Tuberculosis burden in Korea compared with global trends, 1990–2021: a comparative ecological time-trend study

Background

Tuberculosis (TB), caused by Mycobacterium tuberculosis, remains one of the world’s oldest and deadliest infectious diseases [1]. According to the World Health Organization’s Global Tuberculosis Report 2022, TB continues to represent a substantial global health challenge [2]. In 2021, approximately 10.6 million new TB cases were reported worldwide, an increase from 10.1 million in 2020. TB-related deaths rose to 1.6 million in 2021, compared with 1.5 million the previous year. This reversal represents a concerning deviation from the nearly 2% annual decline observed over the past 2 decades. The 3.6% increase in the TB incidence rate in 2021 further underscores setbacks in TB control efforts. Despite advances in TB management, including strengthened national programs and economic development, substantial challenges persist in certain countries and regions. Korea has made considerable progress in TB control over recent decades [3]; however, its burden remains high compared with that of other high-income countries [4]. To address these challenges effectively, it is important to examine Korea’s TB burden in relation to global trends. Such comparisons help identify persistent gaps and areas requiring targeted intervention. These analyses also provide policymakers with evidence to refine prevention strategies and optimize public health resource allocation.

The Global Burden of Disease (GBD) study is a large-scale initiative that quantifies health loss due to diseases, injuries, and risk factors worldwide. Led by the Institute for Health Metrics and Evaluation, it provides standardized data for comparing disease burdens across countries and over time. GBD data have been used to analyze trends in multiple conditions in Korea, including diabetes, Alzheimer’s disease, depression, and asthma [5–8]. Although TB is included in GBD rankings—declining from 10th in 1990 to 49th in 2019 in Korea—comprehensive analyses of long-term TB trends remain limited [9].

Purpose

This study addresses this gap by using GBD 2021 data to provide a comprehensive analysis of TB burden trends in Korea compared with global patterns from 1990 to 2021. We applied joinpoint regression to assess temporal trends, age–period–cohort modeling to evaluate age, period, and cohort effects, and autoregressive integrated moving average (ARIMA) modeling to forecast TB incidence and mortality through 2030 [10–15]. Our findings aim to inform policy planning and strengthen TB control strategies in Korea.

Ethics statement

This study used publicly available, aggregated data that did not contain individual-level or identifiable information. Therefore, ethical approval was not required.

Study design

We conducted a comparative, population-level ecological analysis of GBD 2021 estimates for Korea and the global population from 1990 to 2021. Joinpoint regression was used to evaluate temporal trends, age–period–cohort models assessed age, period, and cohort effects, and ARIMA models forecasted age-standardized incidence and mortality through 2030.

Setting

This ecological time-trend study used publicly available estimates from the GBD study. Data were downloaded from the GBD Results Tool (Institute for Health Metrics and Evaluation) on June 6, 2025. We analyzed annual TB burden estimates for the Republic of Korea and the global population from 1990 through 2021.

Variables

We defined TB burden using 4 GBD 2021 indicators: incidence, prevalence, mortality (deaths), and disability-adjusted life years (DALYs). Analyses included both all-age counts and age-standardized rates per 100,000 population, including the age-standardized incidence rate (ASIR), age-standardized prevalence rate (ASPR), age-standardized mortality rate (ASMR), and age-standardized DALY rate (ASDR). All estimates were reported with 95% uncertainty intervals (UIs) as provided by GBD.

Key stratification variables included calendar year, location (Korea vs. global), sex, and age group (GBD-defined categories). For joinpoint analyses, log-transformed age-standardized rates served as dependent variables, and trends were expressed as annual percent change (APC) and average annual percent change (AAPC) from 1990 to 2021. Age–period–cohort models decomposed effects into age, period, and birth-cohort components. For forecasting, annual age-standardized TB incidence and mortality rates were projected through 2030 (2022–2030) using ARIMA models.

Data source

We obtained data on TB incidence, prevalence, mortality, and DALYs in Korea and globally from 1990 to 2021 using the GBD Results Tool (https://vizhub.healthdata.org/gbd-results/).

