A landmark study published in the New England Journal of Medicine found that cannabidiol (CBD), a non-psychoactive extract from the marijuana plant, slashed monthly seizure frequency nearly in half for children with one of the most devastating forms of epilepsy known to medicine.
In the double-blind, placebo-controlled trial, 120 children and young adults with Dravet syndrome saw their monthly convulsive seizures drop from an average of 12.4 down to just 5.9 after receiving CBD treatment.
Kids on placebo barely moved. Their count went from 14.9 to 14.1.
That is not a marginal difference.
That is families going from a crisis every few days to something approaching a livable routine.
And the science has only grown stronger since then.
What Is Dravet Syndrome and Why Does It Matter?
To understand what this finding means, you need to know what these families are actually living through.
Dravet syndrome is a rare, genetic epileptic encephalopathy, a brain disorder driven by seizures, that typically begins in the first year of life in an otherwise healthy infant.
It is not the kind of epilepsy that fades with age.
Seizures are frequent, prolonged, and stubbornly resistant to most medications.
As the seizures continue, the majority of children develop some degree of developmental disability.
Many lose skills they had already gained.
According to the Dravet Syndrome Foundation, patients face a 15 to 20 percent mortality rate from sudden unexpected death in epilepsy (SUDEP), prolonged seizures, drowning accidents, and infections.
For parents, every day involves a calculation most of us will never have to make.
Lennox-Gastaut syndrome (LGS) is another severe childhood epilepsy in the same category, typically appearing before age four, bringing multiple seizure types and a high rate of intellectual disability.
Together, these conditions represent some of the most medically complex and emotionally grueling diagnoses in all of pediatric medicine.
And until recently, most families had almost nowhere left to turn.
CBD Doesn’t Get Kids High. Here’s What It Actually Does.
This is the part that trips people up, so let’s be direct.
CBD is not THC.
THC is the compound in marijuana responsible for the psychoactive “high.”
CBD, by contrast, has no psychoactive properties and carries no abuse potential.
The NEJM study confirmed this directly. Cannabidiol lacks appreciable affinity or activity at the cannabinoid receptors and does not produce the psychoactivity associated with THC.
The pharmaceutical formulation used in the study is called Epidiolex, a purified oral CBD solution manufactured by GW Pharmaceuticals, which also funded the trial.
It is administered twice daily as an oral liquid, dosed at 20 milligrams per kilogram of body weight per day. That specific dose was determined after an independent safety committee reviewed data from a dose-ranging study testing three separate dose levels and identified the maximum amount that was safe and free from unacceptable side effects.
In 2018, the U.S. Food and Drug Administration approved Epidiolex specifically for the treatment of seizures associated with Dravet syndrome and Lennox-Gastaut syndrome in patients aged two and older, making it the first FDA-approved drug derived from the cannabis plant.
Researchers at NYU Grossman School of Medicine have since identified one key mechanism behind why CBD works.
CBD blocks signals carried by a molecule called lysophosphatidylinositol (LPI), found in brain neurons, which can be hijacked by disease to amplify and sustain seizure activity.
In essence, CBD appears to interrupt a feedback loop, one where seizures trigger more LPI signaling, which encourages more seizures, which raises LPI levels further.
Breaking that cycle is exactly what these children need.
Inside the Study: Who Was Enrolled and How It Worked
This was not a small or loosely designed study.
Researchers recruited patients at 23 centers across the United States and Europe, screening 177 patients before enrolling 120 who met the criteria.
To qualify, patients had to have an established Dravet syndrome diagnosis, be taking at least one anti-seizure medication, and have experienced four or more convulsive seizures in the month before the trial.
The average patient age was 9.8 years, ranging from just over 2 years old to 18.
Before any treatment began, every patient completed a four-week baseline period, during which caregivers recorded daily seizure activity using an interactive voice-response system.
This was not passive data collection. Compliance was rigorous. Adherence to the reporting system hit 97 percent in the CBD group and 98 percent in the placebo group during baseline, with similarly high rates throughout the treatment period.
Patients were then randomly assigned to receive either CBD or a matching placebo solution in addition to their existing medication regimen. All other treatments, including ketogenic diets and vagus-nerve stimulation, were required to remain unchanged for the full duration of the trial.
