridExtra packages, then exported as integrated charts.

2.4. Statistical analysis

All estimates generated by the GBD through its standardized modeling procedures are accompanied by uncertainty intervals (UIs), which serve to quantify the precision of parameter estimations. The 95% UIs used in this study were defined by extracting 500 parameter values from the Bayesian posterior distribution and taking the 2.5th and 97.5th percentiles. Uncertainty was propagated hierarchically across all procedural stages to ensure robust results. This process integrated the impacts of data heterogeneity, model discrepancies, and demographic fluctuations on burden estimates.

3. Results

3.1. Global and SDI-stratified tuberculosis burden trends

From 1990 to 2021, the global age-standardized TB prevalence among WRA declined from 33,784.78 (95% UI: 23,723.09–44,328.49) to 25,792.60 (95% UI: 18,136.50–34,363.76) per 100,000 women, while incidence decreased from 164.32 (95% UI: 100.10–243.78) to 102.95 (95% UI: 63.35–152.84) per 100,000 women (Fig. 1A). High-SDI regions persistently exhibited the lowest prevalence, reaching 13,754.54 (95% UI: 9,393.07–19,038.55) per 100,000 in 2021—significantly below the global average (Fig. 1A). Moreover, middle-SDI regions demonstrated a prevalence of 33,619.55 (95% UI: 24,808.99–42,871.69) per 100,000 in 2008, surpassing low-SDI regions (33,283.16; 95% UI: 23,501.17–43,866.07) for the first time, with this disparity progressively widening through 2021 (Fig. 1A). In 2021, incidence rates remained substantially elevated in low-SDI (173.97; 95% UI: 102.19–268.21) and low-middle-SDI regions (161.44; 95% UI: 98.17–243.49) per 100,000, exceeding global levels (Fig. 1B).

Fig. 1.

Trends in tuberculosis burden among WRA in the world and in different SDI, 1990–2021. (A) Age-standardized tuberculosis prevalence. (B) Age-standardized tuberculosis incidence. (C) Advanced drug-resistant TB prevalence. (D) Drug-susceptible tuberculosis prevalence. (E) Latent tuberculosis infection prevalence. (F) Incidence of drug-resistant TB. (G) Incidence of drug-susceptible TB.

Abbreviations: ASR, age-standardized rate; SDI, socio-demographic index; TB, tuberculosis; WRA, women of reproductive age.

Regarding drug-resistant TB, multidrug-resistant/extensively drug-resistant TB was predominantly concentrated in middle-SDI regions before 2005. However, a pronounced global decline in drug-resistant TB incidence occurred across all SDI strata following 2006, culminating in a 2021 global incidence of 5.65 per 100,000 (Fig. 1C). Post-2015, resurgent increases emerged in low-, middle-, and low-middle-SDI regions, contrasting with the persistent decline observed in high-SDI regions (Fig. 1C). Drug-susceptible TB incidence demonstrated sustained global reductions (Fig. 1D), whereas latent TB infection prevalence progressively increased in low-, middle-, and low-middle-SDI regions (Fig. 1E). Despite SDI regional variations, drug-resistant TB prevalence demonstrated bimodal fluctuations, peaking initially in 2005 (predominantly low-middle SDI) before transitioning to universal decline by 2020 (Fig. 1F).This divergence was particularly evident in high-SDI regions, where drug-susceptible TB incidence (8.38; 95% UI: 6.87–10.32 per 100,000) constituted merely 4% of the rate documented in low-SDI regions (167.64; 95% UI: 135.96–206.93 per 100,000) (Fig. 1G).

3.2. Age-stratified burden patterns and disease-type disparities

Between 1990 and 2021, TB prevalence among WRA demonstrated significant age-stratified and regional heterogeneity. In low-SDI regions, prevalence declined from 36,143.23 (95% UI: 25,154.97–47,422.36) to 20,010.19 (95% UI: 13,701.05–28,065.00) per 100,000 among 15–19-year-olds, while the 20–24-year cohort showed a 23.1% reduction during 2000–2010. High-SDI regions achieved more substantial gains, with 74.9% and 78.1% prevalence declines in 25–29 and 30–34-year-olds respectively (Fig. 2A). Despite maintaining the highest absolute burden, low-SDI regions exhibited pronounced reductions in younger age groups, whereas middle-SDI regions demonstrated persistently higher prevalence among 45–49-year-olds compared to low-SDI counterparts since 2000—revealing a critical “diagnostic-capacity-improvement versus treatment-adherence-deficit” transition paradox (Fig. 2B).