Measurement

Bias

Key potential biases include (1) measurement/modeling bias inherent to GBD estimates (e.g., incomplete notification and misclassification), and (2) comparability bias over time due to evolving diagnostic practices, reporting systems, and cause-of-death coding. To mitigate these concerns, we used a single standardized data source (GBD 2021) for both Korea and global comparisons, focused analyses on age-standardized indicators (ASIR, ASPR, ASMR, and ASDR), and presented estimates with 95% UIs provided by GBD.

Study size

No formal sample-size calculation was performed. We included all available annual GBD 2021 estimates for Korea and the global population from 1990 to 2021 and used these complete time-series data for trend analyses and forecasting.

Statistical methods

We summarized annual TB burden for Korea and the global population (1990–2021) using incidence, prevalence, deaths, and DALYs, reported as all-age counts and age-standardized rates per 100,000 population. Estimates were presented with 95% UIs as provided by the GBD 2021 study. Analyses were conducted overall and stratified by sex and age groups where applicable. Trend assessment, age–period–cohort decomposition, and forecasting were performed using prespecified analytical models described in the Measurement section.

All statistical tests were 2-sided, and statistical significance was defined as P<0.05 when hypothesis testing was applicable. No imputation was performed; analyses used all years with available annual GBD estimates. Trend analysis was performed using the Joinpoint Regression Program (ver. 5.2.0; National Cancer Institute). All other statistical analyses were conducted using R software (ver. 4.3.2; R Foundation for Statistical Computing). Detailed R source code is provided in Suppl. 1.

Description of the burden of TB in Korea and globally

Incidence of TB. Between 1990 and 2021, TB cases in Korea decreased by 43.36%, whereas global cases declined only slightly (2.23%). Korea’s ASIR dropped by 72.2%, while the global ASIR decreased by 40.47%. The AAPC was –4.00% in Korea, compared with –1.65% globally (Table 1).

Table 1.

All-age cases and age-standardized rates of tuberculosis in Korea and globally, 1990 and 2021

Prevalence of TB. From 1990 to 2021, Korea’s TB prevalence decreased by 52.18%, whereas global prevalence increased by 19.09%. Korea’s ASPR decreased by 65.79%, while the global ASPR decreased by 23.07%. The AAPC was –3.40% in Korea and –0.85% globally (Table 1).

Mortality of TB. From 1990 to 2021, global TB deaths declined by 34.63%, whereas Korea showed a substantially steeper reduction of 67.60%. The ASMR also decreased by 65.09% globally and by 90.37% in Korea. The AAPC in the global mortality rate was –3.09%, compared with –7.53% in Korea (Table 1).

DALYs of TB. Between 1990 and 2021, global TB DALYs declined by 43.18%, while Korea experienced a sharper reduction of 81.53%. The ASDR decreased by 64.85% globally and by 92.59% in Korea. The AAPC was –3.32% globally and –8.13% in Korea (Table 1). Decomposition showed that although both years of life lost (YLLs) and years lived with disability declined, the reduction in YLLs accounted for most of the overall decline in DALYs (Suppl. 2).

Overall, these findings indicate that Korea has achieved substantially greater reductions in TB burden than the global average over the past 3 decades.

Joinpoint regression analysis of the TB burden in Korea and globally

From 1990 to 2021, joinpoint regression showed a consistent decline in Korea’s TB burden. The ASDR declined steadily, with a sharp drop from 2003 to 2006 (APC, –12.80; 95% CI, –16.13 to –9.34; P<0.05) (Suppl. 3). Globally, the ASDR also decreased, with the most significant drop from 2006 to 2011 (APC, –4.55; 95% CI, –4.80 to –4.30; P<0.05). Both Korea and the global data showed a downward trend in ASMR. In Korea, it rapidly decreased from 1992 to 1996 (APC, –10.78; 95% CI, –12.77 to –8.75; P<0.05) and from 2002 to 2007 (APC, –11.05; 95% CI, –12.33 to –9.75; P<0.05). The global ASIR also decreased, with a notable reduction from 2006 to 2014, while Korea experienced its most rapid decline from 1990 to 1995. The global ASPR also decreased, with the quickest reduction from 2000 to 2015, while in Korea, it was from 2001 to 2010. Detailed APC, CIs, and P-values are provided in Figure 1, Suppl. 3, and Suppl. 4. Overall, Korea experienced faster and more consistent declines compared to the global trends.

