Global, regional and national burden of epilepsy secondary to four major neonatal disorders: insights from the Global Burden of Disease Study 2021

Introduction

Epilepsy is a disease of the brain characterized by an enduring predisposition to generate epileptic seizures and by the associated neurobiologic, cognitive, psychological, and social consequences (1). The neonatal period represents a critical window for lifelong neurodevelopment. From a developmental origins of health and disease (DOHaD) perspective, fetal and neonatal brain health is shaped by a continuum of factors extending from conception through early childhood (2). Conditions such as preterm birth, neonatal sepsis, hypoxic-ischemic encephalopathy, and hemolytic jaundice not only cause acute harm but also elevate the risk of long-term neurodevelopmental disorders, including secondary epilepsy—one of the most severe complications (3). In 2021, approximately 52 million people worldwide had active epilepsy, of whom 28 million were classified as having secondary epilepsy, with neonatal disorders constituting a notable etiological fraction within the Global Burden of Disease (GBD) comparative risk assessment framework (4,5). Epilepsy related to neonatal disorders exposes individuals to lifelong seizures, progressive cognitive decline, behavioral abnormalities, and reduced social adaptability, thereby increasing familial care burdens and societal healthcare costs (6). Globally, particularly in low-resource settings, uneven distribution of neonatal medical resources perpetuates a prominent long-term burden of perinatal brain injury-related epilepsy, impeding health equity (7).

Existing epidemiological evidence confirms a strong association between neonatal disorders and epilepsy. Preterm infants, due to immature brain development, are highly vulnerable to neurological insults such as hypoxia and infection. The incidence of epilepsy rises steeply with decreasing gestational age. Extremely preterm infants face a higher risk of epilepsy compared to term infants (8). A nationwide cohort study in Denmark demonstrated that early-onset neonatal sepsis increases the risk of childhood epilepsy by approximately two times, with a predisposition toward focal seizures, likely due to infection-induced meningeal inflammation and parenchymal brain damage (9). Severe hypoxic-ischemic events and trauma can lead to neonatal encephalopathy, disrupting cerebral metabolism and connectivity, thereby lowering seizures threshold (10). In untreated hemolytic jaundice, excessive serum bilirubin crosses the blood-brain barrier, causing bilirubin encephalopathy and often drug-resistant epilepsy, with permanent injury to regions such as the basal ganglia (11).

However, previous studies have primarily focused on individual neonatal disorders or been restricted to specific regions and periods, lacking a comprehensive comparative analysis of the relationships between epilepsy and four core neonatal conditions: preterm birth, neonatal sepsis, neonatal encephalopathy, and hemolytic jaundice (8,9). Although the GBD study incorporates some relevant data, detailed analyses of long-term trends, regional variations, and etiological composition of neonatal disorder-related epilepsy remain insufficient (4,12). To address these gaps, this study utilizes the GBD 2021 database to systematically analyze trends in prevalence and years lived with disability (YLDs) of epilepsy attributable to these four neonatal disorders from 1990 to 2021 across global, regional, gender, age, and Socio-Demographic Index (SDI) stratified groups. Combined with inequality assessment, health frontier analysis, and decomposition analysis, this study aims to elucidate burden trends, socioeconomic disparities, potential improvements, and key drivers. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-926/rc).

Methods

Overview

This analysis used publicly available data from the GBD 2021 study, which synthesizes 100,983 data sources to estimate the burden of 371 diseases and injuries globally (13,14). The data are publicly accessible via the Global Health Data Exchange (GHDx) repository (https://ghdx.healthdata.org/). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. We extracted estimates related to epilepsy attributable to four major neonatal disorders: neonatal preterm birth (NPB), neonatal sepsis and other neonatal infections (NS), neonatal encephalopathy due to birth asphyxia and trauma (NE), and hemolytic disease and other neonatal jaundice (HD). All data were accessed on July 25, 2025. Outcomes included prevalence, YLDs, and their corresponding age-standardized rates [age-standardized prevalence rate (ASPR) and age-standardized YLD rate (ASYR)]. All estimates were reported as counts and rates per 100,000 population, with 95% uncertainty intervals (UI). Analyses were stratified by sex, age (following GBD standard age groups from 0–5 to 95+ years), 21 GBD regions, and 5 SDI quintiles (low, low-middle, middle, high-middle, high). SDI is a composite measure of income per capita, educational attainment, and fertility rate (15).