Fig. 2.

Trends in TB burden in WRA in different age groups in the world and in different SDI from 1990 to 2021. (A) TB prevalence in WRA in different age groups (15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49 years). (B) TB incidence in WRA in different age groups (15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49 years).

Abbreviations: SDI, socio-demographic index; TB, tuberculosis; WRA, women of reproductive age.

Mortality and health loss patterns revealed deeper contradictions. TB mortality demonstrated steep age-dependent escalation, peaking among 45–49-year-olds with pronounced geographical inequities—low-SDI regions endured extreme pressures (1990 mortality: 112.39 per 100,000; 95% UI: 95.18–133.88), while low-middle-SDI regions consistently exceeded global averages beyond age 30 (Supplementary Fig. S1A). Low-SDI regions faced severe mortality burdens beyond age 30: while 15–19-year-old mortality decreased from 17.11 (95% UI: 14.36–20.97) in 1990 to 15.07 (95% UI: 12.89–18.42) in 2003, middle-SDI regions surpassed low-SDI mortality among 30–49-year-olds post-2010, with accelerating DALYs accumulation (Supplementary Fig. S1B). Despite modest mortality declines in 15–24-year-olds, older reproductive-age women in resource-constrained settings remained the most vulnerable demographic. The 2021 health loss landscape revealed triple disparities: Geographic survival inequities manifested as 88-fold higher drug-susceptible TB mortality in low-SDI versus high-SDI regions among 25–29-year-olds (11.40 [95% UI: 9.37–14.12] vs. 0.13 [95% UI: 0.10–0.19] per 100,000), with mortality peaks occurring 15 years earlier than in affluent regions; Low-SDI regions exhibited systemic collapse in drug-resistant TB management for older women, with 45–49-year-old mortality surging to 3.11 per 100,000 (95% UI: 1.27–6.48)—6.5 times higher than middle-SDI counterparts (0.48; 95% UI: 0.16–1.01) (Supplementary Fig. S2A). Among 30–34-year-olds in low-SDI regions, drug-resistant TB mortality (1.22 per 100,000; 95% UI: 0.49–2.49) constituted merely 8% of total TB deaths yet generated 74.31 DALYs (95% UI: 30.91–149.09), accounting for 7.8% of total TB health loss (Supplementary Fig. S2B).

3.3. Global distribution and cross-national disparities in TB burden

The 2021 global distribution of TB burden among WRA exhibited profound disparities (Fig. 3A). The bubble plots reveal a consistent pattern across both ASDR and DALYs rates, where the majority of nations maintain low levels of epidemic, while a few countries emerge as extreme outliers. Specifically, the Central African Republic, Lesotho, and Somalia identified as the three highest-burden nations for both mortality and DALYs rates globally. Their age-standardized DALYs rates (Central African Republic: 4,653.41, 95% UI: 2,976.36–6,861.76; Lesotho: 3,877.45, 95% UI: 2,320.91–5,841.47; Somalia: 3,501.00, 95% UI: 1,945.35–6,259.80) represented the extreme peaks of the global epidemic. These rates approached 100-fold elevation above the global median (e.g., China: 37.88; 95% UI: 27.19–50.89) (Fig. 3B).

Fig. 3.

The global burden of TB in 204 countries and territories in 2021. (A) The ASDR of 204 countries and territories. Each bubble represents a country or territory, with its vertical position indicating the mortality level. To highlight the most extreme disparities, the top three high-burden countries—the Central African Republic, Lesotho, and Somalia—are explicitly labeled. (B) The age-standardized DALYs rates of 204 countries and territories. The distribution of bubbles illustrates the variation in DALYs rates across nations. Similar to panel A, only the three highest-burden countries (the Central African Republic, Lesotho, and Somalia) are marked to show the peak of the global epidemic.