Figure 1.

Joinpoint analysis of APC effects for ASDR, ASMR, ASIR, and ASPR of TB by sex in Korea from 1990 to 2021. An asterisk (*) indicates statistically significant changes (P<0.05). (A) ASDR, (B) ASMR, (C) ASIR, (D) ASPR. APC, annual percent change; ASDR, age-standardized disability-adjusted life-year rate; ASMR, age-standardized mortality rate; ASIR, age-standardized incidence rate; ASPR, age-standardized prevalence rate; TB, tuberculosis. This figure was illustrated by the author.

TB Burden by age groups in Korea and globally in 1990 and 2021

Figure 2 and Suppl. 5 compare TB burden by age groups in Korea and globally. In Korea, the incidence peak shifted from ages 20–24 in 1990 to 65–69 in 2021, with the crude incidence rate generally increasing with age in both years. Globally, the peak remained at 20–24 in both years, with a decrease in 0–9 and 65–79 and increases in other age groups. Prevalence in Korea shifted from 30–34 to 50–54, and globally from 20–24 to 30–34. Mortality in Korea rose with age, peaking at 70–74 in 1990 and 80–84 in 2021; Korea’s crude mortality rate (CMR) peaked at ≥95 in both years. Globally, deaths were highest in 0–5 in 1990 and 65–69 in 2021, with CMR peaks at 85–89 in 1990 and 90–94 in 2021. DALYs in Korea peaked at 50–54 in 1990 and 80–84 in 2021, while the crude DALY rate increased consistently with age. Global DALYs peaked at 0–5 in both years. These findings indicate a shift in the TB burden toward older age groups in both Korea and globally.

Figure 2.

Comparison of TB incidence, prevalence, deaths, and DALYs, along with their crude rates, by age group in Korea in 1990 and 2021. (A) Incident cases and CIR; (B) prevalent cases and CPR; (C) deaths and CMR; (D) DALY counts and CDR. Bar charts represent counts, and lines represent crude rates. TB, tuberculosis; DALY, disability-adjusted life year; CIR, crude incidence rate; CPR, crude prevalence rate; CMR, crude mortality rate; CDR, crude disability-adjusted life-year rate. This figure was illustrated by the author.

Sex disparities in the TB burden in different age groups in Korea and globally

Figure 3 and Suppl. 6 present the age- and sex-specific TB burden in 1990 and 2021. In Korea, TB prevalence peaked at ages 25–29 for females and 20–24 for males in 1990, decreasing after age 34 in both sexes. Males had more cases than females up to age 49, while females had more cases thereafter. This trend continued into 2021, except for the 65–74 group. Globally, peak prevalence shifted from 20–24 in 1990 to 30–34 in 2021. Korea TB incidence peaked at 20–24 for both sexes in 1990, shifting to 65–69 in 2021. After age 65, the number of cases decreased in both years. In 1990, males had more cases than females in most age groups, except for the 0–14 and ≥80 groups. By 2021, males still had more cases than females in most age groups, except for 0–14 and ≥85. Globally, males had a higher incidence rate than females in most age groups in both 1990 and 2021. In Korea, TB mortality rates were higher for males in most age groups in 1990. Female mortality peaked in the 75–79 age group, while for males, it peaked in the 70–74 age group. By 2021, Korea TB mortality was higher among females aged 30–89. The peak mortality shifted to 80–84 for males and 85–89 for females. Overall, TB mortality increased with age, peaking before gradually declining. Globally, males had higher mortality than females in most age groups in both years. Korea’s DALYs in 1990 mirrored mortality trends, with males higher in most age groups; DALYs in females peaked at 30–34, while DALYs in males peaked at 50–54. By 2021, both peaked at 80–84. Globally, males had higher DALYs in most age groups in both years. These findings highlight persistent sex disparities in the TB burden across age groups in Korea and worldwide.

