Global, regional, and national burden of epilepsy, 1990–2021: a systematic analysis for the Global Burden of Disease Study 2021

1Epilepsy is one of the most common serious brain conditions of increasing burden that affects individuals of all ages across the globe,, increases risk of premature death up to three times compared with the general population, and is characterised by recurrent, unprovoked seizures due to abnormal excessive or synchronous neuronal activity in the brain. The disease imposes a substantial economic, psychosocial, physical, and mental burden for health systems, societies, and affected individuals and their families.,,

2In 2022, epilepsy was identified by the 75th World Health Assembly and WHO as one of the top priorities in prevention and control of non-communicable diseases, and a special intersectoral global action plan on epilepsy and other neurological disorders for 2022–31 was adopted. To enable evidence-based actions and awareness campaigns, and to strengthen public and private efforts to improve quality of and access to care and reduce the effect of the disease, accurate and regularly updated data on epilepsy incidence, prevalence, death, and disability by age, sex, and location are of crucial importance. From public health perspectives, it is important to provide burden estimates for idiopathic (genetic) epilepsy separately from and combined with secondary epilepsy (epilepsy syndrome due to an underlying abnormality of the brain structure or chemistry) for identifying prognosis and opportunities for prevention efforts, which are clearly different between the two types of epileptic seizures. As in the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) report on epilepsy published in 2019, causes of secondary epilepsy included, among others, stroke, neurodegenerative disorders, infections and inflammatory disorders, brain tumours, traumatic brain injuries, and congenital abnormalities. In this GBD study, these conditions were not considered risks, but rather quantified as sequelae of the underlying causes of secondary epilepsy. Therefore, the causes of secondary epilepsy are more amendable to prevention, but their treatment usually is less successful because they do not address the often severe comorbid disabilities from motor or intellectual impairments. This information might also be used for projections of the burden of epilepsy, which is also important for health-care planning and resource allocation. As emphasised by WHO, an understanding of the development of epilepsy after a brain insult or parasitic infection is crucial to the development of secondary prevention strategies.

3Evidence before this study Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2016 showed that despite a substantial decrease in age-standardised rates of idiopathic epilepsy mortality (24·5% [95% UI 10·8–31·8]) and disability-adjusted life-years (DALYs; 19·4% [9·0–27·7]) from 1990 to 2016, there was a small, although non-substantial, increase in the age-standardised prevalence (6% [–4·0 to 16·7]). The number of survivors with idiopathic epilepsy who remained disabled (as measured by DALYs) had increased: 15·3 million (11·5–19·6) and 11·3 million (8·6–14·1) in 1990 to 25·1 million (19·0–31·4) and 13·1 million (10·0–16·7) in 2016, respectively to prevalence and disability. We searched PubMed for papers from Jan 10, 2016, to Jan 28, 2022, without language restrictions, using the terms (2016/10/01(PDAT): 3000(PDAT)) AND (“epilepsy”(MeSH Terms) OR “epilepsy, partial, motor”(MeSH Terms) OR “epilepsy, benign neonatal”(MeSH Terms) OR “epilepsy, reflex”(MeSH Terms) OR “myoclonic epilepsy, juvenile”(MeSH Terms) OR “epilepsy, frontal lobe”(MeSH Terms) OR “epilepsy, complex partial”(MeSH Terms) OR “epilepsy, post-traumatic”(MeSH Terms) OR “epilepsy, temporal lobe”(MeSH Terms) OR “epilepsy, absence”(MeSH Terms) OR “epilepsy, tonic-clonic”(MeSH Terms) OR “epilepsies, myoclonic”(MeSH Terms) OR “epilepsies, partial”(MeSH Terms) OR epilep*(Title/Abstract)) AND (inciden*(Title/Abstract) OR prevalen*(Title/Abstract)) NOT (animals(MeSH) NOT humans(MeSH)). Previous studies have often been hampered by incomplete data and lack of differentiation between idiopathic and secondary epilepsy. We aimed to overcome these limitations by integrating comprehensive data from population-representative surveys and vital registrations, and offer a more accurate and detailed picture of the global epilepsy burden. Added value of this study This systematic analysis of the GBD 1990–2021 data advances previous GBD estimates on the epilepsy burden and provides the most up-to-date prevalence estimates of not only active idiopathic epilepsy but also active secondary epilepsy on global, regional, and national (204 countries) levels by age and sex for the 1990–2021 period. Implications of all the available evidence These data are important for evidence-based implementation of the WHO Resolution WHA73·10 on integrated (multisectoral) response to epilepsy and other neurological disorders for global health policy and resource allocation. By identifying regions with the highest burden of epilepsy, this study provides crucial data for targeted interventions. Policy makers can use these insights to prioritise funding for epilepsy care, improve access to antiseizure medications, and implement training programmes for health-care providers in underserved regions. Additionally, the study underscores the need for ongoing surveillance and research to track progress and adapt strategies as needed. Urgent efforts must be made by all key stakeholders and decision makers to increase awareness and education about epilepsy, eliminate stigmatisation and discrimination associated with epilepsy, better control secondary causes of epilepsy (eg, stroke, CNS zoonotic diseases, and other infectious diseases), improve access to existing treatments in economically disadvantaged countries or populations, and foster workforce development, especially in low-income countries. Further research on risk factors of idiopathic epilepsy, good-quality long-term epilepsy surveillance studies, and exploration of the possible effect of stigma and cultural differences in seeking medical attention for epilepsy is required, in addition to developing new effective and affordable treatments.

