Daniel sent us this one — he's got a friend in his forties who had a single unexplained seizure, struggles with depression, and was never put on anti-seizure meds. And it shook some assumptions. He thought epilepsy was binary, you either had it or you didn't. He also assumed it clustered at both ends of the intelligence spectrum — high IQ and developmental disorders like autism. So he's asking: are either of those actually true? Is prevalence rising, falling, or flat? And what are the main types and subtypes we're actually talking about here?
The binary assumption is the one that trips up almost everyone. The cultural image is either the grand mal seizure on the floor or nothing. But the reality is this huge gray zone — people who've had one seizure and will never have another, people who have seizures only under very specific conditions, people whose epilepsy is so subtle it looks like daydreaming. Your friend's situation — one unexplained seizure plus depression — that's not an anomaly. That's a pattern the literature keeps bumping into.
The friend might technically have epilepsy, or might not, and both answers are plausible from the same set of facts.
The clinical definition is two unprovoked seizures more than twenty-four hours apart, or one unprovoked seizure with a high risk of recurrence — typically above sixty percent over the next ten years. Or an epilepsy syndrome diagnosis. So one seizure, by itself, usually doesn't meet the threshold. But if an EEG showed epileptiform activity, or if there's a structural brain abnormality, or if the seizure had focal features — starting in one limb, say — then the recurrence risk jumps enough that some neurologists would diagnose epilepsy right then.
"you either have it or you don't" is about as accurate as "you're either depressed or you're not." Which given the friend's depression, there's an irony there.
There is, and the depression connection is one of the most under-discussed parts of this whole topic. People with epilepsy have roughly a one in three lifetime prevalence of depression — about double the general population rate. But it's not just reactive, not just "having seizures is hard so you get depressed." The relationship is bidirectional. Having major depression raises your risk of developing epilepsy by two to six times depending on the study. The neurobiological pathways overlap — neurotransmitter dysregulation, HPA axis dysfunction, structural changes in the hippocampus. Depression and epilepsy share brain territory.
If someone walks into a neurologist's office with one unexplained seizure and a history of depression, the neurologist isn't seeing two separate facts. They're seeing one pattern.
And the Epilepsy Foundation cites a four to seven times higher risk of developing epilepsy in people with depression compared to the general population. That's on the order of smoking and lung cancer. And it's not well known, even among primary care physicians. So your friend's combination — that's a well-documented high-risk profile.
Like adopting a feral cat. Looks like two separate problems, turns out to be one very complicated animal.
actually a perfect way to put it.
Let me ask the obvious question. If depression raises epilepsy risk that dramatically, and depression rates have been climbing — especially in younger populations — shouldn't we see epilepsy prevalence rising too?
That's the logical prediction, and the answer is complicated. According to the CDC, about three point four million people in the US have active epilepsy — roughly one point two percent of the population. Globally, the WHO puts it at around fifty million. In terms of trends, the data doesn't show a dramatic rise in overall prevalence over the past couple of decades. Some studies show a slight increase, some show stability. The CDC's data suggests active epilepsy prevalence among adults has been relatively flat.
Depression's going up but epilepsy isn't. That seems like a contradiction.
It might mean the relationship isn't one-to-one, or that other factors are pushing in the opposite direction. Some historical causes of epilepsy — certain birth injuries, traumatic brain injuries from car accidents without seatbelts, untreated childhood meningitis — those have declined. Better obstetric care, safer cars, vaccination against H. influenzae type B and pneumococcus. So you might have depression-driven risk increasing while injury-and-infection-driven risk decreases, and the net prevalence looks flat. But nobody's done that exact decomposition analysis as far as I know.
What about the IQ piece? The prompt's assumption was that epilepsy clusters at the high end and the low end — geniuses and people with developmental disorders. That feels like one of those ideas that's too neat to be true.
It's a persistent cultural trope. The list of historical figures with epilepsy gets trotted out constantly — Dostoevsky, Van Gogh, Julius Caesar, Napoleon. The "epilepsy and genius" narrative. And on the other end, the association with intellectual disability and autism. Both have some basis in data, but the popular understanding distorts both.