Case definition and cause attribution in GBD 2021

In the GBD 2021 study, epilepsy is defined according to the International Classification of Diseases, Ninth and Tenth Revisions (ICD-9: code 345; ICD-10: codes G40, G41) (16). The burden estimates for epilepsy attributable to the four neonatal disorders (NPB, NS, NE, HD) are derived from the GBD causal inference and comparative risk assessment framework (4,16). This framework models the population-level etiological fraction—the proportion of the total epilepsy burden that can be attributed to these prior neonatal exposures. It synthesizes evidence from epidemiological studies on the relative risk of developing epilepsy following exposure to these neonatal conditions, applies these risks to population-level exposure estimates, and proportionally attributes the total epilepsy burden to these causes while accounting for competing risks. It is crucial to note that “epilepsy attributable to neonatal disorders” represents a population-level, model-based etiological fraction derived from this framework. It does not imply direct, clinically observed causation at the individual level for every case, nor does it preclude the possibility of comorbidity or overlapping etiological pathways between neonatal conditions (e.g., an infant may experience both preterm birth and sepsis). The GBD modeling seeks to disentangle these contributions, but residual uncertainty remains. Detailed methodology for GBD cause attribution and modeling of neonatal disorders and epilepsy is described in the core GBD 2021 publications (4,14,16).

Statistical analysis

Temporal trend analysis

Temporal trends from 1990 to 2021 were summarized using the estimated annual percentage change (EAPC). This was derived from a linear regression model with year as the independent variable (x) and the natural logarithm of the rate [ln (rate)] as the dependent variable (y), modeled as ln (rate) = α + β × year + ε. The EAPC, calculated as (e^β − 1) × 100%, where β denotes the regression coefficient reflecting the annual percentage change; statistical significance of trends was determined using 95% confidence intervals (CI), with non-zero-inclusive CIs indicating significant trends (15). The EAPC provides a summary measure of the average annual rate of change across the entire 1990–2021 period, assuming a constant relative change each year.

Cross-country inequality analysis

Cross-country inequality analysis assessed absolute and relative inequalities in disease burden via the slope index of inequality (SII) and concentration index (CII), quantifying distributional heterogeneities of neonatal disorder-related epilepsy across countries and regions to provide a comprehensive assessment of health disparities. SII, a measure of absolute inequality, captures the absolute gap between the highest and lowest SDI groups. It was computed by ranking populations by SDI and estimating the slope of the relationship between SDI and disease rates via weighted linear regression (weights = population proportion of each country/region). CII, a measure of relative inequality, reflects the concentration of burden across SDI groups. It was calculated as the covariance between the cumulative distribution of SDI-ranked populations and cumulative disease burden, divided by the mean burden (positive values indicate concentration in high-SDI groups; negative values indicate concentration in low-SDI groups, with larger absolute values denoting greater inequality) (17).

The correlation between disease burden and SDI

Spearman’s rank correlation was used as a non-parametric summary measure of the overall association and was not intended to model the potentially non-linear SDI-burden relationship. For the 21 GBD regions, Spearman’s rank correlation analysis was used to further quantify the strength of association between disease burden and regional mean SDI, with correlation coefficients (r) and corresponding P values reported (r>0 indicates positive correlation; r<0 indicates negative correlation) (18). This analysis is ecological in nature, assessing associations at the regional level.

Frontier analysis

Frontier analysis was employed to evaluate health performance across countries with varying SDI, using SDI as the independent variable and ASYR as the dependent variable to construct a “health frontier” (theoretical minimum ASYR at a given SDI level). The frontier was constructed using a non-parametric lower-envelope locally estimated scatterplot smoothing (LOESS) approach fitted to the pooled country-year data [1990–2021] to define the lowest observed ASYR achieved at each SDI level, representing a practical benchmark. To ensure analytical robustness, 1,000 bootstrap samples were generated, and the mean ASYR for each SDI value was computed. Improvement potential for each country/region was assessed by measuring the absolute distance between their 2021 ASYR and the frontier (i.e., efficiency difference: eff_diff = actual ASYR − frontier ASYR), which reflects the gap between observed and optimal values (17). The resulting efficiency gaps are descriptive metrics for benchmarking and should not be interpreted as causal measures of health system performance.

Decomposition analysis

The Das Gupta decomposition method was applied to decompose the total change in YLDs from 1990 to 2021 into three independent driving factors: population growth, aging, and epidemiological changes. The contribution ratio of each factor was calculated as (change attributed to the factor/total change) × 100%. Decomposition followed the GBD-recommended stepwise iteration approach: epidemiological changes were first calculated with fixed population structure, followed by estimation of demographic factor impacts with fixed incidence (17). These analyses served to clarify the underlying drivers of global trends in neonatal disease-related epilepsy burden.

Uncertainty estimation

In the GBD 2021 study, all calculations were conducted 500 times to generate draw-level estimates. Final estimates represent the mean across 500 draws, and 95% UIs are represented by the 2.5th and 97.5th percentile values across the draws. Uncertainty was propagated at each step of the GBD estimation process. For the present analysis, we extracted the point estimates (mean values) of prevalence, YLDs, and age-standardized rates for epilepsy attributable to the four neonatal disorders. All downstream statistical analyses were performed using these point estimates. The 500 draws were not used to propagate uncertainty into these secondary analyses.

This study adhered to the Guidelines for Accurate and Transparent Health Estimates Reporting (GATHER). All data analyses and visualizations were performed using R software (version 4.4.2), with statistical significance defined as two-tailed P<0.05.

