Thyroid disease burden among individuals aged 0–19 years in low-SDI countries, 1990–2035: a systematic analysis of the Global Burden of Disease Study 2023

Introduction

Thyroid diseases encompass a spectrum of non-neoplastic conditions affecting thyroid structure and function, including congenital hypothyroidism, autoimmune thyroiditis, iodine deficiency disorders, and simple goitre (1). During childhood and adolescence, thyroid hormones are critical for normal growth, neurodevelopment, and metabolic regulation; even mild dysfunction can have lasting consequences on cognitive and physical development (2). The aetiology of thyroid disease varies by epidemiological context: in high-income settings, autoimmune thyroiditis (Hashimoto’s disease) is commonly reported, whereas in low Socio-demographic Index (SDI) settings, iodine deficiency remains an important driver of goitre and hypothyroidism (3).

Despite global efforts to eliminate iodine deficiency through universal salt iodization (USI), iodine deficiency persists in many low-SDI countries, where dietary diversity, food fortification coverage, and monitoring capacity remain uneven (4,5). In these settings, iodine deficiency not only causes goitre but also increases the risk of congenital hypothyroidism, which, if untreated, can lead to irreversible intellectual disability (6). Beyond iodine deficiency, autoimmune thyroiditis, subclinical hypothyroidism, and isolated hyperthyrotropinaemia also contribute to pediatric morbidity, but their detection depends heavily on diagnostic capacity and follow-up systems (7,8).

Existing evidence on pediatric thyroid disease in low-resource settings remains fragmented. Long-term trend studies have usually focused on iodine deficiency, thyroid cancer, or selected national and regional populations rather than the full spectrum of non-neoplastic thyroid diseases among individuals aged 0–19 years (4,9,10). Many available studies have used school-based, regional, or cross-sectional designs, which limits comparability across countries and over time because case definitions, diagnostic access, and population coverage vary substantially (9,10). This evidence gap is particularly important from a health-systems perspective. Low-SDI countries are not synonymous with World Bank low-income countries; therefore, clarifying the GBD-defined low-SDI framework is necessary before interpreting the findings. In these settings, potentially treatable thyroid disorders may cause disproportionate disability-adjusted life years (DALYs) when iodized salt monitoring, newborn screening, thyroid function testing, pediatric endocrine follow-up, surgical referral for severe goitre, and reliable levothyroxine supply are unavailable or inequitably distributed.

To address these gaps, we used Global Burden of Disease (GBD) 2023 estimates to provide a standardized assessment of non-neoplastic thyroid disease burden among individuals aged 0–19 years across 50 GBD-defined low-SDI countries. We focused on temporal trends, age-sex patterns, cross-country inequality, frontier efficiency gaps, and projected burden to 2035. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0191/rc).

Methods

Study overview and case definition

This study presents GBD 2023 estimates for non-neoplastic thyroid diseases among individuals aged 0–19 years in 50 countries classified as low-SDI locations in the GBD 2023 framework. SDI is a composite measure based on lag-distributed income per capita, average educational attainment, and total fertility rate (11). In the GBD framework, thyroid diseases are defined as all non-malignant conditions of the thyroid gland, including congenital hypothyroidism (ICD-10 codes E03.0–E03.1), autoimmune thyroiditis (Hashimoto’s disease, E06.3), iodine deficiency disorders (E00–E02), and simple goitre (E04). Thyroid cancer (C73) is classified separately under neoplasms and is excluded from this analysis (12). The study population was stratified into <28 days, 1–5 months, 6–11 months, 12–23 months, 2–4 years, 5–9 years, 10–14 years, and 15–19 years, with males and females analysed separately.

GBD 2023 estimates were produced by an international network of collaborators who identified, reviewed, and modelled available data; across all metrics, GBD 2023 incorporated input from more than 14 000 collaborators from more than 160 countries and territories (12,13). The 2023 endpoint reflects the standard lag in GBD data collation, cause-specific modelling, internal validation, and public release cycles; thus, 2023 represented the most recent complete year available from GBD 2023 at the time this analysis was conducted. The estimates were publicly accessed through the GBD Results Tool (https://vizhub.healthdata.org/gbd-results/). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Data sources and processing

Data sources used to produce GBD 2023 estimates are listed in the GBD 2023 Sources Tool (https://ghdx.healthdata.org/gbd-2023/sources). Metadata and citations for study-specific input data can be obtained by selecting “nonfatal health outcomes” as the component, “thyroid diseases” as the cause, and the 50 low-SDI countries for locations. Input data included vital registration systems, household surveys, disease registries, hospital records, claims or administrative data where available, and published scientific literature. Because historical surveillance and vital registration coverage were sparse in many low-SDI countries, especially in the early years of the 1990–2023 series, GBD did not treat observed primary data as complete enumerations. Instead, data were cleaned, mapped to standardized case definitions, adjusted for known biases where possible, and synthesized using DisMod-MR 2.1, a Bayesian meta-regression framework that borrows strength across age, sex, location, year, and related epidemiological parameters to generate internally consistent prevalence and health-loss estimates (12,13).

