First global tally of early-life epilepsy points to developmental syndromes

Epilepsy that begins in the first weeks and months of life is among the most consequential of childhood neurological conditions, bound up with developmental delay and early death. Yet how common it is, worldwide, has never been pinned down. A systematic review and meta-analysis published on 2 November in JAMA Neurology — nicknamed GENIE — assembles the first consolidated global estimate, and in doing so exposes how patchy the underlying evidence remains [s1].

What the study did

The researchers searched four major databases from inception through 18 December 2025 for population-based studies reporting the incidence or prevalence of neonatal and infantile-onset epilepsies and epilepsy syndromes [s1]. Two reviewers independently selected studies, extracted data and appraised quality using the Joanna Briggs Institute checklist for prevalence studies [s1]. To combine the figures they used a generalised linear mixed-model random-effects meta-analysis, then broke the estimates down by country, World Health Organization region, World Bank income group and diagnostic criteria [s1].

That design matters because epilepsy syndromes in infancy are defined by evolving international criteria, and studies done decades apart or in different health systems do not always count the same children. Pooling them yields a global number, but one assembled from pieces that do not fit together cleanly — a tension the authors are careful to flag.

What it found

Eighty-six population-based studies from 30 countries across four WHO regions met the criteria [s1]. The pooled cumulative incidence of infantile-onset epilepsies within the first two years of life was 236.0 cases per 100,000 live births, with a 95% confidence interval of 164.7 to 338.2 [s1]. The prediction interval — the range a new study might be expected to fall within — was far wider, from 53.0 to 1,044.3, a spread that signals just how much the true rate varies from place to place [s1].

Within that total, the developmental and epileptic encephalopathies (DEEs) stood out as the largest contributor, at an estimated 67.7 cases per 100,000 live births (95% CI, 42.1 to 108.9) [s1]. These are the severe syndromes in which the seizures themselves, alongside the underlying cause, drive a slowing or regression of development. The single most common DEE was infantile epileptic spasms syndrome (IESS), at 31.6 cases per 100,000 live births (95% CI, 27.5 to 36.2) [s1]. Among the self-limited epilepsies — the milder end of the spectrum, which often resolve — self-limited infantile epilepsy was the most frequent, at 13.0 cases per 100,000 live births (95% CI, 9.2 to 18.2) [s1].

The concentration of burden in the developmental encephalopathies is the clinically important part. These are the children for whom early recognition and treatment can matter most, and for whom a rapidly expanding set of genetic diagnoses and targeted therapies is beginning to change what is possible.

How to read it

The wide intervals are not a footnote; they are the finding's main caveat. Between-study heterogeneity was substantial, which is why the authors explicitly caution that the pooled numbers should be read as broad indicators rather than precise, generalisable rates [s1]. A prediction interval stretching from 53 to over 1,000 per 100,000 live births is too broad to anchor planning for any single country.

The deeper problem is where the data come from. Most of the included studies were conducted in high-income countries, with the WHO European Region heavily overrepresented, while the African and Eastern Mediterranean Regions were largely absent [s1]. That matters because the causes of early-life epilepsy — birth asphyxia, infection, metabolic and genetic disorders — are distributed unevenly across the world, and a global estimate built mostly on wealthy countries may understate the burden precisely where perinatal risks are highest. The headline figure is therefore best understood as a well-constructed average over a skewed sample, not a true worldwide rate.

Diagnostic methods compound that unevenness. Identifying an epilepsy syndrome in a newborn increasingly depends on electroencephalography, magnetic resonance imaging and, more and more, genetic testing — tools that are standard in the settings most of these studies came from and scarce in the ones they left out. A study that cannot reach a precise syndrome diagnosis will record fewer cases, so part of the geographic gap in these numbers reflects differences in the ability to detect the condition rather than differences in how often it occurs. The authors call for population-based work in the underrepresented regions, and for longitudinal studies that follow these children to track survival, remission and functional outcomes rather than counting them only at onset [s1].

For families, this is not a study that changes any individual decision; it carries no screening recommendation or treatment implication. Its value is structural. By putting a number — however uncertain — on how often epilepsy begins in infancy, and by showing that developmental encephalopathies dominate it, GENIE gives researchers and health systems a baseline to argue from, and a clear map of the regions where the next generation of population studies is most needed [s1].