Results

Overall burden

From 1990 to 2021, the global burden of epilepsy related to neonatal disorders increased significantly, with heterogeneous patterns across populations, regions, and etiologies. The number of prevalent cases rose from 12.3 million (95% UI: 10.48 million to 14.49 million) to 22.1 million (95% UI: 20.0 million to 24.4 million), accounting for 42.7% of all epilepsy cases and 80.4% of secondary epilepsy cases in 2021. The ASPR increased from 217.7 to 286.0 per 100,000, with an EAPC of 0.98. YLDs increased by 57.8%, from 4.3 million to 6.8 million, while the ASYR rose from 76.2 to 88.8, with an EAPC of 0.56 (Table 1). In 2021, the low-middle SDI region bore the highest burden in terms of ASPR and ASYR, while high-middle and high SDI regions had the lowest. Low SDI regions exhibited the largest increases in both metrics (Figure 1). Geographically, the Caribbean had the highest rates, whereas East Asia and High-income Asia Pacific had the lowest ASPR and ASYR, respectively. Eastern Sub-Saharan Africa showed the largest increases, contrasting with the largest decreases in Australasia (Table 1). At the national level, based on point estimates, Trinidad and Tobago had the highest ASPR and ASYR, while the Central African Republic had the lowest ASPR and Belgium the lowest ASYR (supplementary material 1,2, available at https://cdn.amegroups.cn/static/public/tp-2025-1-926-1.zip).

Figure 1 Proportion distribution of ASPR and ASYR of neonatal diseases-related epilepsy globally and across regions, 1990 vs. 2021. (A) Proportion of ASPR. (B) Proportion of ASYR. ASPR, age-standardized prevalence rate; ASYR, age-standardized years lived with disability rate; SDI, Socio-Demographic Index.

The etiological composition shifted substantially over time. NPB remained the dominant cause globally in 2021, though its proportion declined, a trend more pronounced in lower SDI regions. This decline was paralleled by significant increases in the proportions of NE and NS, with steeper increases associated with lower SDI (Figure 1). Regional exceptions were notable; NE surpassed NPB as the leading cause in East Asia, while NS marginally led in Eastern Europe, indicating distinct etiological profiles across the globe (Figure 2).

Figure 2 Temporal trends of neonatal disorders-related secondary epilepsy burden for both sexes globally and across different SDI regions, 1990–2021. (A) Age-standardized prevalence rate (ASPR). (B) Age-standardized years lived with disability rate (ASYR). SDI, Socio-Demographic Index.

NPB-related epilepsy

The burden of epilepsy attributed to NPB increased globally from 1990 to 2021, with prevalent cases rising from 8.3 million to 12.9 million. The ASPR increased with an EAPC of 0.63, while the ASYR saw a more modest increase (EAPC of 0.17). Disparities were marked, with the low-middle SDI region having the highest ASPR and ASYR in 2021, and South Asia being the highest-burden region. Nationally, Trinidad and Tobago had the highest rates, and Chad the lowest (Figure 3A, Figure S1A). A significant positive correlation was found between NPB-related ASPR and regional SDI (r=0.25, P<0.05). A significant negative correlation was found between NPB-related ASYR and SDI (r=−0.09, P<0.05) (Figure S2A,S2B). Analysis revealed gender and age disparities, with males having higher rates before age 40 years and the burden peaking in childhood (Figure S3A,S4A). Inequality trends were divergent: the absolute gap (SII) in ASPR narrowed, but relative inequality (CI) for ASYR worsened over time (Figure S5A,S6A).

Figure 3 The global ASPR of neonatal disorders-related secondary epilepsy for both sexes in 204 countries and territories in 2021 (A-D). ASPR, age-standardized prevalence rate.

NS-related epilepsy

Epilepsy related to NS also showed a considerable increase, with cases rising from 1.5 million to 3.0 million. The ASPR and ASYR increased with EAPCs of 1.08 and 0.85, respectively. The burden distribution was distinct, as the high-middle SDI region had the highest ASPR in 2021, but the low SDI region had the highest ASYR. Central Europe and Southern Sub-Saharan Africa were high-burden regions, while Romania had the highest national rates and New Zealand the lowest (Figure 3B, Figure S1B). NS-related ASPR and ASYR were both positively correlated with SDI (ASPR: r=0.43, P<0.05; ASYR: r=0.17, P<0.05) (Figure S2C,S2D). The disease burden was higher in males before age 65 years and peaked in the 5–9 years age group (Figure S3B,S4B). Unlike other etiologies, both absolute (SII) and relative (CI) inequality for NS-related epilepsy decreased significantly from 1990 to 2021 (Figure S5B,S6B).