4Evidence before this study

5Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2016 showed that despite a substantial decrease in age-standardised rates of idiopathic epilepsy mortality (24·5% [95% UI 10·8–31·8]) and disability-adjusted life-years (DALYs; 19·4% [9·0–27·7]) from 1990 to 2016, there was a small, although non-substantial, increase in the age-standardised prevalence (6% [–4·0 to 16·7]). The number of survivors with idiopathic epilepsy who remained disabled (as measured by DALYs) had increased: 15·3 million (11·5–19·6) and 11·3 million (8·6–14·1) in 1990 to 25·1 million (19·0–31·4) and 13·1 million (10·0–16·7) in 2016, respectively to prevalence and disability. We searched PubMed for papers from Jan 10, 2016, to Jan 28, 2022, without language restrictions, using the terms (2016/10/01(PDAT): 3000(PDAT)) AND (“epilepsy”(MeSH Terms) OR “epilepsy, partial, motor”(MeSH Terms) OR “epilepsy, benign neonatal”(MeSH Terms) OR “epilepsy, reflex”(MeSH Terms) OR “myoclonic epilepsy, juvenile”(MeSH Terms) OR “epilepsy, frontal lobe”(MeSH Terms) OR “epilepsy, complex partial”(MeSH Terms) OR “epilepsy, post-traumatic”(MeSH Terms) OR “epilepsy, temporal lobe”(MeSH Terms) OR “epilepsy, absence”(MeSH Terms) OR “epilepsy, tonic-clonic”(MeSH Terms) OR “epilepsies, myoclonic”(MeSH Terms) OR “epilepsies, partial”(MeSH Terms) OR epilep*(Title/Abstract)) AND (inciden*(Title/Abstract) OR prevalen*(Title/Abstract)) NOT (animals(MeSH) NOT humans(MeSH)). Previous studies have often been hampered by incomplete data and lack of differentiation between idiopathic and secondary epilepsy. We aimed to overcome these limitations by integrating comprehensive data from population-representative surveys and vital registrations, and offer a more accurate and detailed picture of the global epilepsy burden.

6Added value of this study

7This systematic analysis of the GBD 1990–2021 data advances previous GBD estimates on the epilepsy burden and provides the most up-to-date prevalence estimates of not only active idiopathic epilepsy but also active secondary epilepsy on global, regional, and national (204 countries) levels by age and sex for the 1990–2021 period.

8Implications of all the available evidence

9These data are important for evidence-based implementation of the WHO Resolution WHA73·10 on integrated (multisectoral) response to epilepsy and other neurological disorders for global health policy and resource allocation. By identifying regions with the highest burden of epilepsy, this study provides crucial data for targeted interventions. Policy makers can use these insights to prioritise funding for epilepsy care, improve access to antiseizure medications, and implement training programmes for health-care providers in underserved regions. Additionally, the study underscores the need for ongoing surveillance and research to track progress and adapt strategies as needed. Urgent efforts must be made by all key stakeholders and decision makers to increase awareness and education about epilepsy, eliminate stigmatisation and discrimination associated with epilepsy, better control secondary causes of epilepsy (eg, stroke, CNS zoonotic diseases, and other infectious diseases), improve access to existing treatments in economically disadvantaged countries or populations, and foster workforce development, especially in low-income countries. Further research on risk factors of idiopathic epilepsy, good-quality long-term epilepsy surveillance studies, and exploration of the possible effect of stigma and cultural differences in seeking medical attention for epilepsy is required, in addition to developing new effective and affordable treatments.

10Previous epilepsy burden reports of GBD and GBD-based papers,,, from the 1990–2016 period, and were largely limited to the burden of idiopathic epilepsy. This GBD 2021 study aims to quantify idiopathic and secondary epilepsy prevalence, as well as incidence, death, and disability-adjusted life-years (DALYs) by age, sex, and location (globally, 21 GBD regions and seven super-regions, World Bank country income levels, Socio-demographic Index [SDI], and 204 countries) and their trends from 1990 to 2021. This manuscript was produced as part of the GBD Collaborator Network and in accordance with the GBD Protocol.

11The GBD study systematically models 371 diseases and injuries at the global, regional, and national level (in select countries; also at the subnational level), with subnational analyses in selected countries.,,

12For assessment of mortality due to epilepsy, we used underlying cause-of-death data, making corrections for misclassifications and under-reporting. Data used to estimate epilepsy mortality included vital registration, verbal autopsy, and mortality surveillance data. The International Classification of Diseases (ICD) codes were used to reassign intermediary or unspecified causes to more specific categories. The codes for epilepsy for both ICD-9 (code 345) and ICD-10 (codes G40 and G41) were used. A Cause of Death Ensemble model was used, which combines multiple models to improve mortality estimation accuracy. This is a method produced specifically for cause of death analysis in the GBD study. Further details on the methodology have been published elsewhere, and are provided in appendix 1 (p 3–32).

13The guidelines for epidemiological studies on epilepsy, its classification, and definition from the International League Against Epilepsy, formed the basis for our reference definition. An epilepsy case was defined as someone with an active, recurring condition of epileptic seizures, at least two seizures, unprovoked by any immediate cause, and who has had at least one epileptic seizure in the past 5 years regardless of antiepileptic drug treatment. We used data from additional sources from Jan 10, 2016, to Jan 28, 2022. This latest systematic review included data from Jan 10, 2016, to Jan 28, 2022, because some relevant data for the 1990–2021 period might have been published after 2021. This review yielded 24 new sources on two measures (appendix 1 p 5). The studies included were population-based, representative surveys that reported prevalence, incidence, remission rate, excess mortality rate, relative risk of mortality, standardised mortality ratio, or with-condition mortality rate. Studies that had no clearly defined sample were excluded. Studies that recorded the lifetime recall of epilepsy were crosswalked (the process of adjusting data for known biases) to the reference definition for epilepsy. Using a GBD meta-regression–Bayesian, regularised, trimmed method on the log male:female ratio of prevalence, we split observations where sex was reported for males and females combined into observations for males and females separately. Data that covered an age period of more than 25 years were split into 5-year age bands using the age patterns discerned from DisMod-MR 2·1, a Bayesian meta-regression tool, built on a subset of the epilepsy data with age bands less than 25 years. DisMod-MR 2·1 was also used to model prevalence and incidence for idiopathic and secondary epilepsy combined.