Metrics

We extracted prevalence counts, prevalence rates, DALY counts, and DALY rates for thyroid diseases among individuals aged 0–19 years by age group, sex, country, and year (1990–2023). DALYs are the sum of years of life lost due to premature mortality (YLLs) and years lived with disability (YLDs) (14). YLLs are derived from cause-specific mortality estimates, whereas YLDs are estimated as sequela-specific prevalence multiplied by corresponding disability weights (14). All rates are reported per 100 000 population with 95% uncertainty intervals (UI). SDI values for each country were obtained from GBD 2023.

Age and sex stratification

Age- and sex-specific estimates were compared between 1990 and 2023 to identify demographic patterns and changes over time. The highest burden groups were highlighted, and absolute and relative changes were calculated.

Statistical analysis

Temporal trend analysis

Joinpoint regression (Joinpoint Regression Program, version 5.1.0.0, National Cancer Institute) was used to identify significant changes in prevalence and DALY rates from 1990 to 2023 (15). Conceptually, joinpoint regression fits a series of connected log-linear segments and identifies calendar years in which the slope changes statistically, allowing clinical readers to distinguish long periods of gradual change from short periods of acceleration, stagnation, or reversal. The annual percent change (APC) was calculated for each segment with 95% confidence intervals (CI). The average annual percent change (AAPC) over the entire study period was computed as a weighted average of segment-specific APCs. Trends were classified as increasing if the APC or AAPC and its 95% CI were both greater than 0, decreasing if both were less than 0, and stable if the 95% CI included 0.

Correlation with SDI

Spearman rank correlation was used to assess the relationship between thyroid disease burden (prevalence rate and DALY rate) and SDI across the 50 countries in 2023 (16). Correlation coefficients (R) with 95% CIs and P values were computed; statistical significance was set at P<0.05.

Inequality analysis

Cross-country inequality was quantified using the slope index of inequality (SII) and concentration index (17). SII measures absolute inequality in burden across the SDI gradient, estimated via weighted regression of disease rates on the relative rank of SDI. A negative SII indicates that disease burden is disproportionately concentrated in lower-SDI countries. The concentration index measures relative inequality by quantifying the extent to which disease burden is concentrated among countries with lower versus higher SDI; it ranges from −1 to 1, with negative values indicating concentration among disadvantaged populations and positive values indicating concentration among advantaged populations. These indices are widely used in health inequality research due to their ability to reflect both absolute and relative disparities across socioeconomic gradients. Both indices were calculated for 1990 and 2023 to assess changes over time.

Frontier analysis

Frontier analysis was conducted to evaluate health system performance in achieving minimal thyroid disease burden relative to socioeconomic development levels. Using non-parametric data envelopment analysis, we established a theoretical frontier representing the lowest achievable prevalence and DALY rates at each SDI level based on the performance of best-practice countries. The efficiency difference (eff_diff) was calculated as the difference between a country’s observed rate and the frontier-predicted rate at its SDI level, representing the avoidable disease burden and potential for improvement. Countries with small efficiency differences were identified as performance benchmarks for their SDI level (18).

Projection analysis

Bayesian age-period-cohort (BAPC) models were applied to project prevalence and DALY burden from 2023 to 2035 based on historical age-specific estimates from 1990 to 2023. The projection analysis followed the BAPC framework implemented in the R BAPC package, which fits a Bayesian APC model using a Poisson likelihood and integrated nested Laplace approximation (INLA) (19). In this framework, the age-period data are organized in a Lexis structure, with rows representing calendar periods and columns representing age groups.

For each metric based on counts, the model can be expressed as:

Ya,t∼Poisson(μa,t)log(μa,t)=log(Pa,t)+ηa,tηa,t=α+αa+βt+γc[1]

Where Y_a,t denotes the observed burden count in age group a and calendar year t, P_a,t denotes the corresponding population denominator, and µ_a,t denotes the expected burden count. The term log(P_a,t) is included as an offset so that the model estimates age-specific burden rates. The linear predictor η_a,t includes an intercept α, an age effect α_a, a period effect β_t, and a cohort effect γ_c, where the birth cohort is determined by c=t−a.

The age effect captures biological development and age-specific vulnerability. The period effect captures calendar-time influences that affect all age groups, such as changes in diagnostic access, iodine nutrition programmes, health-service coverage, or pandemic-related service disruption. The cohort effect captures differences shared by individuals born in the same period, reflecting early-life exposures or cohort-specific risks. Consistent with the BAPC framework, smoothing priors were specified for age, period, and cohort effects, with second-order random walk priors used to reflect the assumption that adjacent effects change gradually over time. INLA was used to approximate posterior distributions and generate projected counts and rates with 95% prediction intervals.

Calculating uncertainty

Uncertainty was propagated throughout the estimation process by taking 250 draws from posterior distributions. Mean estimates represent the mean across draws, and 95% UI were calculated as the 2.5th and 97.5th percentiles of the draws. For projection analysis, 95% prediction intervals reflect both model and parameter uncertainty (12). All data analysis and visualization were conducted using R (version 4.4.2).