NE-related epilepsy

NE-related epilepsy demonstrated the most rapid growth among all etiologies. Prevalent cases surged from 2.1 million to 5.4 million, with the ASPR and ASYR increasing at EAPCs of 2.08 and 1.74, respectively. The low SDI region experienced the largest increases. Central Latin America and Eastern Sub-Saharan Africa had the highest regional rates in 2021, with Mexico and Uganda having the highest national ASPR and ASYR, respectively; Portugal had the lowest rates (Figure 3C, Figure S1C). NE-related ASPR and ASYR were both positively correlated with SDI (ASPR: r=0.54, P<0.05; ASYR: r=0.30, P<0.05) (Figure S2E,S2F). The burden was consistently higher in males before age 70 years and peaked in the 5–9 years age group (Figure S3C,S4C). Inequality analysis revealed a pivotal shift, with the burden transitioning from being concentrated in high-SDI groups to being concentrated in low-SDI groups over the study period (Figure S5C,S6C).

HD-related epilepsy

Although the smallest in absolute terms, the burden of HD-related epilepsy increased, with cases rising from 0.5 million to 0.8 million. The ASPR and ASYR saw modest increases with EAPCs of 0.73 and 0.25. In 2021, the low-middle and low SDI regions had the highest ASPR and ASYR, respectively, with Central Asia being the highest-burden region. Albania and Afghanistan had the highest national ASPR and ASYR (Figure 3D, Figure S1D). HD-related ASPR and ASYR were both negatively correlated with SDI (ASPR: r=−0.31, P<0.05; ASYR: r=−0.49, P<0.05) (Figure S2G,S2H). Unlike other etiologies, females had higher ASPR and ASYR than males across all age groups, with the peak burden in the 0–5 years age group (Figure S3D,S4D). Concerningly, both absolute and relative inequalities for HD-related epilepsy worsened or persisted, indicating a growing concentration of this burden in the most vulnerable populations (Figure S5D,S6D).

Decomposition and frontier analyses (Figure 4)

Figure 4 Decomposition analysis of population-level determinant YLDs changes (aging, population growth, epidemiological changes) for neonatal disorders-related secondary epilepsy, and frontier analysis of the relationship between SDI and ASYR, globally, across SDI regions, and in 204 countries and territories, 1990–2021. (A,C,E,G) depict the decomposition analysis of YLDs for epilepsy related to four neonatal disorders (neonatal preterm birth, neonatal sepsis and other neonatal infections, neonatal encephalopathy due to birth asphyxia and trauma, and hemolytic disease and other neonatal jaundice). Black dots represent the total change contributed by all three components. A positive value for each component indicates a corresponding positive contribution to YLDs, while a negative value indicates a corresponding negative contribution to YLDs. In panels (B), (D), (F), and (H) (which depict frontier analysis results), the color change from light blue [1990] to dark blue [2021] on the left side of the figures represents the change over years. On the right side of panels (B), (D), (F), and (H), each point represents a specific country or territory in 2021; the frontier line is shown in black, and the top 15 countries and territories with the largest deviations from the frontier are marked in brown. Blue indicates low-SDI regions/countries with the smallest deviations from the frontier, while red indicates high-SDI regions/countries with the largest deviations from the frontier. The color of the dots indicates the direction of change in ASYR from 1990 to 2021: orange dots represent decreases, and green dots represent increases. ASYR, age-standardized years lived with disability rate; SDI, Socio-Demographic Index; YLD, years lived with disability.

Decomposition analysis revealed that the drivers of increasing YLDs varied by etiology (Figure 4A,4C,4E,4G). For NPB, population growth was the primary driver (131.13%). For NS, the increase was driven by a combination of population growth and epidemiological deterioration. For NE, epidemiological changes (61.69%) and population growth (45.4%) were the predominant contributors. For HD, population growth was the main driver, partially offset by epidemiological improvements (e.g., better jaundice management).

Frontier analysis identified substantial variation in efficiency gaps across countries (Figure 4B,4D,4F,4H). For NPB, high-middle SDI countries like Trinidad and Tobago had the largest efficiency gap (eff_diff =128.28), whereas low-SDI countries like Chad (eff_diff =6.91) were closer to their expected performance. For NS, high-middle SDI countries like Romania (eff_diff =49.01) had the greatest potential for improvement. For NE, low-SDI countries like Uganda (eff_diff =49.80) had the largest room for improvement. For HD, Afghanistan (eff_diff =16.08) had the largest efficiency gap, while high-SDI countries like the USA were close to the theoretical optimum.

Discussion

This study, leveraging the GBD 2021 comparative risk assessment framework, provides the first systematic quantification of the global epilepsy burden attributed to four major neonatal disorders from 1990 to 2021. Our principal findings indicate that epilepsy resulting from these conditions constitutes a substantial and growing component of global neurological disability. This aligns with the broader context of nervous system disorders being the leading cause of global disease burden (12). Specifically, our estimated 22.1 million cases of neonatal disorder-related epilepsy in 2021 contributes significantly to the overall burden, where neonatal disorders constitute a notable etiological fraction of secondary epilepsy. We observed significant increases in both prevalence and YLDs over the 32-year period, with marked heterogeneity across etiologies, regions, and socioeconomic strata. NPB remained the leading contributor globally, but its proportional share declined, while the burdens linked to NE and NS increased more rapidly, particularly in low-SDI regions. Decomposition analysis revealed that population growth was the predominant driver of rising YLDs overall, with epidemiological changes playing a more prominent role for encephalopathy- and sepsis-related epilepsy.