14In GBD 2021, overall epilepsy was split into idiopathic epilepsy, in which the underlying cause is unknown or genetic in nature, and secondary epilepsy, in which the underlying cause is known (eg, epilepsy due to abnormality of the brain structure or chemistry). We make explicit estimates of secondary epilepsy due to neonatal, cerebral malaria, neonatal tetanus, meningitis, cystic echinococcosis, cysticercosis, and neonatal conditions. The majority of our epidemiological data sources use the International League Against Epilepsy 1985 proposal for classification of epilepsies and epileptic syndromes, definition for idiopathic (unknown cause but generally considered to be genetically determined) epilepsy,,,,, and therefore, for this review we also used this definition. Our systematic review covering Jan 10, 2016, to Jan 28, 2022, discovered no additional unique sources of the proportion of idiopathic epilepsy due to genetic or unknown causes beyond the 89 already identified, covering 18 of 21 world regions. Not all sources use MRI or CT scans as well as electroencephalograms to diagnose secondary epilepsy. Studies that did not use advanced diagnostic methods were readjusted to the study region where all possible diagnostic methods for secondary epilepsy were used using a binary covariate for study quality based on whether the study explicitly described use of neuroimaging diagnostics across all study participants. A mixed-effects model with random effects on super-region (the 21 world regions aggregated into seven groups defined in GBD) was built using these data. The prediction of the proportion of idiopathic epilepsy obtained from this model for each year and location combination was then used in conjunction with the incidence and prevalence results from the DisMod-MR 2·1 model to calculate incidence and prevalence for idiopathic and secondary epilepsy considered separately and combined. Consistent with previous GBD 2019 report on burden of epilepsy, secondary epilepsy was quantified as long-term consequences of meningitis, tetanus, malaria, cysticercosis, cystic echinococcosis, preterm birth complications, neonatal encephalopathy, neonatal sepsis, and neonatal haemolytic disease. Secondary epilepsy from other causes, such as brain cancer, traumatic brain injury, congenital anomalies, or stroke, was not quantified explicitly but assumed to be subsumed in the severity distributions and corresponding disability weights for those conditions.

15Disability-adjusted life-years (DALYs) are the sum of years of life lost and years lived with disability (YLDs). Uncertainty is propagated through each computation step by sampling 500 draws at each step. By ordering the draws, we were able to use the 2·5 and the 97·5 percentile values to form the uncertainty intervals. Differences between two draw sets are significant if the uncertainty level of the difference does not include zero. YLDs were calculated as prevalence multiplied by the category-specific disability weight. Further details on the methodology have been published elsewhere and are in appendix 1 (pp 3–32). This report adheres to the GATHER recommendations.

16The funder of the study had no role in study design, data collection, data analysis, data interpretation, or the writing of the report.

17In 2021, there were 51·7 million (95% UI 44·9–58·9) people with epilepsy (idiopathic and secondary combined) globally (table 1), and the global age-standardised prevalence rate was 658 per 100 000 (569–748). Idiopathic epilepsy (appendix 1 pp 25–36) had an age-standardised prevalence rate of 307 per 100 000 (235–389) globally, with 24·2 million (18·5–30·7) prevalent cases (46·9% [41·1–52·1] of epilepsy from idiopathic and secondary epilepsy combined). Secondary epilepsy had a global age-standardised prevalence rate of 350 per 100 000 (322–380) and a combined number of 27·5 million (25·2–29·9) prevalent cases (53·1% [50·7–56·2] of epilepsy from idiopathic and secondary epilepsy combined).

18In 2021, the age-standardised prevalence of epilepsy from idiopathic and secondary epilepsy combined ranged from 445 per 100 000 (95% UI 312–583) in North Korea to 1374 (945–1773) in Trinidad and Tobago (table 1). By World Bank country income level, the highest age-standardised prevalence of epilepsy from idiopathic and secondary epilepsy combined (appendix 1 p 50) was observed in low-income countries (LICs; 672 per 100 000 [563–789]), followed by lower-middle-income countries (LMICs; 665 per 100 000 [570–765]); high-income countries (HICs; 662 per 100 000 [544–779]); and upper-middle-income countries (UMICs; 626 per 100 000 [539–718]). There were no significant differences in the age-standardised prevalence of idiopathic and secondary epilepsy between GBD super-regions, except for South Asia, where the prevalence of secondary epilepsy of 407 per 100 000 (350–464) was over 1·5 times greater than the prevalence of idiopathic epilepsy (260 [191–330]).

19Over the 1990–2021 period (table 1), there was a substantial increase in age-standardised prevalence of secondary and combined epilepsy (14·5% [95% UI 5·3 to 25·5] and 10·8% [1·1 to 21·3]), but a small increase in the age-standardised prevalence of idiopathic epilepsy (6·9% [–9·7 to 25·5]). Over the past three decades (table 1), the largest increase in combined epilepsy prevalence was observed in Indonesia (50·2% [18·8 to 94·5]), where the prevalence of combined epilepsy remains one of the lowest in the world since 1990, while the largest decrease was observed in Russia (12·8% [2·5 to 21·6]). From 1990 to 2021, no country showed substantial increase in the age-standardised prevalence of idiopathic epilepsy, and only Russia had a substantial decrease in the age-standardised prevalence of idiopathic epilepsy (15·2% [1·9 to 30·1]).

20There was no substantial sex difference in the prevalence of epilepsy from idiopathic and secondary epilepsy combined (males: 685 per 100 000 [95% UI 593–778]; females: 631 [548–721]) and idiopathic epilepsy (males: 322 [247–405]; females 293 [223–373]; figure 1). The prevalence of idiopathic epilepsy increased from birth (174 per 100 000 [109–276]; age 0–6 days) to age 19 years (343 [227–487]; age 15–19 years), then decreased to age 50 (233 [150–339]; age 50–54 years), and then increased with age steeply after age 55 years, with no substantial sex differences in the age pattern. Although age and sex prevalence patterns were largely similar for idiopathic and secondary epilepsy, secondary epilepsy prevalence showed consistently greater values than idiopathic epilepsy up to age 65 years, after which the age-specific prevalence of idiopathic epilepsy was substantially greater than that for secondary epilepsy (figure 1; table 1).