Results

Overview of thyroid diseases burden

In 1990, the total population aged 0–19 years in the 50 low-SDI countries was approximately 440.5 million; by 2023, this denominator had increased to approximately 847.7 million. The total number of prevalent cases of thyroid disease among individuals aged 0–19 years increased from 6.22 million (95% UI: 4.38–8.80 million) in 1990 to 11.39 million (95% UI: 7.87–16.58 million) in 2023, while the prevalence rate slightly decreased from 1,412.19 per 100,000 population (95% UI: 994.91–1,997.73) to 1,343.90 per 100,000 (95% UI: 928.43–1,956.07). In terms of DALYs, total DALYs increased from 0.19 million (95% UI: 0.13–0.29 million) to 0.32 million (95% UI: 0.23–0.49 million), whereas the DALY rate decreased from 44.14 per 100,000 (95% UI: 28.83–65.94) to 38.25 per 100,000 (95% UI: 26.71–57.93) (Tables S1,S2).

Substantial heterogeneity in disease burden was observed across low-SDI countries. In 2023, the countries with the highest prevalence rates were Haiti (2,752.05 per 100,000), Vanuatu (2,569.71 per 100,000), and Somalia (2,236.46 per 100,000), while the lowest prevalence rates were observed in Mali (1,060.04 per 100,000), Cameroon (1,086.61 per 100,000), and Rwanda (1,088.56 per 100,000). The countries with the highest DALY rates were Haiti (85.60 per 100,000), Niger (72.05 per 100,000), and South Sudan (68.85 per 100,000), whereas the lowest DALY rates were found in Nepal (19.15 per 100,000), Bangladesh (20.62 per 100,000), and Bhutan (21.75 per 100,000) (Tables S1,S2; Figure S1).

Distribution by age and sex

In 2023, the prevalence and number of prevalent cases of thyroid disease in low-SDI countries showed marked age and sex disparities (Figure 1A,1B). Females had higher numbers of cases and prevalence rates than males across all age groups, with both metrics increasing with age and peaking in the 15–19-year age group. In this age group, the number of prevalent cases in females was 2.83 million (95% UI: 1.65–4.59 million), with a prevalence rate of 3,065.69 per 100,000 (95% UI: 1,793.04–4,974.97); in males, the number was 1.37 million (95% UI: 0.79–2.27 million), with a prevalence rate of 1,470.33 per 100,000 (95% UI: 849.90–2,432.72). The highest DALY rates were observed in females aged 15–19 years (69.93 per 100,000, 95% UI: 40.75–120.01), followed by females aged 10–14 years (44.10 per 100,000, 95% UI: 25.75–74.64). Notably, DALY rates in the neonatal period (<28 days) were high in both sexes, with males at 150.89 per 100,000 (95% UI: 49.74–305.90) and females at 108.51 per 100,000 (95% UI: 38.80–216.47).

Figure 1 Age and sex differences in thyroid disease burden among individuals aged 0–19 years in low-SDI countries. (A,B) Prevalence and DALY rates by age group and sex in 2023. (C,D) Compare prevalence and DALY rates by age group between 1990 and 2023. DALYs, disability-adjusted life years; SDI, Socio-demographic Index; UI, uncertainty interval.

Compared to 1990, the number of prevalent cases in all age groups generally increased in 2023, with the most pronounced increase in the 15–19-year age group (from 2.03 to 4.20 million). However, prevalence rates decreased in most age groups; the rate in the 15–19-year age group declined from 2,479.87 to 2,263.89 per 100,000, with other age groups showing declines of varying magnitudes. Regarding DALYs, counts in 2023 were higher than in 1990 for all age groups, with the highest count in the 15–19-year age group (90,409, 95% UI: 54,092–156,721). DALY rates decreased in most age groups compared to 1990, with the most substantial decline in the 5–9-year age group. However, the DALY rate in the neonatal period remained high (130.23 per 100,000, 95% UI: 58.07–218.98) and showed a smaller reduction (Figure 1C,1D).

Temporal trends

From 1990 to 2023, the overall prevalence rate of thyroid disease among individuals aged 0-19 years in low-SDI countries remained relatively stable, with an AAPC of −0.06 (95% CI: −0.13 to 0.01). However, this stable trend was not linear. Joinpoint regression analysis revealed three distinct trend phases: a slow increase from 1990–2000 (APC =0.63, 95% CI: 0.51–0.75); a significant decrease from 2000 to 2014 (APC =−0.91, 95% CI: −1.00 to −0.83); and a subsequent increase from 2014 to 2023 (APC =0.49, 95% CI: 0.32–0.66). The overall DALY rate showed a significant decreasing trend over the study period, with an AAPC of −0.46 (95% CI: −0.53 to −0.40), representing an average annual decline of 0.46%. Joinpoint regression identified five phases for DALY rate trends: a slow decline from 1990 to 1999 (APC =−0.29, 95% CI: −0.36 to −0.22); an accelerated decline from 1999 to 2005 (APC =−2.04, 95% CI: −2.19 to −1.89); a decelerated decline from 2005 to 2013 (APC =−0.75, 95% CI: −0.84 to −0.65); a significant increase from 2013 to 2020 (APC =1.14, 95% CI: 1.02–1.26); and a return to decline from 2020 to 2023 (APC =−0.81, 95% CI: −1.26 to −0.36) (Figure 2).