The treatment phase ran for 14 weeks: two weeks of dose escalation followed by 12 weeks of maintenance dosing.
After that, solutions were gradually tapered over 10 days, and nearly all of the 108 patients who completed the trial, 105 of them, chose to enter a long-term open-label extension study.
That last detail matters. When patients who have already completed a rigorous trial overwhelmingly choose to continue on a medication, it says something real about how their families experienced the results.
But Here’s What Most People Still Get Wrong
When people hear “marijuana extract for kids,” the conversation often stalls right there.
The association with recreational drug use is so strong that it can overshadow what the evidence actually shows.
Surprisingly, the truth is quite different from the headline fear.
The real issue is not that CBD is too experimental or too risky.
The real issue is that for children with drug-resistant epilepsy, the existing alternatives had already failed.
Look at the patients enrolled in the NEJM trial. Before entering the study, they had tried a median of four anti-seizure medications each, with some having attempted as many as 26. At the time of enrollment, they were still taking a median of three medications simultaneously.
Those medications included clobazam, valproate, stiripentol, levetiracetam, and topiramate. Some of the most commonly used anti-seizure drugs in pediatric neurology.
These were not naive patients who had tried one drug and given up.
They were exhausted families who had tried nearly everything.
A real-world study published in Epilepsy and Behavior involving over 550 children with severe epilepsy found that more than half of patients treated with purified CBD saw their seizures cut by at least 50 percent.
Fourteen percent became completely seizure-free.
The study followed patients for a median of 22 months across 10 medical centers and included infants as young as six months old.
Those are not preliminary numbers buried in a small lab trial.
Those are real children, real families, and real outcomes measured over nearly two years.
Another study from the UK Medical Cannabis Registry, published in Neuropediatrics, found that over 90 percent of children treated with a combination of CBD and THC experienced a 50 percent or greater reduction in seizures at six months.
Even CBD alone, without any THC, produced meaningful, measurable relief for children whose prior medications had given them almost none.
What the Data Actually Showed
The NEJM trial measured several outcomes, and across nearly every one, CBD outperformed placebo.
The primary finding was clear. Monthly convulsive seizures dropped by a median of 38.9 percent in the CBD group, compared to just 13.3 percent in the placebo group.
That gap, adjusted for baseline differences between groups, came to 22.8 percentage points in favor of CBD, a result that was statistically significant with a p-value of 0.01.
But the numbers kept building.
Among patients in the CBD group, 43 percent saw their seizure frequency drop by at least 50 percent, compared to 27 percent in the placebo group.
Three patients receiving CBD became completely seizure-free during the treatment period. No patients in the placebo group reached that threshold.
When researchers looked at total seizures across all types, not just convulsive ones, CBD again came out ahead. The adjusted reduction in total seizure frequency was 28.6 percent in the CBD group, compared to just 9 percent in the placebo group.
One of the more meaningful measures used in the study was the Caregiver Global Impression of Change, a seven-category scale that asked parents and caregivers to rate whether their child’s overall condition had improved or worsened.
This is not a cold statistical number. It is a parent describing their lived reality.
62 percent of caregivers in the CBD group said their child’s overall condition had improved, compared to just 34 percent in the placebo group.
That difference was statistically significant and, arguably, more important than any frequency table. It captured what it actually felt like to live in that household week after week.
The Numbers Behind the Hope
It helps to put the scale of this in perspective.
Dravet syndrome affects approximately 1 in 15,700 infants born in the United States.
Lennox-Gastaut syndrome is estimated to account for 3 to 4 percent of all childhood epilepsy cases.
These are classified as “rare” diseases, but the families affected would not describe the experience as rare.
For them, it is the entire world.
Of the 120 patients enrolled in the NEJM trial, 114 had documented developmental delays. Of those, 56 had delays described as severe or profound. The other 58 had mild to moderate delays.
This context is important. CBD was not being tested in a relatively healthy population with manageable seizures.
It was being tested in children for whom seizures had already taken a measurable neurological toll, and it still worked.
A separate year-long open-label study at NYU Langone tracking 214 patients between ages 1 and 30 found a median 36.5 percent reduction in monthly motor seizures across 12 weeks of treatment.
