CBD and Childhood Epilepsy: What 63 Studies Reveal

When a child’s seizures become less frequent, what else changes? Can they participate more comfortably in daily activities, communicate more easily, or stay alert through a lesson? These questions help explain why counting seizures, although essential, cannot capture the entire value of an epilepsy treatment.

A new systematic review, Cannabidiol in pediatric drug-resistant epilepsies: A systematic review and meta-analysis of seizure responder rates and neurobehavioral outcome reporting, examines this challenge.¹ Accepted in October 2026 by Epilepsy Research, it brings together evidence on cannabidiol, or CBD, while asking how often researchers look beyond seizure control.

The results provide reasons for continued investigation. They also identify a practical opportunity: making future studies more useful to families by measuring how treatment affects children’s functioning alongside their seizures.

Why Pediatric CBD Research Needs Broader Measures

Drug-resistant epilepsy involves seizures that remain inadequately controlled despite appropriate medication treatment. Some childhood epilepsy syndromes also involve substantial difficulties with development, learning, and behavior. These challenges can reflect the underlying condition, epileptic activity, treatment effects, and their interactions.

CBD is used as an additional treatment, meaning it is added to an existing regimen. In the United States, prescription Epidiolex is approved for seizures associated with Lennox-Gastaut syndrome, Dravet syndrome, and tuberous sclerosis complex in patients aged one year and older.

Those specific indications matter. The Epidiolex prescribing information describes a standardized medicine with defined dosing, monitoring requirements, and potential interactions. Evidence supporting its use cannot simply be transferred to retail CBD oils or gummies.

Seizure counts remain a necessary starting point. However, measuring treatment through seizure frequency alone is like evaluating a school day only by attendance: the number matters, but it does not explain everything that happened while the child was there.

What the CBD Epilepsy Review Examined

The review identified 63 studies described across 92 reports. Secondary publications from the same research programs were linked to their original studies to reduce double counting. Most studies, 56 of 63, were observational rather than randomized trials.

For the quantitative analysis, the author pooled 37 studies involving 3,439 assessed patients. The main outcome was the proportion achieving at least a 50% reduction in seizures. This threshold identifies a “responder,” but it does not mean the person became seizure-free.

The analysis generally selected the earliest stable assessment after the initial dose-adjustment period. This approach matters because later measurements can disproportionately represent people who benefit enough, and tolerate treatment well enough, to continue.

Although the review focused on pediatric epilepsy, some included cohorts also contained adults. An analysis restricted to exclusively pediatric cohorts produced a broadly similar result, helping assess whether mixed ages substantially changed the overall estimate.

What the Seizure Findings Actually Mean

The pooled responder rate was 48.4%. Put plainly, the analysis estimated that roughly half of assessed patients experienced at least a halving of their seizures while receiving additional CBD treatment.

That is encouraging in populations whose epilepsy has resisted treatment. However, it is not a measurement of how much benefit CBD produced compared with placebo. Establishing that difference requires comparing treatment and control groups within randomized trials.

The pooled response proportion was 41.3% in the CBD arms of randomized trials and 49.8% in observational studies. Observational research captures everyday care, but changes can also reflect other treatments, natural fluctuations in seizures, and differences in who remains under observation.

The following figures come entirely from the review.

Responses also varied substantially between studies. The review’s prediction interval ranged from 29.4% to 67.8%, illustrating how much the responder proportion might differ in another comparable setting. The overall average therefore provides context, rather than a reliable forecast for an individual child.

Why Fewer Seizures Do Not Answer Every Question

The reporting gap is the review’s most important contribution. Nearly every study measured seizures, while only five reported cognitive outcomes. This does not show that CBD lacks cognitive benefits. It shows that the available research cannot confidently characterize them.

Consider a hypothetical child whose seizures fall by half. That improvement could be valuable even if learning remains unchanged. Another child might become more engaged during daily activities, but researchers would still need to determine whether this reflected fewer seizures, changes to other medicines, better sleep, or another influence.

These possibilities require separate measurements. Otherwise, an improvement in one area can be mistakenly presented as evidence of improvement everywhere.

Related coverage of CBD and stress-linked memory research raises another boundary: findings from animal experiments cannot establish developmental benefits in children with epilepsy. Different populations and experimental models answer different questions.

Sleep deserves similar precision. MyCannabis’s explanation of CBD and sleep evidence distinguishes subjective experiences from measured sleep changes. For pediatric epilepsy research, recording sleep and daytime alertness separately would help avoid treating drowsiness as proof of better rest.

How Better Studies Could Reveal More Useful Benefits

A productive next step would be to agree on a small set of outcomes that studies measure consistently. Seizure diaries could sit alongside age-appropriate assessments of cognition, behavior, and daily functioning.

The review calls for standardized longitudinal assessment, meaning repeated measurements over time. Building on that recommendation, a useful study framework would ask:

• How do seizure frequency and everyday functioning change together?
• Which changes persist, and what happens to patients who stop treatment?
• How do adverse effects and adjustments to other medicines influence the results?

These questions would help distinguish an overall treatment benefit from a single favorable number. Caregiver observations could contribute valuable information, especially where formal testing is difficult, while consistent assessment methods would make results easier to compare.

Recent research illustrates why that broader framework matters. A 2026 study of purified CBD beyond approved epilepsy indications examined records from a pediatric epilepsy center, including patients with diagnoses outside the FDA-labeled categories. Such work can identify promising directions, but retrospective observations cannot establish equivalent effectiveness across conditions.

A separate Canadian study of highly purified CBD in refractory pediatric epilepsy, published online in January 2026, examined experience within a specialist program. As clinical experience expands, using compatible outcome measures would make individual studies more informative collectively.

Where the Review’s Conclusions Need Restraint

The review itself has limitations. Its database search was confined to PubMed/MEDLINE, supplemented by citation searching. It was not prospectively registered, used a single reviewer, and did not perform a formal assessment with validated risk-of-bias tools. Relevant research may have been missed, and methodological weaknesses were not comprehensively graded.

Longer-term response estimates also require care. Studies assessing patients beyond six months reported higher responder proportions than shorter studies. However, all long-term estimates came from observational research. Patients who stopped because treatment was ineffective or difficult to tolerate could be underrepresented.

This is similar to judging a course only by the students who finished it. Their results remain informative, but understanding the full experience requires accounting for those who left.

Safety belongs in that same assessment. Prescription CBD can cause sedation, gastrointestinal problems, and liver enzyme elevations, and it can interact with other medicines. Specialist monitoring helps clinicians evaluate benefit and tolerability together. This review does not provide a basis for families to substitute consumer CBD products or change a child’s treatment independently.

The Next Question for CBD and Childhood Epilepsy

The evidence supports a constructive direction: preserve rigorous seizure measurement while broadening the definition of treatment success. A reduction in seizures can be meaningful on its own, and better research could clarify when additional benefits accompany it.

The next study should therefore ask more than how many children cross a response threshold. It should follow what changes in their daily lives, for whom, and for how long, while accounting for adverse effects and treatment discontinuation. Answering those questions would turn encouraging averages into knowledge that better reflects the children receiving care.

References:

¹ Praticò, A. D. (2026). Cannabidiol in pediatric drug-resistant epilepsies: A systematic review and meta-analysis of seizure responder rates and neurobehavioral outcome reporting. Epilepsy Research. Advance online publication. https://doi.org/10.1016/j.eplepsyres.2026.107935