Figure 2 Joinpoint regression analysis of thyroid disease prevalence and DALY rates, 1990–2023. Joinpoint regression identifying significant changes in the trends of prevalence and DALY rates of thyroid diseases among individuals aged 0–19 years in low-SDI countries, 1990–2023. Annual percent changes and 95% confidence intervals are shown for each significant segment. APC, annual percent change; DALYs, disability-adjusted life years; SDI, Socio-demographic Index.

At the country level, trends in prevalence rates exhibited significant heterogeneity. Among the 50 countries, 30 showed increasing trends in prevalence (23 statistically significant), and 20 showed decreasing trends (16 statistically significant). The most pronounced increases were observed in South Sudan (AAPC =0.73, 95% CI: 0.66–0.81), Yemen (0.48, 95% CI: 0.43–0.54), and the Democratic Republic of the Congo (0.33, 95% CI: 0.29–0.37); while the largest decreases were in Ethiopia (AAPC =−0.63, 95% CI: −0.74 to −0.52), Cambodia (−0.46, 95% CI: −0.50 to −0.42), and Guinea (−0.38, 95% CI: −0.46 to −0.31) (Table S1). DALY rate trends were also heterogeneous. Eleven countries showed increasing DALY rate trends (7 statistically significant), and 39 showed decreasing trends (33 statistically significant). The most substantial declines were in the United Republic of Tanzania (AAPC =−1.54, 95% CI: −1.87 to −1.21), Mozambique (−1.43, 95% CI: −1.70 to −1.16), and Rwanda (−1.42, 95% CI: −1.91 to −0.93); while the most marked increases were in Sao Tome and Principe (AAPC =0.85, 95% CI: 0.40–1.31), South Sudan (AAPC =0.58, 95% CI: 0.29–0.86), and Afghanistan (AAPC =0.57, 95% CI: 0.28–0.86) (Table S2).

Correlation with SDI

In 2023, there was no significant correlation between the prevalence rate of thyroid disease among individuals aged 0–19 years in low-SDI countries and SDI (R=0.07, 95% CI: −0.20 to 0.33, P=0.64). Conversely, the DALY rate showed a significant negative correlation with SDI (R=−0.36, 95% CI: −0.61 to −0.02, P=0.01), suggesting that the health loss due to thyroid disease was more severe in countries with lower socioeconomic development (Figure S2).

Inequality analysis

Analysis using concentration curves, the SII, and the concentration index showed that the SII for prevalence decreased from 298.91 (95% CI: −46.44 to 644.27) in 1990 to 168.72 (95% CI: −211.53 to 548.98) in 2023, while the concentration index increased from −0.00 (95% CI: −0.25 to 0.22) to 0.15 (95% CI: −0.17 to 0.35). For DALY rate, the SII changed from −18.52 (95% CI: −34.45 to −2.59) in 1990 to −22.24 (95% CI: −35.71 to −8.77) in 2023, and the concentration index shifted from −0.05 (95% CI: −0.29 to 0.16) to 0.08 (95% CI: −0.20 to 0.32) (Figure 3).

Figure 3 Cross-country inequality in thyroid disease burden by SDI. Concentration curves and SII for prevalence (A,B) and DALY rates (C,D) of thyroid diseases among individuals aged 0–19 years across 50 low-SDI countries, 1990 and 2023. SII and CI with 95% confidence intervals are shown. CI, concentration index; DALYs, disability-adjusted life years; SDI, Socio-demographic Index; SII, slope index of inequality.

Frontier analysis

In terms of prevalence, the countries with the largest efficiency differences (observed rate minus frontier rate) in the low-SDI region were Haiti (1,743.44 per 100,000), followed by Vanuatu (1,560.90 per 100,000), Cambodia (1,222.47 per 100,000), Somalia (1,182.88 per 100,000), and Mauritania (1,168.93 per 100,000). The countries with the smallest efficiency differences were Mali (51.50 per 100,000), Cameroon (77.80 per 100,000), Rwanda (79.69 per 100,000), Uganda (83.50 per 100,000), and Zimbabwe (97.92 per 100,000). For DALY rate, the largest efficiency differences were observed in Haiti (66.43 per 100,000), followed by South Sudan (49.17 per 100,000), Guinea-Bissau (48.13 per 100,000), Sao Tome and Principe (47.83 per 100,000), and Liberia (46.54 per 100,000). The smallest efficiency differences were found in Nepal (0.06 per 100,000), Zimbabwe (4.05 per 100,000), Solomon Islands (10.94 per 100,000), Democratic Republic of the Congo (11.62 per 100,000), and Central African Republic (14.26 per 100,000) (Figure 4).

Figure 4 Frontier analysis of health system performance in thyroid disease control. Frontier analysis evaluating health system efficiency in achieving minimal thyroid disease burden relative to sociodemographic development. Each point represents a low-SDI country. The solid line represents the theoretical frontier—the lowest achievable prevalence (A) or DALY rate (B) at each SDI level based on best‑performing countries. The efficiency difference (vertical distance from the frontier) indicates avoidable burden. DALYs, disability-adjusted life years; SDI, Socio-demographic Index.