Figure 3.

Comparison of TB prevalence, incidence, mortality, and DALYs by sex and age groups in Korea in 1990 and 2021. (A) Prevalence in 1990; (B) incidence in 1990; (C) mortality in 1990; (D) DALYs in 1990; (E) prevalence in 2021; (F) incidence in 2021; (G) mortality in 2021; (H) DALYs in 2021. TB, tuberculosis; DALY, disability-adjusted life year. This figure was illustrated by the author.

Figure 4 and Suppl. 7 compare the TB burden by sex in Korea from 1990 to 2021. The ASPR consistently decreased for both sexes, with males consistently higher. The sex gap widened from 1990 to 1996, peaked, then narrowed before rising again after 2017, reaching another peak in 2020 before decreasing in 2021 (Suppl. 8). This pattern contrasts with the global trend, where the sex gap in ASPR decreased over time (Suppl. 9). The ASIR decreased steadily for both sexes, with males consistently having higher ASIRs than females. However, the gap between the sexes narrowed over this period, aligning with global trends. ASMR and ASDR showed similar trends to the ASIR, although sex disparity patterns in Korea differed from global trends.

Figure 4.

Comparison of all-age cases and age-standardized rates of prevalence, incidence, mortality, and DALYs by sex in Korea, 1990–2021. (A) Prevalent cases and ASPR; (B) incident cases and ASIR; (C) deaths and ASMR; (D) DALY counts and ASDR. Bar charts represent counts, and lines represent age-standardized rates. DALY, disability-adjusted life year; ASPR, age-standardized prevalence rate; ASIR, age-standardized incidence rate; ASMR, age-standardized mortality rate; ASDR, age-standardized disability-adjusted life-year rate. This figure was illustrated by the author.

Age–period–cohort analysis in Korea and globally

For age effects, incidence in Korea decreased until age 22, rose between ages 23–38, decreased until age 78, increased from ages 79 to 82, and then decreased again at ages ≥83. Globally, a similar pattern was observed: incidence decreased until age 22, increasing until age 38, and then declined steadily. In Korea, prevalence declined until ages 18–22, increased until age 38, and then decreased steadily. Globally, prevalence increased until age 42, followed by a consistent decline. In Korea, mortality showed a slight increase in ages 28–32 and 88–92, while otherwise declining steadily. Globally, mortality decreased until age 28, increased until age 78, and then declined thereafter. In Korea, the DALYs decreased until age 28, increased until age 32, and then steadily declined. Globally, they decreased until age 28, rose to age 38, and then steadily decreased (Suppl. 10, 11).

For period effects, the rate ratio (RR) is defined as the ratio of the indicator rate in 2021 to that in 1990, where rates are cases per population per year. In Korea, the RRs for incidence, prevalence, mortality, and DALYs all declined over time. Globally, mortality and DALYs RRs also declined steadily, but incidence RR decreased until 2014 and then rose. In contrast, the prevalence RR increased throughout (Suppl. 10, 11).

For cohort effects, both incidence and prevalence RR increased until the 1922 cohort, declined until 1932, rose again until 1942, and then steadily decreased thereafter. Mortality and DALYs RR increased until the 1907 cohort and then declined consistently. A similar trend was observed globally, with mortality and DALYs RR increasing until 1902 and then declining thereafter (Suppl. 10, 11).