21Globally, in 2021, there were approximately 3·3 million (95% UI 2·4–4·1) new cases of active idiopathic epilepsy (age-standardised incidence rate 42·8 per 100 000 [31·2–53·7]), 140 000 deaths (117 000–153 000; age-standardised death rate 1·7 [1·5–1·9]), and 13·9 million DALYs (10·7–17·6; age-standardised DALYs rate 177·8 [137·7–225·9]) due to active idiopathic epilepsy (table 2; appendix 1 pp 25–50). Among global deaths and DALYs from idiopathic and secondary epilepsy combined, idiopathic epilepsy accounted for 0·21% (0·17–0·22) of deaths and 0·48% (0·38–0·59) of DALYs, ranking it as the 72nd most common cause of death and 44th most common cause of death and disability combined in the world. Over the past three decades, there was a substantial reduction in the age-standardised death and DALY rates in both males and females (table 2; figure 2; appendix 1 pp 50,52). There were no substantial sex differences in the incidence and DALY rates of idiopathic epilepsy (appendix 1 pp 50,52) in 2021, but death rates in males (2·1 [1·8–2·4]) were substantially greater than in females (1·4 [1·0–1·5]). The global age-standardised prevalence in 2021 was not substantially different between males and females: 322 per 100 000 (247–405) for males and 293 per 100 000 (223–373) for females. The prevalence of idiopathic epilepsy has not changed substantially for either males (302 per 100 000 [227–379]) or females (274 per 100 000 [205–347]) since 1990. The highest prevalence of idiopathic epilepsy in 2021 was in Ecuador, with an age-standardised prevalence of 711 per 100 000 (226–1141). North Korea had the lowest age-standardised prevalence of 178 per 100 000 (48–305).

22Geographical variations in the age-standardised YLD rates were four-fold, with the highest estimates (figure 3B) in sub-Saharan Africa (particularly in Gabon [470 per 100 000; 220–815]) and central and Latin America (particularly in Guyana [400 per 100 000; 195–654]), and lowest in western Europe (particularly in Italy [105 per 100 000; 54–187]). Lower age-standardised YLD rates were observed in the regions of high-income Asia Pacific, east Asia, Australasia, and eastern Europe. Similar to the geographical differences in age-standardised prevalence and YLDs of epilepsy from idiopathic and secondary epilepsy combined, the age-standardised prevalence of idiopathic epilepsy showed four-fold geographical variations (figure 3A; appendix 1 pp 25–36), with the highest rates in some sub-Saharan African countries (Gabon: 688 per 100 000 [146–1158]; Angola: 542 per 100 000 [143–917]; and Zambia: 547 per 100 000 [148–944]), Latin America countries (Ecuador: 711 per 100 000 [226–1141]), central Latin America (Mexico: 583 per 100 000 [401–754]), some western European countries (Germany: 539 per 100 000 [163–816]), and central Asia (Kazakhstan, Uzbekistan, and Turkmenistan, with a range of 460–479 per 100 000 [116–770]). The lowest rates were in North Korea (178 per 100 000 [48–305]), Yemen (197 per 100 000 [37–352]), Bangladesh (199 per 100 000 [59–356]), Russia (211 per 100 000 [144–282]), Somalia (212 per 100 000 [34–460]), and China (215 per 100 000 [150–279]). The lowest age-standardised prevalence of epilepsy from idiopathic and secondary epilepsy combined was observed in east Asia (especially North Korea: 445 per 100 000 [312–583]; Indonesia: 460 per 100 000 [381–549]; and China: 464 per 100 000 [393–539]), and eastern Europe (Russia 498 per 100 000 [404–592]), whereas the highest age-standardised prevalence was in some countries of the Latin America and Caribbean region (Trinidad and Tobago: 1374 per 100 000 [945–1773]; and Dominica: 1262 per 100 000 [807–1647]) and some sub-Saharan African countries (Gabon: 1360 per 100 000 [823–1833]; and Equatorial Guinea: 1284 per 100 000 [750–1754]; table 1).

23The age-standardised prevalence of idiopathic epilepsy showed four-fold geographical variations (figure 1A; appendix 1 pp 25–36), with the highest rates in some sub-Saharan African countries (Gabon: 688 per 100 000 [146–1158]; Angola: 542 [143–917]; and Zambia: 547 [148–944]), Latin America countries (Ecuador: 711 [226–1141]), central Latin America (Mexico: 583 [401–754]), some western European countries (Germany: 539 [163–816]), and central Asia (Kazakhstan, Uzbekistan, and Turkmenistan, with a range of 460–479 [116–770]). The lowest rates were in North Korea (178 [48–305]), Yemen (197 [37–352]), Bangladesh (199 [59–356]), Russia (211 [144–282]), Somalia (212 [34–460]), and China (215 [150–279]). The lowest age-standardised prevalence of epilepsy from idiopathic and secondary epilepsy combined was observed in east Asia (especially in North Korea: 445 per 100 000 [312–583]; Indonesia: 460 [381–549]; and China: 464 [393–539]), and eastern Europe (Russia: 498 [404–592]), whereas the highest age-standardised prevalence was in some countries of the Latin America and Caribbean region (Trinidad and Tobago: 1374 [945–1773]; and Dominica: 1262 [807–1647]) and some sub-Saharan African countries (Gabon: 1360 [823–1833]; and Equatorial Guinea: 1284 [750–1754]; table 1).