Abbreviations: DALY, disability-adjusted life year; ASDR, age-standardized death rate; TB, tuberculosis.

Cross-national comparisons further revealed markedly lower burdens in G7 countries—France: 5.41 (95% UI: 4.81–6.15), Germany: 2.84 (95% UI: 2.34–3.46), Canada: 2.96 (95% UI: 2.60–3.40) DALYs/100,000 (Supplementary Fig. S3A)—whereas persistently high burdens persisted in BRICS countries, particularly India (753.93; 95% UI: 632.42–886.29) and South Africa (1,338.15; 95% UI: 1,116.72–1,570.71) (Supplementary Fig. S3B), confirming the inverse correlation between economic development and disease burden. East African Community states exhibited higher proportional years lived with disability contributions (e.g., Uganda: 93.08/678.55 ≈ 13.72%) compared to other sub-Saharan African countries (e.g., Democratic Republic of Congo: 212.26/1,979.02 ≈ 10.73%) (Supplementary Fig. S3C), indicating regional heterogeneity in TB-related disability patterns.

3.4. Risk factor attribution patterns for TB mortality

Global analysis of TB mortality risk factors among WRA revealed a significant shift from 1990 to 2021, with metabolic risks gaining prominence over behavioral risks, particularly high body mass index (BMI) rising to become the leading risk factor (Supplementary Fig. S4).

In 2021, the risk attribution of TB deaths in WRA at different SDI levels showed distinct differences: metabolic risks dominated in high SDI areas, especially in the 45–49 age group, with high BMI accounting for 24.11% (95% UI: 6.83%–47.05%) of TB deaths; in contrast, alcohol use accounted for only 9.48% (95% UI: −8.31%–33.87%) and high fasting blood glucose 9.55% (95% UI: 6.36%–13.98%). These data indicate that obesity and dysglycemia need to be addressed in resource-sufficient settings (Fig. 4A). The attribution pattern in the medium SDI region showed a mixed “metabolic + behavioral” characteristic, with high BMI accounting for 11.26% (95% UI: 2.14%–26.85%), alcohol use and high processed meat intake accounting for 5.28% (95% UI: −2.37%–24.67%) and 0.17% (95% UI: 0.04%–0.30%), respectively, reflecting the impact of changes in diet and lifestyle during economic development on the burden of TB deaths (Fig. 4A). Although the risk of lifestyle behaviors increased in low SDI areas, metabolic risk still ranked first. The attributable rate of death for women aged 45–49 years due to high BMI was 7.62% (95% UI: 1.55%–20.13%), alcohol use was 6.66% (95% UI: −2.38%–27.66%), and high fasting blood glucose was 5.12% (95% UI: 3.57%–6.90%), indicating that metabolic factors should not be ignored even in the most vulnerable groups (Fig. 4A). Among different types of TB, deaths from drug-susceptible TB were mainly related to smoking and malnutrition (such as low vegetable intake), while deaths from drug-resistant TB were more driven by metabolic diseases and low physical activity, especially in medium- and high-SDI areas, where the contribution of metabolic factors was most obvious (Fig. 4B).

Fig. 4.

Sankey diagram analysis of risk factor attribution for tuberculosis deaths in 2021. (A) Stratified by 5 territories and age group (15–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49 years), showing the proportion of tuberculosis deaths attributable to behavioral and metabolic risk factors. (B) Stratified by tuberculosis type (drug-susceptible tuberculosis, drug-resistant tuberculosis) and age group (same as above), showing the proportion of tuberculosis deaths attributable to the same behavioral and metabolic risk factors mentioned above.

Abbreviation: SDI, socio-demographic index.