Forecasting TB incidence and mortality in Korea and globally

Using the R forecast package, ARIMA models were applied to TB incidence and mortality rates in Korea and globally, with model specifications and parameters provided in Suppl. 12. Based on the selected models, forecasts for 2022–2030 indicate that in Korea, the ASIR will approach zero by 2030, with male ASMR decreasing to 4.37 per 100,000 and female ASMR to 1.90 per 100,000. Globally, the male ASIR is projected to decline to 96.54 per 100,000, and 73.31 per 100,000 for females. The global ASMR is projected to fall to 10.64 per 100,000 for males and 4.61 per 100,000 for females (Suppl. 13, 14).

Key results

Using data from the GBD 2021 database, we evaluated TB burden in Korea and globally from 1990 to 2021 using incidence, prevalence, mortality, and DALYs. Over this period, Korea experienced faster declines in TB burden than global trends. Incident cases decreased by 43.36%, and ASIR decreased by 72.2% (AAPC, –4.00%), compared with a 40.47% decline in global ASIR (AAPC, –1.65%). Prevalence decreased by 52.18% in Korea but increased globally. Mortality and DALYs decreased by 67.60% and 81.53%, respectively, in Korea. Age-standardized mortality declined by 90.37%. The TB burden shifted toward older adults and remained higher among males. Forecasts suggest near-elimination by 2030.

Interpretation and comparison with previous studies

Although both Korea and the global population experienced overall declines across these indicators, a slight increase in new cases and mortality was observed in 2021 [3], likely reflecting healthcare disruptions related to the coronavirus disease 2019 pandemic [16,17]. These findings indicate that the global TB burden remains insufficiently controlled and underscore the need for sustained, targeted public health interventions worldwide.

In contrast to global trends, Korea experienced a greater reduction in TB burden from 1990 to 2021, which may reflect the impact of robust national initiatives such as the “2030 TB Elimination Plan” and the National Strategic Plan for TB control. ARIMA forecasts suggest that TB incidence in Korea may approach zero by 2030, consistent with the potential effectiveness of these efforts. However, Korea continues to have a relatively high TB burden, which may be related to socioeconomic conditions and risk factors—particularly high smoking prevalence, alcohol consumption, and diabetes prevalence—that are associated with increased susceptibility [18–23]. These findings underscore the importance of integrating behavioral and metabolic health interventions into TB control programs. The decline in period effects is consistent with the scale-up of national TB strategies, including the expansion of public–private mix collaboration, which has strengthened treatment adherence and outcomes through standardized patient monitoring [3]. In addition, cohort effects may reflect population aging, as older generations exposed to poverty and overcrowding in the post-war era may be more likely to experience reactivation of latent TB infection with immunosenescence [3]. Our findings also indicate age- and sex-related disparities in TB burden in Korea and globally. In Korea, TB incidence was higher in younger age groups, increased with age, and then decreased in the oldest age groups. Globally, TB prevalence was highest among middle-aged adults, whereas mortality rates were higher among children younger than 5 years. Sex differences were also evident, with males consistently showing higher incidence and mortality. This disparity may be related to higher prevalence of risk behaviors (e.g., smoking and alcohol use) among males, greater occupational exposure, and differences in healthcare-seeking behavior [3,18,20]. Collectively, these results support the need for age- and sex-sensitive TB control strategies.

Implications and suggestions for further studies

Overall, although Korea has made substantial progress in TB control, achieving elimination will require comprehensive strategies that address underlying determinants and population-specific risk factors. Despite limitations related to the quality and update frequency of the GBD dataset, this study provides insights into long-term TB trends and future projections [24]. Future research should incorporate more granular data, including International Classification of Diseases, 10th edition–based TB subtypes and region-specific estimates, to better capture disease heterogeneity and regional disparities. Such efforts are essential for developing targeted, equitable, and effective public health responses in the continued fight against TB.

Conclusion

From 1990 to 2021, Korea experienced a significant decline in its TB burden, surpassing global improvements in incidence, mortality, and DALYs. Nevertheless, the burden has increasingly affected older adults and remains disproportionately higher among males. To achieve the projected near-elimination of TB by 2030 while maintaining equity, sustained and targeted prevention, screening, and treatment efforts focused on vulnerable populations are imperative.