24Age-standardised idiopathic epilepsy incidence, prevalence, death, and DALY rates per 100 000 people in GBD regions by SDI quintiles are presented in figure 2. Globally, across all SDI quintiles in almost all GBD regions, there was a trend towards reduction of age-standardised idiopathic epilepsy death and DALY rates, but an increase in the age-standardised idiopathic epilepsy incidence rate and some increase in age-standardised death rates in the high SDI quintile, although not substantially (appendix 1 pp 25–36, 53). The bulk of the idiopathic epilepsy incidence (82·1%), prevalence (80·4%), deaths (84·7%), and DALYs (87·9%) occurred in LMICs. From 1990 to 2021 (appendix 1 pp 51, 53), age-standardised death rates substantially increased in HICs (13·9% [3·4–20·3]) but substantially reduced in UMICs (41·1% decline [34·2–47·9]), LMICs (21·5% decline [10·7–31·5]), and LICs (15·6% decline [1·4–26·7]). Over the same period, UMICs and LMICs also had a substantial reduction in age-standardised DALY rates (30·3% [15·8–42·1] and 17·3% [2·2–28·7] decrease, respectively), and similar trend patterns were observed in age-standardised DALY rates by SDI quintiles—some reduction in the rates in high and lower SDI regions (appendix 1 p 54). However, no substantial changes from 1990 to 2021 were observed in age-standardised incidence rates and prevalence in any of the World Bank country income levels (appendix 1 p 53).

25Although the global age-standardised incidence rates and prevalence of idiopathic epilepsy did not change substantially from 1990 to 2021 (12·3% [95% UI –4·8 to 32·6] and 6·9% [–9·7 to 25·5], respectively), the age-standardised death and DALY rates over that time reduced substantially (15·8% [8·8–22·8] and 14·5% [4·2–24·2] decline, respectively; appendix 1 pp 25–36). Age-standardised death rates in LICs (4·6 per 100 000 [3·9–5·6]) and LMICs (2·2 per 100 000 [1·7–2·5]), and age-standardised rates of DALYs in LICs (313·6 per 100 000 [248·0–392·3]) were greater than those in HICs and UMICs (LICs: 1·2 per 100 000 [1·1–1·2]; HICs and UMICs: 1·0 per 100 000 [0·9–1·1]; LICs: 129·2 per 100 000 [87·7–189·7]; and HICs and UMICs: 136·1 per 100 000 [97·8–182·2], respectively), especially in males (appendix 1 p 52). In 2021 (appendix 1 pp 25–36), there were large between-country variations in the age-standardised death rate of idiopathic epilepsy: the lowest rates were in Viet Nam (0·1 per 100 000 [0·0–0·3]) and San Marino (0·1 per 100 000 [0·1–0·2]) and the highest rates in Zambia (12·9 per 100 000 [9·5–17·1]) and Somalia (10·3 per 100 000 [6·7–16·9]). Similarly, there were large variations in the age-standardised DALY rates: the lowest rates were in San Marino (67·0 per 100 000 [20·3–146·5]), Russia (72·8 per 100 000 [48·2–110·6]), and Viet Nam (79·1 per 100 000 [18·1–169·2]), and the highest rates were in Zambia (746·5 per 100 000 [505·8–1031·4]) and Somalia (505·5 per 100 000 [305·5–770·1]).

26In 2021, the lowest age-standardised incidence rate of idiopathic epilepsy (appendix 1 pp 25–36) was observed in North Korea (21·7 per 100 000 [95% UI 5·9–38·7]), Bangladesh (25·7 per 100 000 [7·4–45·5]), Papua New Guinea (27·9 per 100 000 [6·6–50·8]), and China (28·2 per 100 000 [19·0–37·9]), with the highest incidence rates in Ecuador (94·9 per 100 000 [29·9–160·5]), Germany (91·8 per 100 000 [27·5–140·5]), Equatorial Guinea (84·9 per 100 000 [20·6–140·9]), and Gabon (82·8 per 100 000 [17·8–140·9]).

27Globally in 2021, across all ages, 0·7% of the population had active epilepsy. The overall global prevalence of active epilepsy from idiopathic and secondary epilepsy combined in our study (658 per 100 000 [95% UI 569–748]) is similar to the results of a recent meta-analysis of 197 epilepsy prevalence studies (638 per 100 000 [557–730]). Our estimates of the combined number of prevalent active epilepsy cases (51·7 million people), with the majority of the cases in LMICs (83·7%), were similar to those estimated by WHO in 2024. Similar to previous studies, we found no substantial difference in the age-standardised prevalence of active epilepsy between LMICs and HICs. However, from 1990 to 2021, there was a substantial increase in the age-standardised prevalence of secondary epilepsy in southeast Asia and sub-Saharan Africa, and a substantial decrease in age-standardised death and DALY rates from idiopathic epilepsy. We also found that the age-standardised prevalence of secondary epilepsy as well as age-standardised death and DALY rates of idiopathic epilepsy are higher in LICs than in HICs, but the age-standardised prevalence of idiopathic epilepsy or all-cause epilepsy does not have a substantial difference by income level. The observed reduction in death and DALY rates from idiopathic epilepsy might be related to improved access and treatment of idiopathic epilepsy.

28The substantial increase in the prevalence of secondary epilepsy in non-high-income regions was likely related to the greater exposure of the population of these countries to perinatal risk factors and higher rates of CNS zoonotic and other infections., In addition, poorer treatment (limited availability of antiseizure medications and access to specialist antiepileptic services) might also contribute to the greater age-standardised prevalence and DALYs in LICs and some LMICs compared with HICs and countries with higher SDI. It is also possible that case verification of epilepsy in high-income regions is better than in non-high-income regions, which might contribute to the observed differences. The true gap between the burden of epilepsy in HICs or high SDI countries and LICs or low SDI countries is probably even greater because of possible under-reporting of cases of epilepsy in LICs or low SDI countries, often due to stigmatisation of the disease in many countries.