The observed burden trajectories and their complex association with socioeconomic development demand interpretation through a lens that extends beyond acute neonatal care. The positive correlations between SDI and ASPR for preterm birth, sepsis, and encephalopathy likely encapsulate a “survivorship-disability paradox” (19): advances in neonatal intensive care in higher-resource settings improve survival after these insults, thereby enlarging the population at risk for long-term neurological sequelae like epilepsy (6,20). For NE and NS, this paradox is particularly salient: global incidence has declined, yet epilepsy-related YLDs have risen sharply—a pattern unequivocally driven by increased survival rather than increased disease occurrence (21,22). Conversely, the consistent negative correlation for HD underscores it as a condition of persistent inequity, where gaps in access to basic preventive interventions such as prenatal screening and phototherapy remain concentrated in low-resource settings (23,24). Importantly, the distinct patterns for ASYR further suggest that the ability to mitigate disability severity, through access to antiseizure medications, rehabilitation, and chronic care, is variably distributed and not solely determined by national wealth (7,25).

The shifting etiological composition and pronounced regional disparities must be understood within the context of the DOHaD and life-course theory. The four neonatal disorders studied are not isolated acute events but rather time-sensitive biomarkers of complex disease pathways. A DOHaD perspective emphasizes that social, structural, and environmental drivers of health shape brain development across the life course (2,26). It is critical to recognize that neonatal encephalopathy itself has seen a global reduction in incidence and mortality, yet its contribution to childhood epilepsy has increased dramatically, suggesting improvements in acute survival may outpace gains in long-term neuroprotection (21). Notably, this burden is not purely intrapartum in origin; as emphasized by the American College of Obstetricians and Gynecologists, up to 70% of cases of neonatal encephalopathy have antepartum contributions, underscoring that the path to epilepsy often begins before labor (27). This underscores that the exposome affecting the woman, placenta, and fetus plays a fundamental role in etiopathogenesis. The developing fetal neural exposome is influenced by the continuity of reproductive and pregnancy-related factors affecting both the woman and her partner, and is shaped by toxic stressor interplay representing gene-environment interactions with either positive or negative outcomes (2). Therefore, a life-course approach is indispensable. Effective prevention requires continuity of interventions from preconception through the first 1,000 days (28,29). The GBD estimates, while valuable for quantifying burden, are derived from populations that largely predate modern neurocritical interventions and do not capture detailed data on many social determinants, highlighting a gap where future causal inference studies using directed acyclic graphs are needed (30).

The inequality analyses reveal a critical divergence between trends in disease prevalence and those in lifelong disability. For preterm birth-related epilepsy, while prevalence gaps between socioeconomic groups narrowed, disability gaps widened. This indicates that gains in survival, potentially facilitated by the diffusion of basic neonatal care technologies (31), have not been matched by equivalent investments in the long-term neurodevelopmental follow-up and supportive care systems needed to mitigate disability severity in lower-resource settings (7,32). The pronounced “burden shift” of encephalopathy and sepsis towards low-SDI regions over the study period starkly visualizes the consequences of unequal access to quality perinatal care and infection control (21,22). These patterns are not inevitable but are direct consequences of policy and investment priorities (33).

An intersectional approach is needed to fully understand disparities, moving beyond biological sex and SDI to consider how gender, ethnicity, geography, and access to care intersect to create compounded vulnerabilities (34). The gender disparities observed, with a higher burden in males at younger ages, mirror the higher incidence of the underlying neonatal disorders in males (35) and may be further influenced by distinct molecular mechanisms of epileptogenesis that confer differential seizure susceptibility between sexes (36); however, their interaction with social determinants warrants further study (37). Furthermore, a life-course perspective on women’s brain health is essential: early-life neurodevelopmental insults may interact with later-life factors to shape the higher observed incidence of dementias, depression, and stroke in aging women, highlighting the need for synergistic neuroprotective interventions across the lifespan (38).