Give me the actual data, not the Dostoevsky fan club.
On the developmental disorders side — yes, epilepsy is substantially more common in people with autism. Roughly twenty to thirty percent of autistic individuals also have epilepsy, compared to one to two percent in the general population. That's a ten to twenty-fold increase. And the relationship goes the other way too — children with epilepsy have higher rates of autism. There's shared genetic architecture, shared neurodevelopmental pathways. Conditions like tuberous sclerosis complex produce both. So that half of the assumption is basically correct.
The genius half?
There's no good epidemiological evidence that high IQ people have elevated epilepsy rates. What exists instead is a centuries-old cultural association reflecting a few things. One, epilepsy has historically been romanticized — the "sacred disease," seizures as divine visions. Two, a handful of extremely famous creative people had epilepsy, and confirmation bias does the rest. You remember Dostoevsky, you don't remember the millions of ordinary people with epilepsy. Three, there's been research on personality traits associated with temporal lobe epilepsy specifically — Geschwind syndrome, a cluster including hypergraphia, heightened religious or philosophical preoccupation, altered sexuality. That fed into the "epileptic personality" idea popular in mid-twentieth-century psychiatry. But none of that maps onto higher IQ. If anything, the cognitive burden of frequent seizures, especially in childhood, can depress measured IQ — a few points on average in some studies. The "epileptic genius" is mostly mythology with a few memorable anecdotes.
One half of the assumption holds up, the other half is basically folklore dressed up in a Dostoevsky quote.
That's about the size of it. And even the autism-epilepsy link, which is real, gets oversimplified. It's not that one causes the other. They share underlying mechanisms. Specific genetic syndromes — like Dravet syndrome, caused by SCN1A mutations — produce both epilepsy and features of autism. It's one underlying process manifesting in multiple ways.
Which brings us back to the friend. One seizure, depression, no diagnosis. What are the actual types we should be aware of? Because when most people hear "epilepsy," they picture the tonic-clonic on the ground. But you mentioned focal seizures that look like daydreaming.
Let me walk through the major categories, because this is where the binary assumption really falls apart. The International League Against Epilepsy divides seizures into three broad categories: focal onset, generalized onset, and unknown onset. And each of those splits into motor and non-motor, then into more specific types.
Start with focal. Those are the ones that don't look like what people expect.
Focal seizures start in one area of the brain. About sixty percent of epilepsies are focal. The key distinction is between focal aware seizures — what used to be called simple partial — where consciousness is preserved, and focal impaired awareness seizures — complex partial, in the old terminology — where consciousness is altered.
Someone having a focal aware seizure is fully conscious and aware it's happening.
Yes, and the symptoms depend entirely on where in the brain it starts. Motor cortex — jerking in one limb. Sensory cortex — tingling or numbness. Temporal lobe — which is the most common focal epilepsy location — you get an incredible range of subjective experiences. A rising sensation in the stomach. Déjà vu or jamais vu. A sudden intense emotion, often fear. Hallucinations of smell or taste. Some people describe a specific memory surfacing involuntarily. These can last from a few seconds to a couple of minutes. The person is fully present but experiencing something deeply strange.
These are the ones that can look like daydreaming from the outside.
Especially when they progress to impaired awareness. The person might stop talking mid-sentence, stare blankly, make repetitive movements with their mouth or hands — lip-smacking, picking at clothes, what are called automatisms. They're not unconscious but they're not fully there. It lasts a minute or two, then they come back, often confused or tired. No convulsions, no falling. If you didn't know what you were looking at, you'd think they were distracted or odd.
These can go completely undiagnosed for years.
There are case reports of people whose temporal lobe epilepsy wasn't diagnosed until their forties or fifties because the seizures were so subtle. They were written off as anxiety attacks, migraines, or just "spells." If your friend's unexplained seizure was actually his first recognized seizure — but he's been having focal aware episodes for years that nobody connected — that's a very common story.