29We also found three-fold to four-fold geographical variations in the prevalence of idiopathic and secondary epilepsy combined, secondary epilepsy, and idiopathic epilepsy, with slightly greater age-standardised prevalence in males compared with females. Substantial geographical variations in the prevalence of active epilepsy, with the predominance in LMICs, and slightly higher rates in males were also shown in other studies,,, whereas some studies found significant greater burden of epilepsy in males compared with females. Congruent with previous studies, we found that the prevalence of idiopathic epilepsy was relatively low early in life, increased during adolescence, and decreased after age 30 years, but unlike previous studies that showed fairly constant prevalence after age 30 years,, we found a substantial increase in the prevalence of idiopathic epilepsy after age 55 years, especially noticeable for secondary epilepsy. The large increase in the prevalence of secondary epilepsy in the elderly (after age 90 years) is likely to be related to the increase in the prevalence of stroke, brain injuries, and neurodegenerative disorders in this age group. Another likely reason is the reluctance to investigate underlying causes of epilepsy in the elderly, particularly if they have dementia, stroke, or other degenerative diseases. Many older people only have a CT scan that can only detect gross structural lesions, and cannot reliably detect temporal lobe lesions, whereas MRI is the investigation of choice in epilepsy. This investigation is often not done in older people because the need to keep still for longer and can sometimes require a general anaesthetic.

30In 2021, active idiopathic epilepsy led to almost 14 million DALYs, or 0·5% of DALYs from idiopathic and secondary epilepsy combined; the age-standardised incidence, death, and DALY rates of active idiopathic epilepsy were 42·8 per 100 000 (95% UI 31·2–53·7), 1·7 per 100 000 (1·5–1·9), and 177·8 per 100 000 (137·7–225·9), respectively. Although the age-standardised death and DALY rates of idiopathic epilepsy substantially decreased over time (appendix 1 p 52), the age-standardised incidence rate showed a non-significant trend towards increasing. Greater incidence of epilepsy (mainly epilepsy from idiopathic and secondary epilepsy combined) in LMICs was also found in a recent meta-analysis. The very low age-standardised incidence and death rates in some LMICs, such as Viet Nam and San Marino, is difficult to explain, but at least part of it might be related to the possible effect of stigma and cultural or organisational differences in epilepsy reporting. Data sources for epilepsy are also very limited in some countries. For example, there were no non-fatal epilepsy data source in San Marino, and the last cause-of-death source was from 2005. There were two non-fatal data sources in Viet Nam, both from rural areas, and only two cause-of-death sources (no vital registration system). Similarly, the causes for decreasing death and DALY rates (particularly in LMICs) are not clear but might be related to the emerging better treatment, improved identification of less severe events, with lower death rates of epilepsy and CNS infections in these countries., The increasing incidence of idiopathic epilepsy (particularly in middle-income countries) might reflect better identification of cases of idiopathic epilepsy in these countries over that time period.

31Globally, between 1990 and 2021, there was a substantial increase not only in prevalent cases of all types of epilepsy, but also an increase in the age-standardised prevalence of secondary and combined epilepsy, whereas the age-standardised prevalence of idiopathic epilepsy did not change substantially. As prevalence of many neurological conditions (eg, cerebral malaria, neonatal encephalopathy, neonatal sepsis, and nervous system cancer) have increased over the past three decades, it is probably not surprising that the prevalence of secondary epilepsy, often related to them, also increased from 1990 to 2021.

32The effect of the COVID-19 pandemic on patients with epilepsy has been substantial, including increased poor COVID-19 outcomes, mental health challenges, and difficulties in the self-management of epilepsy. For example, a survey in Brazil reported worsening of seizure control due to cancellation of appointments and challenges in access to medications. COVID-19 was also associated with decreased confidence of health care in remote management of epilepsy, worsening or aggravating of pre-existing epilepsy,,, and the risk of de novo seizures, which exceeds the risk of seizure or epilepsy after influenza. These observations highlight the necessity of further research on the long-term consequences of decreased epilepsy diagnosis and care, and its subsequent increase in epilepsy-associated mortality. Addressing these issues can help the scientific community and health-care policy makers at both global and national levels recognise gaps and insufficiencies. This awareness can lead to better preparedness for managing similar global health challenges in the future, thereby reducing the additional burden on epilepsy patients.

33The major strength of this study is that it provides the most up-to-date prevalence estimates of active idiopathic epilepsy and active secondary epilepsy, as well as the prevalence of active epilepsy from idiopathic and secondary epilepsy combined on the global, regional, and national (204 countries) levels by age and sex for the 1990–2021 period. These data are of crucial importance for health-care planning, prevention, resource allocation, and workforce development. However, there are some general and epilepsy-specific limitations of the study detailed in our previous 1990–2016 GBD epilepsy burden paper. The most important limitation of the study is the scarcity of reliable population-based epidemiological studies on various types of epilepsy in most countries of the world. In addition, the GBD study cannot provide analysis of all causes of secondary epilepsy (eg, stroke, degenerative diseases, and zoonotic diseases) and various phenotypes of idiopathic epilepsy (eg, juvenile myoclonic epilepsy, childhood absence epilepsy, juvenile absence epilepsy, and genetic generalised epilepsy) due to the scarcity of such reliable estimates in most countries. Although from previous research we know that the most common structural causes of secondary epilepsy include various neurological conditions, including traumatic brain injury, stroke, CNS zoonotic disorders, neuroinfectious diseases, neurodegenerative diseases, brain tumour, and various neural development lesions, accounting for remaining aetiologies explicitly would be desirable in future GBD rounds. We also acknowledge that many studies on epilepsy included in the GBD analysis are not nationally representative but instead focus on smaller populations within a geographic location and we had limited ability to adjust for quality of studies. Another important limitation was the scarcity of reliable data on risk factors of idiopathic epilepsy sufficient for the GBD modelling.