The findings, while ecological, suggest actionable, albeit cautiously framed, implications for policy and clinical practice. For preterm birth, strategies must combine strengthened prenatal care with mandated, structured neurodevelopmental follow-up programs for preterm survivors to manage long-term sequelae (39). For neonatal sepsis, reducing the epilepsy burden requires a multifaceted approach: prioritizing infection prevention and control bundles in maternity units, ensuring access to clean delivery, promoting early recognition and referral, and investing in accessible diagnostics to guide appropriate antibiotic use and combat antimicrobial resistance (40,41). For neonatal encephalopathy, scaling up training in evidence-based delivery room resuscitation is paramount. Therapeutic hypothermia, while proven effective in high-income settings, requires cautious implementation in resource-limited environments; the HELIX trial demonstrated harm in South Asian settings, underscoring the need for context-specific adaptation and further research (42,43). For hemolytic disease, ensuring universal access to low-cost, high-impact interventions like bilirubin screening and phototherapy can prevent kernicterus-associated epilepsy (23,44). Crucially, health frontier analysis challenges simplistic resource determinism, demonstrating that outcomes are mediated by health system design. Strategic integration of services, continuity of care, and community-based follow-up can optimize outcomes within resource constraints (45,46).

The main strengths of this study include the use of standardized, comprehensive GBD data allowing global and regional comparisons over a long time horizon, and the application of multiple analytical lenses (trends, inequality, frontier, decomposition) to elucidate different dimensions of the burden. However, there are several limitations in this study. First, estimates rely on heterogeneous input data and modeling assumptions. Under-ascertainment and misclassification of both epilepsy and neonatal disorders—particularly in low-resource settings and in earlier years—may bias prevalence and YLD estimates and affect their temporal comparability. Second, “epilepsy attributable to neonatal disorders” reflects GBD cause-attribution modeling rather than individual-level causal inference. It represents a population-level etiological fraction derived from synthesized epidemiological evidence and does not imply direct clinical causation for every case. Therefore, associations with SDI and between-country comparisons should be interpreted as ecological and hypothesis-generating, not causal. Third, the estimates for epilepsy prevalence decades after a neonatal insult, particularly in adulthood, are highly model-dependent. They rely on extrapolations from limited long-term follow-up data, which is especially scarce for low- and middle-income countries. Furthermore, while the GBD model estimates point prevalence of active epilepsy, its approach to attributing lifelong active epilepsy to a neonatal event involves assumptions that may not fully capture the complex natural history of epilepsy, including the potential for remission as suggested by long-term follow-up studies in high-income settings. This contributes additional uncertainty to estimates of lifelong burden. Fourth, neonatal conditions frequently co-occur (e.g., prematurity, infection, encephalopathy, jaundice), and attribution may be affected by competing risks and residual overlap across causes, leading to uncertainty in etiological fractions. Fifth, trend metrics based on log-linear assumptions (EAPC) may oversimplify non-linear changes over time; future work could explore segmented trends or joinpoint analyses. Sixth, the frontier analysis uses SDI as a proxy for development and health-system capacity, but does not fully capture within-country inequalities, care quality, coverage of key interventions, or governance factors. The resulting “efficiency gaps” should be interpreted descriptively, not as causal measures of health system performance. Seventh, decomposition analyses partition changes into population growth, aging, and epidemiological change, but may not fully account for interactions among these components. Eighth, GBD burden metrics do not directly incorporate treatment access, epilepsy severity phenotypes, antiseizure medication coverage, or rehabilitation availability. Therefore, observed disability patterns may reflect both disease epidemiology and unmeasured variation in care. Ninth, our analysis is based on the GBD modeling framework, which provides synthesized global estimates. While this allows for standardized comparisons, the high proportion of secondary epilepsy attributed to neonatal disorders represents a global aggregate that may not reflect local etiological realities. In specific geographic settings hyperendemic for infections such as onchocerciasis or neurocysticercosis, these parasitic diseases are established predominant causes of epilepsy (47,48). Therefore, the relative contribution of etiologies varies dramatically by location, and our global analysis should be complemented by local epidemiological studies for targeted intervention planning. Tenth, relatedly, our study did not involve a systematic review of individual population-based epilepsy surveys, which are vital for understanding local etiological profiles, especially in regions with high burdens of infection-related epilepsy. Finally, country-level point estimates, particularly for smaller nations or those with limited input data, should be interpreted with caution due to potentially wide UIs. Comparisons between individual countries, even those with similar health systems, should consider this uncertainty.

Conclusions

In summary, this study delineates a significant and growing global burden of epilepsy attributable to major neonatal disorders, characterized by profound and evolving socioeconomic and geographical inequalities. The findings underscore that this burden is substantially modifiable by the quality, equity, and integration of healthcare systems across the reproductive, perinatal, neonatal, and child developmental continuum. Moving forward, mitigating this burden requires a dual strategy: first, targeted strengthening of prevention and acute care for neonatal disorders in high-burden, low-resource settings; and second, a systemic commitment to integrating lifelong neurodevelopmental surveillance and support into maternal and child health programs globally. Future research should employ life-course frameworks and causal inference methods to further elucidate modifiable risk pathways and evaluate the cost-effectiveness of integrated intervention packages from preconception through childhood.