Based on the analysis of transnational TB mortality risk factors, the G7 and BRICS countries are mainly driven by metabolic diseases, high BMI and high fasting blood sugar become the core risks, while Russia deviates significantly from this pattern due to alcohol and smoking; the East African Community is completely different, with alcohol exposure dominating the risk pattern, and diet and exercise risks have little impact globally. This sharp contrast reveals the differentiation of prevention and control paths: industrial countries need to integrate TB and metabolic disease management, while Africa urgently needs to prioritize curbing alcohol abuse (Supplementary Fig. S5).

3.5. Projected TB burden trajectories to 2030

Leveraging 1990–2021 global burden data and 2021–2030 projection models for reproductive-age women (15–49 years), total DALYs declined from 13.661 million (95% UI: 12.436–15.088 million) to 7.827 million (95% UI: 7.070–8.731 million), achieving a 2.1% annual decline rate, while the DALYs rate is projected to reach 320.1 per 100,000 (95% UI: 290.4–358.2) in 2030 (Fig. 5A). Prevalence exhibited parallel secular declines, decreasing from 33,599.9 per 100,000 (95% UI: 28,864.6–39,362.1) in 1990 to 25,950.5 per 100,000 (95% UI: 22,556.2–30,220.0) in 2021, representing a cumulative reduction of 22.8% (Fig. 5B). Concurrently, TB-related deaths demonstrated sustained decline from 219,900 cases (95% UI: 199,500–243,100) in 1990 to 123,500 cases (95% UI: 110,800–138,600) in 2021—yielding an annualized reduction rate of 1.7%—with mortality projected to decrease further to 5.2 per 100,000 (95% UI: 4.8–5.7) by 2030 (Fig. 5C).

Fig. 5.

Historical trends and projections of the global tuberculosis burden among WRA (observed in 2021, projected for 2022–2030). (A) Global projected trends in tuberculosis-related DALYs. (B) Global projected trends in tuberculosis-related prevalence. (C) Global projected trends in tuberculosis-related death.

Abbreviations: DALY, disability-adjusted life year; WRA, women of reproductive age.

4. Discussion

This study provides a comprehensive assessment of TB burden among WRA across global and SDI-stratified populations. Despite a significant decline in global prevalence since 1990, striking disparities persist across age groups, regions, and TB subtypes. Notably, the resurgence of drug-resistant TB in low- and middle-SDI regions—particularly in sub-Saharan Africa—alongside rising mortality in older WRA and divergent regional risk patterns, underscores critical gaps in current control strategies. These findings underscore the need for context-specific precision interventions. Addressing these inequities requires integrated strategies that dismantle the structural and gendered barriers hindering women's access to the full continuum of TB care—from diagnosis to prevention. The age-stratified distribution of TB burden in WRA reflects the combined influence of biological vulnerability and sociobehavioral exposure. The incidence peak among 15–24-year-olds aligns with a period of immunological maturation and increased exposure risk associated with sexual debut and educational transitions—consistent with prior GBD findings on impaired immune responsiveness during adolescence.¹⁹ Among women aged 20–24 years, occupational exposures and crowded living conditions may further elevate risk, while pregnancy-related immunosuppression likely contributes to increased drug-resistant TB incidence in the 25–34-year cohort. Indeed, our analysis identified a 2.25% higher drug-resistant TB mortality in this group compared to their 15–19-year-old counterparts, reinforcing the importance of TB screening during preconception and antenatal care.²⁰ The rising mortality and DALYs burden among women aged 35–49 years appears closely linked to comorbid conditions such as diabetes and HIV, combined with diagnostic delays. In low-SDI regions, 45–49-year-olds experienced a 17.46% higher DALYs burden relative to the 20–24-year reference group—highlighting the cumulative impact of metabolic decline, inadequate healthcare access, and weak social support systems in the later reproductive years.²¹ The prevalence gap between middle- and low-SDI regions among women aged 45–49 years likely reflects ascertainment bias. While restricted testing in low-SDI settings leads to systemic under-ascertainment, higher rates in middle-SDI regions are largely driven by superior diagnostic infrastructure and improved case detection.