34In conclusion, our estimates of incidence, death, prevalence, and DALYs show diverging trends in the burden of epilepsy in the world, with the bulk of the burden residing in LMICs. Urgent efforts must be made by all key stakeholders and decision makers to increase awareness and education about epilepsy, eliminate stigmatisation and discrimination associated with epilepsy, better control secondary causes of epilepsy (stroke, CNS zoonotic diseases, and other infectious diseases), improve access to existing treatments in economically disadvantaged countries or populations, and foster workforce development, especially in LMICs. Such initiatives are important for the implementation of the WHO intersectoral global action plan on epilepsy and other neurological disorders 2022–2031 and Universal Health Coverage,, and particularly for LMICs, in which three-quarters of people with epilepsy do not get the treatment they need, and access to specialised neurological care is very limited. Further research on risk factors of idiopathic epilepsy, good-quality long-term epilepsy surveillance studies, and examination of the possible effects of stigma and cultural differences on seeking medical attention for epilepsy, as well as developing new effective and affordable treatments, need to be explored.

35To download GBD data used in these analyses, please visit the GBD 2021 Sources Tool website. To download forecasted estimates used in these analyses, please visit the GBD visualisation tools at https://collab2021.healthdata.org/gbd-compare/.

36Editorial note: The Lancet Group takes a neutral position with respect to territorial claims in published maps and institutional affiliations.