Acknowledgments

We highly appreciate the efforts of GBD 2021 collaborators.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-926/rc

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-926/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2025-1-926/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. The 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. Fisher RS, Acevedo C, Arzimanoglou A, et al. ILAE official report: a practical clinical definition of epilepsy. Epilepsia 2014;55:475-82. [Crossref] [PubMed]
2. Barker DJ. The origins of the developmental origins theory. J Intern Med 2007;261:412-7. [Crossref] [PubMed]
3. Mwaniki MK, Atieno M, Lawn JE, et al. Long-term neurodevelopmental outcomes after intrauterine and neonatal insults: a systematic review. Lancet 2012;379:445-52. [Crossref] [PubMed]
4. Global, regional, and national burden of epilepsy, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Public Health 2025;10:e203-27. [Crossref] [PubMed]
5. Ketata I, Ellouz E, Mizouri R. Impact of prenatal, neonatal, and postnatal factors on epilepsy risk in children and adolescents: a systematic review and meta-analysis. Acta Epileptol 2024;6:1. [Crossref] [PubMed]
6. Oh A, Thurman DJ, Kim H. Independent role of neonatal seizures in subsequent neurological outcomes: a population-based study. Dev Med Child Neurol 2019;61:661-6. [Crossref] [PubMed]
7. Kapoor D, Garg D, Beriwal N, et al. Clinico-Etiologic Profile of Children and Adolescents with Drug-Resistant Epilepsy in a low-Resource Setting: 10 Years’ Experience. J Child Neurol 2023;38:315-20. [Crossref] [PubMed]
8. Li W, Peng A, Deng S, et al. Do premature and postterm birth increase the risk of epilepsy? An updated meta-analysis. Epilepsy Behav 2019;97:83-91. [Crossref] [PubMed]
9. Andersen M, Matthiesen NB, Murra M, et al. Early-Onset Neonatal Infection and Epilepsy in Children. JAMA Netw Open 2025;8:e2519090. [Crossref] [PubMed]
10. Nevalainen P, Metsäranta M, Toiviainen-Salo S, et al. Neonatal neuroimaging and neurophysiology predict infantile onset epilepsy after perinatal hypoxic ischemic encephalopathy. Seizure 2020;80:249-56. [Crossref] [PubMed]
11. Kaur A. Rohit, Aran KR. Unraveling the dual role of bilirubin in neurological Diseases: A Comprehensive exploration of its neuroprotective and neurotoxic effects. Brain Res 2025;1851:149472. [Crossref] [PubMed]
12. Global, regional, and national burden of disorders affecting the nervous system, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet Neurol 2024;23:344-81. [Crossref] [PubMed]
13. Murray CJL. The Global Burden of Disease Study at 30 years. Nat Med 2022;28:2019-26. [Crossref] [PubMed]
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. [Crossref] [PubMed]
15. Clegg LX, Hankey BF, Tiwari R, et al. Estimating average annual per cent change in trend analysis. Stat Med 2009;28:3670-82. [Crossref] [PubMed]
16. Global burden of 288 causes of death and life expectancy decomposition in 204 countries and territories and 811 subnational locations, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021. Lancet 2024;403:2100-32. [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. Guan SY, Zheng JX, Feng XY, et al. Global burden due to modifiable risk factors for autoimmune diseases, 1990-2021: Temporal trends and socio-demographic inequalities. Autoimmun Rev 2024;23:103674. [Crossref] [PubMed]
19. Lawn JE, Blencowe H, Oza S, et al. Every Newborn: progress, priorities, and potential beyond survival. Lancet 2014;384:189-205. [Crossref] [PubMed]
20. Pisani F, Spagnoli C, Falsaperla R, et al. Seizures in the neonate: A review of etiologies and outcomes. Seizure 2021;85:48-56. [Crossref] [PubMed]
21. Liu T, Mei H, Zheng J, et al. Global, Regional, and National Burden of Neonatal Encephalopathy due to Birth Asphyxia and Trauma, Its Attributable Risk Factors, 1990-2021: Results from the Global Burden of Disease Study 2021. Neuroepidemiology 2025; Epub ahead of print. [Crossref]
22. Li J, Xiang L, Chen X, et al. Global, regional, and national burden of neonatal sepsis and other neonatal infections, 1990-2019: findings from the Global Burden of Disease Study 2019. Eur J Pediatr 2023;182:2335-43. [Crossref] [PubMed]
23. Slusher TM, Zipursky A, Bhutani VK. A global need for affordable neonatal jaundice technologies. Semin Perinatol 2011;35:185-91. [Crossref] [PubMed]
24. Vidavalur R, Bhutani VK. Health and societal infant mortality burden of neonatal hemolytic disorders. Semin Fetal Neonatal Med 2025;30:101620. [Crossref] [PubMed]