Projected trends to 2035

Based on historical trends from 1990 to 2023, the BAPC model was used to project the prevalence and DALY burden among individuals aged 0–19 years in low-SDI countries to 2035 (Figure 5). The total number of prevalent cases is projected to increase from 11.39 million in 2023 to 18.37 million in 2035 (95% UI: 0.18–36.56 million). The projected mean prevalence rate in 2035 was 1,181.47 per 100,000 (95% UI: 11.47–2,351.48). Total DALYs are projected to increase from 0.32 million in 2023 to 0.37 million in 2035 (95% UI: 0.002–0.75 million), whereas the projected mean DALY rate in 2035 was 24.10 per 100,000 (95% UI: 15.85–32.35).

Figure 5 Projected burden of thyroid diseases among individuals aged 0–19 years to 2035. BAPC projections of thyroid disease prevalence and DALY rates among individuals aged 0–19 years in 50 low-SDI countries, 2023–2035. Solid lines represent historical trends (1990–2023); dashed lines and shaded areas represent projected mean values and 95% prediction intervals. ASDR, age-standardized DALY rate; ASPR, age-standardized prevalence rate; BAPC, Bayesian age-period-cohort; DALYs, disability-adjusted life years; SDI, Socio-demographic Index.

Discussion

This study provides a comprehensive assessment of non-neoplastic thyroid disease burden among individuals aged 0–19 years in 50 GBD-defined low-SDI countries using data from the GBD 2023 study. By applying joinpoint regression, inequality analysis, frontier analysis, and BAPC models, we systematically examined temporal trends, demographic patterns, cross-country disparities, health system performance gaps, and future projections. Our findings fill a critical gap in understanding thyroid disease burden in pediatric populations living in resource-limited settings and provide an evidence base for targeted interventions and resource allocation.

From 1990 to 2023, the overall prevalence rate of thyroid disease among individuals aged 0–19 years in low-SDI countries remained relatively stable. However, the absolute number of prevalent cases nearly doubled, increasing from 6.22 to 11.39 million. This divergence indicates that health systems must manage many more affected individuals even when individual-level risk does not increase (20). In contrast, DALY rates declined significantly over the study period, with an average annual reduction of 0.46%. This improvement is consistent with global progress in iodine nutrition: the proportion of households consuming adequately iodized salt in low- and middle-income countries increased from approximately 20% in 1990 to over 85% by 2020 (21).

The three-phase trend in prevalence—an increase from 1990 to 2000 (APC 0.63), a decrease from 2000 to 2014 (APC −0.91), and a subsequent increase from 2014 to 2023 (APC 0.49)—requires careful interpretation. The early increase may reflect improved case ascertainment and health information systems rather than a true rise in incidence. The decline during 2000–2014 coincides with the global scale-up of USI following the 1990 World Summit for Children and the 2002 UN Special Session on Children, which galvanised international commitment to eliminating iodine deficiency (22). The recent increase since 2014 is concerning because it may signal stagnation or partial reversal of these gains, particularly in settings where enforcement, monitoring, or salt supply chains are vulnerable. Our finding that conflict-affected countries such as South Sudan (AAPC 0.73) and Yemen (AAPC 0.48) experienced rising prevalence supports this interpretation, underscoring the need for sustained political commitment and robust monitoring systems. The COVID-19 pandemic represents an important potential external disturbance during the final years of the study period. Although we did not model COVID-19 as a separate exposure, the non-linear AAPC and joinpoint analyses provide a framework for observing whether pandemic-era years coincided with changes in slope. In our results, DALY rates increased during 2013–2020 (APC 1.14) and then declined during 2020–2023 (APC −0.81). This aggregate decline does not rule out pandemic-related disruption, because lockdowns, reduced antenatal care, interruptions in routine screening, medicine shortages, and weakened USI monitoring may have affected specific countries or subgroups without producing a uniform regional reversal. Recent repeated cross-sectional evidence also suggests that the COVID-19 period could alter iodine nutrition status among pregnant women and infants/toddlers, supporting the biological and service-delivery plausibility of such interference (23).

Marked age and sex disparities were observed in thyroid disease burden. Females had higher prevalence and DALY rates than males across all age groups, peaking in adolescence. In the 15–19-year age group, the prevalence rate in females was 3,065.69 per 100,000, compared with 1,470.33 per 100,000 in males; the DALY rate in females was 69.93 per 100,000, compared with 27.85 per 100,000 in males. This pattern aligns with the natural history of autoimmune thyroiditis, which typically manifests during puberty and is two to four times more common in females (24). The female predominance is biologically plausible given the immunomodulatory effects of estrogen and the higher prevalence of autoimmune thyroiditis in females, as documented in settings where autoimmune testing is routine (25). In low-SDI settings, where such testing is less available, these cases may be underdiagnosed or misclassified as simple goitre, contributing to the observed high prevalence and potentially increasing the risk of disease progression (7). The higher DALY rate in adolescent females may reflect the consequences of untreated autoimmune thyroiditis and hypothyroidism, including effects on educational attainment, reproductive health, and long-term quality of life. These findings support targeted detection through school or primary-care platforms, particularly for adolescent females (26).