We identified a hyperendemic “triangular epicenter”—comprising the Central African Republic, Lesotho, and Somalia—where the TB burden is nearly 100 times the global median (DALYs > 3,500 per 100,000). This disparity is driven by a lethal convergence of factors, including hyperendemic HIV (e.g., 19.3% in Lesotho), conflict-induced healthcare collapse, and severe malnutrition (affecting 6.6% of children in the Central African Republic). Such synergistic forces not only accelerate disease progression but also create formidable barriers to effective care.^22,23

In contrast, G7 countries maintain DALYs below 5 per 100,000 through robust universal health coverage and precision public health strategies, underscoring the central role of socioeconomic development in infectious disease control.²⁴ While BRICS countries have made progress in TB control, the pace of improvement remains modest compared to high-income Asia-Pacific countries.²⁵ India, for example, displays a uniquely “youth-concentrated” TB burden among WRA, with 41% of incident cases occurring in women aged 20–34 years. Despite implementation of the National Tuberculosis Elimination Programme, the country faces ongoing challenges, including funding shortfalls, inadequate diagnostic infrastructure, underreporting, high attrition from treatment programs, and pervasive social stigma.²⁶ Since 2002, South Africa has overtaken India in TB age-standardized mortality rates, driven largely by a syndemic of TB and HIV.²⁵ National HIV prevalence stands at 13.5%, rising to over 30% among pregnant women attending antenatal care, with 8,400 pregnant women diagnosed with drug-susceptible TB in 2016 alone.^27,28 Yet, the exclusion of pregnant women from clinical trials has left critical gaps in evidence regarding the safety and efficacy of drug-resistant TB treatment during pregnancy—constraining the development of evidence-based care guidelines for this vulnerable group.29, 30, 31

Managing TB during pregnancy requires addressing both biological and clinical hurdles. The Th1/Th2 imbalance necessary for fetal tolerance paradoxically attenuates the immunological vigilance required to contain Mycobacterium tuberculosis, elevating the risk of disease reactivation and adverse maternal outcomes.³² Beyond systemic risks, women genital tuberculosis remains a neglected threat to reproductive health and a primary cause of tubal factor infertility.³³ Clinical management is further complicated by diagnostic delays—as symptoms like fatigue often overlap with normal pregnancy physiology—and critical evidence gaps in medication safety, particularly for drug-resistant strains, due to the historical exclusion of pregnant women from clinical trials.^9,34 Consequently, as advocated by WHO guidelines, integrating sex-sensitive TB screening into routine antenatal care is imperative to dismantle these barriers, preserve reproductive potential, and ensure optimal maternal-fetal health.³⁵

This study revealed marked spatial and temporal variation in the distribution of drug-resistant TB. Prior to 2005, middle-SDI regions bore the highest burden, driven by limited access to second-line anti-tubercular drugs and suboptimal treatment adherence. Following a transient global decline, a resurgence of drug-resistant TB emerged after 2015 in low-SDI regions—exposing vulnerabilities in international supply chains, including stockouts of second-line regimens, and the breakdown of preventive infrastructure in conflict-affected areas.^36,37 In contrast, high-SDI regions sustained reductions in drug-resistant TB incidence through the early implementation of molecular diagnostics (e.g., Xpert MTB/RIF) and adherence to directly observed treatment short-course protocols—highlighting the value of precision diagnostic and delivery systems in containing resistant strains.²⁰ Of particular concern is the rising drug-resistant TB burden among women aged 40–49 years, which likely reflects the convergence of metabolic comorbidities and complex drug interactions that undermine treatment efficacy and exacerbate toxicity. Our risk attribution analysis further clarifies this trend, showing that metabolic risk factors contribute significantly to the TB burden, particularly in the 40–49 age group and higher SDI regions. Sankey diagram results confirm that metabolic risks account for a substantial proportion of deaths in both drug-susceptible and drug-resistant TB, identifying metabolic health as a critical variable in TB outcomes. This metabolic-TB link is driven by dual mechanisms. Biologically, high BMI triggers meta-inflammation—a state of chronic, low-grade systemic inflammation—which leads to immunometabolic dysfunction.³⁸ Sociologically, this shift reflects the global “nutrition transition”.³⁹ The resulting “syndemic”—the convergence of metabolic dysfunction and infectious threats—has caused metabolic risks to supersede undernutrition as the leading driver of TB burden in middle-to-high SDI populations.⁴⁰ These findings underscore the urgent need to develop simplified, comorbidity-adapted treatment regimens tailored for older reproductive-age women—particularly in resource-limited settings where polypharmacy is often unavoidable and clinical trial data are scarce.⁴¹