37S Afzal reports support for the present manuscript from King Edward Medical University for study material, research articles, valid data sources and authentic real time information for this manuscript. S Afzal also reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from King Edward Medical University and collaborative partners including University of Johns Hopkins, University of California, University of Massachusetts, KEMCAANA, KEMCA_UK; support for attending meetings and/or travel from King Edward Medical University; participation on a Data Safety Monitoring Board or Advisory Board with National Bioethics Committee Pakistan, King Edward Medical University Ethical Review Board, Ethical Review Board Fatima Jinnah Medical University and Sir Ganga Ram Hospital as a Member of the Technical Working Group on Infectious Diseases; leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid with the Pakistan Association of Medical Editors, the Faculty of Public Health Royal Colleges UK (FFPH) as Fellow, the Society of Prevention, Advocacy And Research, King Edward Medical University. (SPARK), and with the Pakistan Society of Infectious Diseases as a member; other financial or non-financial interest serving as Dean of Public Health and Preventive Medicine King Edward Medical University, as Chief Editor Annals of King Edward Medical University since 2014, as Director of Quality Enhancement Cell King Edward Medical University, as Principal School of Artificial Intelligence at International Level, as Fellow of Faculty of Public Health United Kingdom, as Advisory Board Member and Chair Scientific Session KEMCA-United Kingdom, as Chairperson International Scientific Conference KEMCAANA United States, as Member of the Research and Publications Higher Education Commission HEC Pakistan, as Member of the Research and Journals Committee Pakistan Medical and Dental Council, as Member of the National Bioethics Committee Pakistan, as Member of the Corona Experts Advisory Group (Punjab), as Member Dengue Experts Advisory Group (Punjab), and as Chair of the Punjab Residency Program Research Committee; all outside the submitted work. R Ancuceanu reports consulting fees from Abbvie and Merck Romania; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Abbvie, Laropharm, Reckitt, and Merck Romania; support for attending meetings and/or travel from Merck Romania; all outside the submitted work. S Bhaskar reports grants or contracts from Japan Society for the Promotion of Science (JSPS), Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) for a Grant-in-Aid for Scientific Research (KAKENHI) (Grant ID: 23KF0126) and from JSPS and the Australian Academy of Science for a JSPS International Fellowship (Grant ID: P23712); leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid with Rotary District 9675, Sydney, Australia as District (Chair, Diversity, Equity & Inclusion), Global Health & Migration Hub Community, Global Health Hub Germany, Berlin, Germany (Chair, Founding Member and Manager), PLOS One, BMC Neurology, Frontiers in Neurology, Frontiers in Stroke, Frontiers in Public Health, Journal of Aging Research, Neurology International, Diagnostics, & BMC Medical Research Methodology (Editorial Board Member), College of Reviewers, Canadian Institutes of Health Research (CIHR), Government of Canada (Member), World Headache Society, Bengaluru, India (Director of Research), Cariplo Foundation, Milan, Italy (Expert Adviser/Reviewer), National Cerebral and Cardiovascular Center, Department of Neurology, Suita, Osaka, Japan (Visiting Director), Cardiff University Biobank, Cardiff, UK (Member, Scientific Review Committee), Rotary Reconciliation Action Plan (Chair); all outside the submitted work. S Cortese reports grants or contracts from NIHR and the European Research Agency; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from the British Association of Psychopharmacology (BAP), Canadian ADHD Resource Alliance (CADDRA), Medice, and the Association for Child and Adolescent Mental Health (ACAMH); all outside the submitted work. A Hassan reports consulting fees from Novartis, Sanofi Genzyme, Biologix, Merck, Hikma Pharma, Janssen, Inspire Pharma, Future Pharma, and Elixir Pharma; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Novartis, Allergan, Merck, Biologix, Janssen, Roche, Sanofi Genzyme, Bayer, Hikma Pharma, Al Andalus, Chemipharm, Lundbeck, Inspire Pharma, Future Pharma and Habib Scientific Office, and Everpharma; support for attending meetings and/or travel from Novartis, Allergan, Merck, Biologix, Roche, Sanofi Genzyme, Bayer, Hikma Pharma, Chemipharm, Al Andalus, and Clavita Pharm; leadership or fiduciary role in other board, society, committee or advocacy group, paid or unpaid, as vice president of MENA Headache Society, board member of the Headache Chapter of the Egyptian Society of Neurology, board member of Multiple Sclerosis Chapter of the Egyptian Society of Neurology, member of the Committee of Education of the International Headache Society (IHS), membership committee of IHS, and regional committee of IHS; all outside the submitted work. I Ilic reports supports from the present manuscript from the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (project no. 175042, 2011-2023). M Ilic reports support for the present manuscript from the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (project no. 451-03-47/2023-01/200111). K Krishan reports non-financial support from the UGC Centre of Advanced Study, CAS II, awarded to the Department of Anthropology, Panjab University, Chandigarh, India, outside the submitted work. H R Marateb reports grants or contracts paid to their institution from Beatriu de Pinós post-doctoral program from the Office of the Secretary of Universities and Research from the Ministry of Business and Knowledge of the Government of Catalonia program: 2020 BP 00261, outside the submitted work. L Monasta reports support for the present manuscript from the Italian Ministry of Health (Ricerca Corrente 34/2017), payments made to the Institute for Maternal and Child Health IRCCS Burlo Garofolo. S K Panda reports support for the present manuscript via salary from Siksha ‘O’ Anusandhan (Deemed to be University). S K Panda also reports grants or contracts from DST-GOVT. OF ODISHA (Letter No. 3444/ST) outside the submitted work. R Passera reports participation on a Data Safety Monitoring Board or Advisory Board as a member of the Data Safety Monitoring Board of the clinical trial “Consolidation with ADCT-402 (loncastuximab tesirine) after immunochemotherapy: a phase II study in BTKi-treated/ineligible Relapse/Refractory Mantle Cell Lymphoma (MCL) patients” - FIL, Fondazione Italiana Linfomi, Alessandria; leadership or fiduciary roles in board, society, committee or advocacy groups as Member of the EBMT Statistical Committee, European Society for Blood and Marrow Transplantation, Paris (F) (unpaid) and as a past member of 2020-2023 (biostatistician) of the IRB/IEC Comitato Etico AO SS. Antonio e Biagio Alessandria-ASL AL-VC (paid reimbursement of expenses); all outside the submitted work. I Rautalin reports support for the present manuscript from Sigrid Juselius Foundation, Finnish Medical Foundation, Sakari Alhopuro Foundation, Finnish Foundation for Cardiovascular Research, Maud Kuistila Foundation for personal research grants with no role in the design and conduct of the study; in the collection, management, analysis, and interpretation of the data; or in the manuscript's preparation, review, or approval. U Saeed reports support for the present manuscript from the Ontario Graduate Scholarship awarded at the University of Toronto, Canada. Y L Samodra reports leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid with Benang Merah Research Center (bmrc.id) as Co-founder, outside the submitted work. V Sharma reports other financial or non-financial support from DFSS (MHA)‘s research project (DFSS28(1)2019/EMR/6) at Institute of Forensic Science & Criminology, Panjab University, Chandigarh, India, outside the submitted work. J A Singh reports consulting fees from ROMTech, Atheneum, Clearview Healthcare Partners, American College of Rheumatology, Yale, Hulio, Horizon Pharmaceuticals, DINORA, ANI/Exeltis, USA Inc., Frictionless Solutions, Schipher, Crealta/Horizon, Medisys, Fidia, PK Med, Two labs Inc., Adept Field Solutions, Clinical Care Options, Putnam Associates, Focus Forward, Navigant Consulting, Spherix, MedIQ, Jupiter Life Science, UBM LLC, Trio Health, Medscape, WebMD, and Practice Point Communications, and the National Institutes of Health; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Simply Speaking; support for attending meetings and/or travel from OMERACT as past steering committee member; participation on a Data Safety Monitoring Board or Advisory Board, unpaid, with the FDA Arthritis Advisory Committee; leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid, with OMERACT as past steering committee member, the Veterans Affairs Rheumatology Field Advisory Committee as Chair, and the UAB Cochrane Musculoskeletal Group Satellite Center on Network Meta-analysis as Editor and Director; stock or stock options in Atai Life Sciences, Kintara Therapeutics, Intelligent Biosolutions, Acumen Pharmaceutical, TPT Global Tech, Vaxart pharmaceuticals, Atyu Biopharma, Adaptimmune Therapeutics, GeoVax Labs, Pieris Pharmaceuticals, Enzolytics Inc., Seres Therapeutics, Tonix Pharmaceuticals Holding Corp., Aebona Pharmaceuticals, and Charlotte's Web Holdings, Inc. and previously-owned stock options in Amarin, Viking and Moderna Pharmaceuticals; all outside the submitted work. R Tabarés-Seisdedos reports grants or contracts from the Valencian Regional Government's Ministry of Education (PROMETEO/CIPROM/2022/58) and the Spanish Ministry of Science, Innovation and Universities (PID2021-129099OB-I00) outside the submitted work. J H V Ticoalu reports leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid with Benang Merah Research Center (bmrc.id) as Co-founder, outside the submitted work. S J Tromans reports grants or contracts paid to their institution, University of Leicester, from NHS Digital, via the Department of Health and Social Care as part of the 2023 Adult Psychiatric Morbidity Survey team collecting epidemiological data on community-based adults living in England; leadership or fiduciary roles in other board, society, committee or advocacy groups, paid or unpaid, with the Neurodevelopmental Psychiatry Special Interest Group and Psychiatry of Intellectual Disability Faculty at the Royal College of Psychiatrist as Academic Secretary, with BMC Psychiatry, Advances in Autism, Advances in Mental Health and Intellectual Disability, and Progress in Neurology and Psychiatry as Editorial Board Member, and with Psychiatry of Intellectual Disability Across Cultures (Oxford University Press) as Editor; all outside the submitted work. S Wiebe reports grants or contracts from Alberta Strategy for patient-oriented research and Epilepsy Canada; consulting fees from UCB Pharm, Eisai, Sunovion, and Liva Nova for educational grants paid to their institution; payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Torrent Pharma and Biopas; participation on Advisory Boards with Jazz Pharmaceuticals and Paladyn Labs; leadership or fiduciary roles in board, society, committee or advocacy groups, paid or unpaid with the International League Against Epilepsy as Executive Committee Member; all outside the submitted work. M Zielińska reports other financial or non-financial interests as an AstraZeneca employee outside the submitted work.