The neonatal period (<28 days) exhibited strikingly high DALY rates, with a much smaller reduction between 1990 and 2023 compared with other age groups. This persistent burden is an important marker of incomplete maternal-neonatal thyroid health services. Untreated congenital hypothyroidism leads to irreversible intellectual disability, yet recent worldwide coverage evidence indicates that most newborns globally are still not screened for congenital hypothyroidism, with the largest gaps concentrated in settings where health-system and laboratory capacity remain limited (27). Even where screening exists, coverage is often low because of limited facility-based deliveries, delayed sample transport, and inadequate infrastructure for confirmatory testing and follow-up (28). Maternal iodine deficiency remains a major contributor: studies from sub-Saharan Africa have documented that up to 40% of pregnant women have insufficient iodine intake, placing their newborns at risk (29). The slow decline in neonatal DALY rate indicates that interventions targeting pregnancy and the neonatal period have lagged behind progress in other age groups. Dried blood spot testing for congenital hypothyroidism, integrated into immunization or maternal health platforms, could offer a scalable solution in low-resource settings (30). Strengthening maternal iodine nutrition through antenatal supplementation and ensuring universal access to iodized salt should be prioritised alongside the phased introduction of newborn screening programmes.

Substantial heterogeneity in disease burden was observed across low-SDI countries, highlighting the importance of national context. High-burden countries included Haiti, Vanuatu, Somalia, Niger, and South Sudan. For countries such as Haiti, Somalia, and South Sudan, published evidence describes fragile health systems, humanitarian stressors, iodine deficiency, or limited primary health-care capacity, which may amplify thyroid-related health loss once disease occurs (31-33). These country-level interpretations should remain cautious because the present analysis was ecological and based on GBD aggregate estimates rather than direct measurements of iodized salt coverage, screening coverage, or treatment access. Conversely, several low-SDI countries showed smaller frontier efficiency gaps. The low prevalence efficiency gap in Mali and the near-zero DALY-rate efficiency gap in Nepal suggest that better-than-expected outcomes are possible even at low-SDI levels. These findings should be interpreted as signals for policy learning rather than proof of a single causal programme. Potential contributors include sustained iodine monitoring, primary health-care delivery platforms, community-based maternal and child health services, and partner-supported nutrition programmes (34-37). Countries with large efficiency gaps, such as Haiti and South Sudan, may require priority support for iodized salt quality monitoring, neonatal screening capacity, referral pathways, and reliable thyroid medication supply.

DALY rate showed a significant negative correlation with SDI, consistent with the broader literature demonstrating that socioeconomic development is a key determinant of health outcomes (38). Higher SDI is associated with better healthcare access, higher educational attainment, and improved nutrition—all of which contribute to lower disease burden. However, prevalence rate showed no significant correlation with SDI, suggesting that the occurrence of thyroid disease may be less sensitive to SDI than its consequences. Once disease develops, outcomes depend critically on health system capacity to diagnose and treat effectively. This finding aligns with the concept of the “prevention paradox” and underscores the need for both primary prevention (iodine supplementation) and secondary prevention (early detection and treatment) (4,26). Integrating comprehensive thyroid health services—including diagnosis, treatment, and follow-up—into universal health coverage (UHC) packages is essential to bridge the gap between disease occurrence and adverse outcomes.

Inequality analysis revealed complex but coherent patterns. The SII for prevalence decreased from 298.91 in 1990 to 168.72 in 2023, indicating a narrowing of the absolute gap in disease occurrence across the SDI gradient. However, the concentration index increased from approximately zero to 0.15, suggesting that relative prevalence inequality shifted slightly toward countries with higher SDI within this low-SDI stratum, although CI were wide. For DALY rate, the SII remained negative and statistically different from zero in both 1990 and 2023, indicating persistent absolute disadvantage in countries with lower SDI. In contrast, the DALY-rate concentration index crossed zero in both years, so relative inequality should be interpreted cautiously rather than as a robust directional shift. This divergence between prevalence and DALY inequality mirrors our earlier observation that disease occurrence is relatively evenly distributed across low-SDI countries, whereas health loss depends more strongly on health-system capacity to diagnose and treat disease once it occurs (38,39).

Based on historical trends from 1990 to 2023, BAPC models projected that the total number of prevalent cases in low-SDI countries will continue to increase, reaching 18.37 million by 2035, whereas the projected mean prevalence rate is expected to be 1,181.47 per 100,000. In contrast, the projected mean DALY rate is expected to decline further to 24.10 per 100,000. These projections reflect anticipated population growth and possible continued improvements in disease management, but they also imply growing demand for thyroid-related services as the absolute number of affected individuals expands. However, these projections assume that historical relationships among age, period, and cohort effects continue. If iodized salt coverage erodes because of policy neglect, conflict, supply-chain disruptions, or if climate change exacerbates food insecurity, these gains could be reversed (27,31,33). The wide prediction intervals around prevalence projections highlight the need for robust monitoring systems and adaptive planning.