This study has several limitations. While our projections suggest a continued global decline in TB burden among WRA, they do not fully capture the potential widening of regional disparities. The post-2015 resurgence of drug-resistant TB in low-SDI regions, for example, may be further amplified by funding constraints, rising HIV co-infection rates, and conflict-related health system disruptions^42,43 —factors that are not explicitly modeled. Additionally, the 2020–2021 fluctuations reflect a pandemic-induced “diagnostic gap” coupled with a rebound in mortality, driven by service disruptions and treatment interruptions.⁴⁴ While the ARIMA model effectively captures historical trends, its univariate nature cannot account for such future external shocks, including policy shifts, economic fluctuations, or climatic factors. This divergence from our “business-as-usual” ARIMA projections highlights the severe impact of COVID-19 on global TB surveillance and survival among WRA. Moreover, climate change-driven food insecurity and population displacement may jeopardize recent gains in TB control. Future modeling efforts should incorporate a broader range of dynamic and context-specific variables—such as political instability, nutritional status, HIV prevalence, and healthcare access—to improve predictive accuracy and guide more equitable and resilient TB prevention strategies, particularly for vulnerable populations such as WRA.

5. Conclusions

The global TB among WRA arises from a lethal convergence of biological vulnerability and entrenched social inequity. While high-SDI regions demonstrate the efficacy of robust health systems and precision interventions, lower-SDI regions must adopt integrated models that pair clinical treatment with structural social interventions.⁴⁵ Moving forward, global efforts should prioritize closing diagnostic gaps and embedding sex-sensitive frameworks across the entire care continuum—from prevention to cure. Such actions are essential to safeguard the health rights of women and accelerate the elimination of TB in this critically underserved population.

CRediT authorship contribution statement

Ting Liu: Writing – original draft, Investigation, Visualization. Lei Zhang: Writing – original draft, Investigation, Visualization. Jiao Liu: Writing – original draft, Visualization, Investigation. Yao Liu: Writing – review & editing, Writing – original draft. Anying Xiong: Writing – review & editing, Visualization. Qin Ran: Writing – review & editing, Visualization. Xiang He: Writing – review & editing, Visualization. Li Jiang: Writing – review & editing, Supervision. Guoping Li: Writing – review & editing, Supervision, Conceptualization, Funding acquisition. Junyi Wang: Writing – review & editing, Supervision, Conceptualization, Funding acquisition.

Informed consent

Not applicable.

Organ donation

Not applicable.

Ethical statement

This analysis utilized publicly available, aggregated data from the Global Burden of Disease 2021 study and did not involve direct interaction with human subjects. Therefore, ethics approval was not required for this secondary data analysis.

Data availability statement

The data that support the findings of this study are derived from the Global Burden of Disease Study 2021, a public domain resource available through the Institute for Health Metrics and Evaluation (IHME) at http://ghdx.healthdata.org/gbd-results-tool. The specific datasets analyzed during the current study are available from the corresponding author upon reasonable request.

Animal treatment

Not applicable.

Generative AI

During the preparation of this work, the authors used ChatGPT to assist with language refinement, grammar correction, and improvement of clarity and readability. After using this tool, the authors carefully reviewed, revised, and edited all generated content and take full responsibility for the integrity, accuracy, and originality of the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (82300088, 82370022, 82200079), China Postdoctoral Science Foundation (2025M772262), Chengdu Medical Research Project (2025003), The Third People’s Hospital of Chengdu Scientific Research Project (2023PI01, 2023PI15, 2023PI12) and The Third People’s Hospital of Chengdu Clinical Research Program (CSY-YN-03-2024-008, CSY-YN-03-2024-027, CSY-YN-04-2025-012).

Declaration of Interest Statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.