For children averaging 30 seizures a month at the start, that translated to cutting seizure frequency roughly in half.
What About the Side Effects?
This is where the conversation deserves honesty rather than either alarm or dismissal.
Side effects in the CBD group were more common than in the placebo group, and the trial reported them transparently.
93 percent of patients in the CBD group experienced at least one adverse event, compared to 75 percent in the placebo group.
The most frequently reported issues, occurring in more than 10 percent of CBD patients, included vomiting, fatigue, fever, upper respiratory infection, decreased appetite, lethargy, somnolence, and diarrhea.
Somnolence, or excessive sleepiness, was the most common single adverse event, occurring in 36 percent of the CBD group versus 10 percent in the placebo group.
Notably, 18 of the 22 patients who experienced somnolence in the CBD group were also taking clobazam, another anti-seizure medication, suggesting a potential drug interaction rather than a standalone CBD effect.
Elevated liver enzyme levels were reported in 12 patients in the CBD group, all of whom were also taking valproate. Most of these resolved while the patients continued taking CBD, suggesting a transient metabolic effect rather than lasting liver damage. This is still a signal that warrants monitoring, particularly in patients on valproate.
Eight patients withdrew from the trial due to adverse events in the CBD group, compared to one in the placebo group.
These are real considerations. Families and neurologists need to weigh them carefully.
But for parents of children who are having 12, 20, or even 30 seizures a month and who have already tried multiple failed medications, a 40 to 50 percent reduction is not a consolation prize.
It is a life-changing shift.
Research published in ScienceDirect measuring quality of life in Dravet and LGS patients found that seizure-free days had the greatest impact on quality of life more than any other factor measured.
Every seizure prevented is not just a medical data point.
It is a morning where a child wakes up safely.
It is a parent who slept through the night.
What This Means for Families Right Now
CBD treatment for epilepsy is not a fringe idea anymore.
It is FDA-approved, peer-reviewed, and increasingly standard in pediatric neurology clinics across the United States.
A May 2025 study found that nearly half of pediatric patients prescribed Epidiolex achieved a significant reduction in seizures, with 49 percent experiencing greater than a 25 percent drop.
The NEJM trial also confirmed that CBD did not negatively affect sleep quality. There was no significant difference between groups on the sleep disruption score or the Epworth Sleepiness Scale, which is important for children whose development and daily functioning depend heavily on consistent sleep.
There were also no instances of suicidal ideation recorded among the 77 patients who completed the Columbia Suicide Severity Rating Scale, a standard safety screening tool used in psychiatric and neurological drug trials.
These reassurances matter. Parents worried about adding another drug to an already complicated regimen deserve the full picture.
And the full picture, while not without caveats, is genuinely encouraging.
The Research Keeps Building
Scientists are not done.
Researchers at NYU are now investigating whether CBD efficacy can be maintained at lower doses, which could help reduce the risk of side effects.
Longer-term studies are underway for Dravet syndrome, Lennox-Gastaut syndrome, and other forms of treatment-resistant epilepsy.
There is also growing interest in whether early treatment initiation matters.
The 2025 real-world study specifically noted that including infants under one year old offered early evidence that starting CBD treatment sooner may help reduce the neurological toll of seizures during a critical developmental window.
The brain’s early years are not recoverable.
Every seizure that fires unchecked during that window carries consequences that ripple outward for years.
That is the quiet urgency behind all of this research.
A Closing Thought
For decades, the words “marijuana” and “medicine for children” could barely appear in the same sentence without triggering an argument.
The science has quietly, methodically moved past that debate.
What CBD research in pediatric epilepsy has revealed is something worth sitting with.
A compound once dismissed as a drug of abuse has now been clinically tested across 23 medical centers on two continents, scrutinized in one of the world’s most rigorous scientific journals, and approved by the FDA.
And it works.
Not for every child. Not without side effects. Not as a cure.
But for families who watched their children have seizure after seizure while medication after medication failed, it offers something that clinical trials rarely measure but everyone in those households can feel.
A little more ordinary time.
The question worth asking now is not whether CBD works.
The question is how quickly researchers can make it work better, for more children, with fewer compromises.