There are several limitations in this study. First, data availability in low-SDI countries is often sparse, and estimates rely on statistical modelling, which introduces uncertainty, particularly in conflict-affected settings and countries with weak health information systems. Second, GBD 2023 does not distinguish between subtypes of non-neoplastic thyroid disease, limiting our ability to identify specific aetiological drivers and design targeted interventions. Third, although the AAPC and joinpoint analyses captured non-linear changes across the whole time series, the analysis could not directly measure COVID-19-related disruptions to antenatal care, USI monitoring, laboratory testing, or medicine supply chains; therefore, pandemic-period findings should be interpreted descriptively rather than causally. Fourth, frontier analysis assumes that the best-performing countries represent achievable targets, but contextual factors such as environmental iodine levels, political stability, and health-system organization may affect the feasibility of replicating their success elsewhere. Fifth, projections assume continuation of historical trends and do not account for potential major policy changes, new UHC initiatives, more affordable diagnostic technologies, or negative economic shocks. Sixth, unmeasured confounders such as individual dietary practices, actual access to iodized salt at the household level, healthcare-seeking behaviours, and cultural beliefs may influence the observed associations.

Conclusions

The burden of non-neoplastic thyroid diseases among individuals aged 0–19 years in low-SDI countries is substantial and growing in absolute terms, with persistent sex and age disparities and wide cross-country inequalities. While overall DALY rates are declining, the high neonatal burden and large efficiency gaps in many countries signal missed opportunities for prevention and treatment. Achieving further reductions will require targeted investments in maternal and child health, including strengthening maternal iodine nutrition, implementing dried blood spot newborn screening for congenital hypothyroidism, establishing school-based detection for adolescent females, and ensuring affordable access to levothyroxine and other essential thyroid medicines. Health-system strengthening in the most disadvantaged countries, alongside policy learning from benchmark countries such as Mali and Nepal, is essential to close efficiency gaps. Integrating thyroid health into UHC-oriented maternal, newborn, child, and adolescent health packages offers a practical pathway to reduce avoidable disability.

Acknowledgments

We appreciate the high-quality data provided by the Global Burden of Disease study 2023 collaborators.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0191/rc