25. Singh G, Sander JW. The global burden of epilepsy report: Implications for low- and middle-income countries. Epilepsy Behav 2020;105:106949. [Crossref] [PubMed]
26. Fleming TP, Watkins AJ, Velazquez MA, et al. Origins of lifetime health around the time of conception: causes and consequences. Lancet 2018;391:1842-52. [Crossref] [PubMed]
27. Executive summary: Neonatal encephalopathy and neurologic outcome, second edition. Report of the American College of Obstetricians and Gynecologists’ Task Force on Neonatal Encephalopathy. Obstet Gynecol 2014;123:896-901.
28. Britto PR, Lye SJ, Proulx K, et al. Nurturing care: promoting early childhood development. Lancet 2017;389:91-102. [Crossref] [PubMed]
29. Keats EC, Das JK, Salam RA, et al. Effective interventions to address maternal and child malnutrition: an update of the evidence. Lancet Child Adolesc Health 2021;5:367-84. [Crossref] [PubMed]
30. Dahabreh IJ, Bibbins-Domingo K. Causal Inference About the Effects of Interventions From Observational Studies in Medical Journals. JAMA 2024;331:1845-53. [Crossref] [PubMed]
31. Liang X, Lyu Y, Li J, et al. Global, regional, and national burden of preterm birth, 1990-2021: a systematic analysis from the global burden of disease study 2021. EClinicalMedicine 2024;76:102840. [Crossref] [PubMed]
32. Newton CR, Garcia HH. Epilepsy in poor regions of the world. Lancet 2012;380:1193-201. [Crossref] [PubMed]
33. Yan W, Qin C, Tao L, et al. Association between inequalities in human resources for health and all cause and cause specific mortality in 172 countries and territories, 1990-2019: observational study. BMJ 2023;381:e073043. [Crossref] [PubMed]
34. Scher MS, Eyre H, Donn S, et al. An interdisciplinary fetal neonatal neurology collaborative promotes integrative life-course brain health. Front Neurol 2025;16:1725289. [Crossref] [PubMed]
35. Xu L, Shen Y. Disease burden and risk factors of neonatal disorders in China from 1990 to 2021: An analysis of the global burden of disease study 2021 data. Int J Gynaecol Obstet 2025; Epub ahead of print. [Crossref]
36. Reddy DS, Thompson W, Calderara G. Molecular mechanisms of sex differences in epilepsy and seizure susceptibility in chemical, genetic and acquired epileptogenesis. Neurosci Lett 2021;750:135753. [Crossref] [PubMed]
37. Oduenyi C, Banerjee J, Adetiloye O, et al. Gender discrimination as a barrier to high-quality maternal and newborn health care in Nigeria: findings from a cross-sectional quality of care assessment. BMC Health Serv Res 2021;21:198. [Crossref] [PubMed]
38. Scher MS, Ludington S, Smith V, et al. Brain care bundles applied over each and successive generations. Semin Fetal Neonatal Med 2024;29:101558. [Crossref] [PubMed]
39. Ohuma EO, Moller AB, Bradley E, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet 2023;402:1261-71. [Crossref] [PubMed]
40. Sands K, Carvalho MJ, Portal E, et al. Characterization of antimicrobial-resistant Gram-negative bacteria that cause neonatal sepsis in seven low- and middle-income countries. Nat Microbiol 2021;6:512-23. [Crossref] [PubMed]
41. Strunk T, Molloy EJ, Mishra A, et al. Neonatal bacterial sepsis. Lancet 2024;404:277-93. [Crossref] [PubMed]
42. Arnautovic T, Sinha S, Laptook AR. Neonatal Hypoxic-Ischemic Encephalopathy and Hypothermia Treatment. Obstet Gynecol 2024;143:67-81. [Crossref] [PubMed]
43. Thayyil S, Pant S, Montaldo P, et al. Hypothermia for moderate or severe neonatal encephalopathy in low-income and middle-income countries (HELIX): a randomised controlled trial in India, Sri Lanka, and Bangladesh. Lancet Glob Health 2021;9:e1273-85. [Crossref] [PubMed]
44. Olusanya BO, Kaplan M, Hansen TWR. Neonatal hyperbilirubinaemia: a global perspective. Lancet Child Adolesc Health 2018;2:610-20. [Crossref] [PubMed]
45. Erismann S, Gami JP, Ouedraogo B, et al. Effects of a four-year health systems intervention on the use of maternal and infant health services: results from a programme evaluation in two districts of rural Chad. BMC Public Health 2021;21:2304. [Crossref] [PubMed]
46. Montgomery RJ, Scudder E, Tulloch C, et al. Constrained optimization: evaluating possible packages of community health interventions with competing resource requirements in Galmudug, Somalia. Health Policy Plan 2025;40:566-77. [Crossref] [PubMed]
47. Van Cutsem G, Siewe Fodjo JN, Hadermann A, et al. Onchocerciasis-associated epilepsy: Charting a path forward. Seizure 2026;135:105-14. [Crossref] [PubMed]
48. Sharma A, Jena D, Shabil M, et al. Trends and forecasting of cysticercosis burden in South Asia: Insights from global burden of disease data (1990-2021). Parasitol Int 2025;109:103092. [Crossref] [PubMed]