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0191/prf

Funding: This research was supported by Zhejiang Provincial Natural Science Foundation of China (No. ZCLY24H0301).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0191/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Taylor PN, Albrecht D, Scholz A, et al. Global epidemiology of hyperthyroidism and hypothyroidism. Nat Rev Endocrinol 2018;14:301-16. [Crossref] [PubMed]
2. Zoeller RT, Rovet J. Timing of thyroid hormone action in the developing brain: clinical observations and experimental findings. J Neuroendocrinol 2004;16:809-18. [Crossref] [PubMed]
3. Zimmermann MB, Boelaert K. Iodine deficiency and thyroid disorders. Lancet Diabetes Endocrinol 2015;3:286-95. [Crossref] [PubMed]
4. Liang D, Wang L, Zhong P, et al. Perspective: Global Burden of Iodine Deficiency: Insights and Projections to 2050 Using XGBoost and SHAP. Adv Nutr 2025;16:100384. [Crossref] [PubMed]
5. Peng F, Lin X, Liu S, et al. Epidemiological trends and disparities in iodine, vitamin A, and iron deficiencies among children aged 0-14 years globally, 1990-2021. Front Nutr 2025;12:1622945. [Crossref] [PubMed]
6. Delange F. The role of iodine in brain development. Proc Nutr Soc 2000;59:75-9. [Crossref] [PubMed]
7. Rallison ML, Dobyns BM, Meikle AW, et al. Natural history of thyroid abnormalities: prevalence, incidence, and regression of thyroid diseases in adolescents and young adults. Am J Med 1991;91:363-70. [Crossref] [PubMed]
8. Biondi B, Cappola AR, Cooper DS. Subclinical Hypothyroidism: A Review. JAMA 2019;322:153-60. [Crossref] [PubMed]
9. Abebe Z, Gebeye E, Tariku A. Poor dietary diversity, wealth status and use of un-iodized salt are associated with goiter among school children: a cross-sectional study in Ethiopia. BMC Public Health 2017;17:44. [Crossref] [PubMed]
10. Nouri Saeidlou S, Babaei F, Ayremlou P, et al. Has iodized salt reduced iodine-deficiency disorders among school-aged children in north-west Iran? A 9-year prospective study. Public Health Nutr 2018;21:489-96.
11. Gu X, Liu T, Qian R, et al. Global burden of developmental intellectual disability in children and adolescents with congenital heart disease, 1990-2021: a population-based study. Transl Pediatr 2026;15:7. [Crossref] [PubMed]
12. Burden of 375 diseases and injuries, risk-attributable burden of 88 risk factors, and healthy life expectancy in 204 countries and territories, including 660 subnational locations, 1990-2023: a systematic analysis for the Global Burden of Disease Study 2023. Lancet 2025;406:1873-922.
13. Global burden of 292 causes of death in 204 countries and territories and 660 subnational locations, 1990-2023: a systematic analysis for the Global Burden of Disease Study 2023. Lancet 2025;406:1811-72.
14. Global incidence, prevalence, years lived with disability (YLDs), disability-adjusted life-years (DALYs), and healthy life expectancy (HALE) for 371 diseases and injuries in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2024;403:2133-61.
15. Kim HJ, Fay MP, Feuer EJ, et al. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med 2000;19:335-51. [Crossref] [PubMed]
16. Spearman C. The proof and measurement of association between two things. Int J Epidemiol 2010;39:1137-50. [Crossref] [PubMed]
17. Bai Z, Han J, An J, et al. The global, regional, and national patterns of change in the burden of congenital birth defects, 1990-2021: an analysis of the global burden of disease study 2021 and forecast to 2040. EClinicalMedicine 2024;77:102873. [Crossref] [PubMed]
18. Liu T, Chen Z, Ge J, et al. The evolving burden of childhood meningitis in low- and middle-income countries, 1990-2021: a decomposition and frontier analysis. Eur J Pediatr 2025;184:679. [Crossref] [PubMed]
19. Riebler A, Held L. Projecting the future burden of cancer: Bayesian age-period-cohort analysis with integrated nested Laplace approximations. Biom J 2017;59:531-49. [Crossref] [PubMed]
20. Global age-sex-specific fertility, mortality, healthy life expectancy (HALE), and population estimates in 204 countries and territories, 1950-2019: a comprehensive demographic analysis for the Global Burden of Disease Study 2019. Lancet 2020;396:1160-203.
21. Zimmermann MB, Andersson M. GLOBAL ENDOCRINOLOGY: Global perspectives in endocrinology: coverage of iodized salt programs and iodine status in 2020. Eur J Endocrinol 2021;185:R13-21. [Crossref] [PubMed]
22. Andersson M, Takkouche B, Egli I, et al. Current global iodine status and progress over the last decade towards the elimination of iodine deficiency. Bull World Health Organ 2005;83:518-25.
23. Jia W, Du Y, Qin Q, et al. The Impact of the COVID-19 Pandemic on Iodine Nutrition Status of Pregnant Women and Infants/Toddlers: Repeated Cross-Sectional Studies in Zhengzhou, China. Biol Trace Elem Res 2026;204:2117-29. [Crossref] [PubMed]
24. Brown RS. Autoimmune thyroiditis in childhood. J Clin Res Pediatr Endocrinol 2013;5:45-49. [Crossref] [PubMed]
25. de Vries L, Bulvik S, Phillip M. Chronic autoimmune thyroiditis in children and adolescents: at presentation and during long-term follow-up. Arch Dis Child 2009;94:33-7. [Crossref] [PubMed]
26. Ford G, LaFranchi SH. Screening for congenital hypothyroidism: a worldwide view of strategies. Best Pract Res Clin Endocrinol Metab 2014;28:175-87. [Crossref] [PubMed]
27. Arrigoni M, Zwaveling-Soonawala N, LaFranchi SH, et al. Newborn screening for congenital hypothyroidism: worldwide coverage 50 years after its start. Eur Thyroid J 2025;14:e240327. [Crossref] [PubMed]
28. Therrell BL Jr, Padilla CD. Newborn screening in the developing countries. Curr Opin Pediatr 2018;30:734-9. [Crossref] [PubMed]
29. Wang T, Tong J, Liu Y, et al. Global, regional, and national burden of iodine deficiency for women of reproductive age, 1990-2021: a systematic analysis based on the Global Burden of Disease Study 2021. Front Nutr 2025;12:1577169. [Crossref] [PubMed]
30. Rose SR, Wassner AJ, Wintergerst KA, et al. Congenital Hypothyroidism: Screening and Management. Pediatrics 2023;151:e2022060419. [Crossref] [PubMed]
31. Dorilien C, Sanabria CAP. The healthcare system in Haiti. Front Public Health 2025;13:1603076. [Crossref] [PubMed]
32. Abdi YH, Abdi MS, Bashir SG, et al. The State of Public Health in Somalia: Top 5 Challenges and Strategies for Improvement. Public Health Chall 2025;4:e70138. [Crossref] [PubMed]
33. Chuot CC, Galukande M, Ibingira C, et al. Iodine deficiency among goiter patients in rural South Sudan. BMC Res Notes 2014;7:751. [Crossref] [PubMed]
34. Binagwaho A, Farmer PE, Nsanzimana S, et al. Rwanda 20 years on: investing in life. Lancet 2014;384:371-5. [Crossref] [PubMed]
35. Condo J, Mugeni C, Naughton B, et al. Rwanda's evolving community health worker system: a qualitative assessment of client and provider perspectives. Hum Resour Health 2014;12:71. [Crossref] [PubMed]
36. Walley J, Lawn JE, Tinker A, et al. Primary health care: making Alma-Ata a reality. Lancet 2008;372:1001-7. [Crossref] [PubMed]
37. Gelal B, Chaudhari RK, Nepal AK, et al. Iodine deficiency disorders among primary school children in eastern Nepal. Indian J Pediatr 2011;78:45-8. [Crossref] [PubMed]
38. Marmot M, Allen J, Bell R, et al. WHO European review of social determinants of health and the health divide. Lancet 2012;380:1011-29. [Crossref] [PubMed]
39. Marmot M, Friel S, Bell R, et al. Closing the gap in a generation: health equity through action on the social determinants of health. Lancet 2008;372:1661-9. [Crossref] [PubMed]