What about generalized onset? That's the dramatic stuff.
Generalized seizures involve both hemispheres from the start. The motor ones everyone knows — tonic-clonic, formerly grand mal. Tonic phase is stiffening, clonic phase is rhythmic jerking. Usually lasts one to three minutes. Then there are tonic seizures — just stiffening — and clonic seizures — just jerking — and myoclonic seizures, very brief shock-like jerks. And atonic seizures, where muscle tone suddenly drops and the person collapses — drop attacks.
The non-motor generalized seizures?
Used to be called petit mal. Brief staring spells, typically five to ten seconds. The person stops, stares, then resumes like nothing happened. No confusion afterward. Most common in children, and they can happen dozens or even hundreds of times a day. A kid with untreated absence epilepsy might be written off as inattentive for years before anyone realizes.
Hundreds of times a day. That's staggering.
The record is something like several hundred in a twenty-four-hour period. And each one is just a few seconds of brain activity that looks — on EEG — like a three-per-second spike-and-wave pattern so distinctive it's basically diagnostic. One of the most striking patterns in all of clinical neurophysiology.
To put this together for the friend — he had one seizure that never got explained. It could have been a first and only tonic-clonic that came out of nowhere, it could have been the first recognized manifestation of a focal epilepsy that's been simmering subtly for years, or it could have been a provoked seizure that doesn't reflect an underlying epilepsy at all. And the depression is either a separate issue or part of the same neurobiological picture.
That's exactly the differential. And the reason he's not on anti-seizure medication is probably because the neurologist weighed the recurrence risk against the side effect burden and decided medication wasn't warranted after a single seizure. That's consistent with guidelines. The standard approach is not to treat after a first unprovoked seizure unless there are high-risk features — abnormal EEG, structural lesion, focal seizure onset, nocturnal seizure. Without those, the recurrence risk is around thirty to forty percent over the next few years. Many neurologists and patients choose to watch and wait.
It's not medical neglect. It's a calculated gamble.
And it's a gamble that's reasonable to take. Anti-seizure medications are not benign. The older ones — phenytoin, carbamazepine, valproate — have significant side effect profiles. Even the newer ones — levetiracetam, lamotrigine — can cause cognitive slowing, mood changes, fatigue. Levetiracetam in particular has a known psychiatric side effect profile — irritability, aggression, what patients call "Keppra rage." If your friend already struggles with depression, you'd think carefully before adding a medication that might worsen his mood.
That's a real clinical term?
It's the colloquial name, but it's well documented. Levetiracetam can cause behavioral side effects in ten to fifteen percent of patients, ranging from irritability to outright aggression. Pyridoxine — vitamin B6 — is sometimes co-prescribed to mitigate it. And lamotrigine, interestingly, has the opposite profile — it has mood-stabilizing properties and is sometimes used as a second-line treatment for bipolar depression. So if your friend ever did need medication, lamotrigine might be a natural choice given the depression history.
The prompt also asked about prevalence trends. You said it's roughly flat, but is that the whole story?
There are nuances. In high-income countries, prevalence is relatively stable or slightly declining. In low and middle-income countries, it's probably undercounted but may be increasing as diagnostic capacity improves — you're not seeing more epilepsy, you're seeing more diagnoses. The treatment gap is enormous. The WHO estimates that in low-income countries, up to seventy-five percent of people with epilepsy don't receive treatment. That's tens of millions of untreated people.
Seventy-five percent. That's not a gap, that's a chasm.
It's driven by lack of neurologists, cost of medications, and stigma. Epilepsy carries immense stigma in many parts of the world. It's attributed to demonic possession, to witchcraft, to moral failing. People are ostracized, they can't marry, they can't work. In some places, the traditional healer is the first and only point of contact. The epilepsy community has been fighting this for decades, and progress is slow.
What about in the US specifically? Is there a demographic pattern?
The CDC data shows that active epilepsy is more common among non-Hispanic Black adults and among people with lower household incomes and less education. That's probably reflecting disparities in access to care, higher rates of risk factors like traumatic brain injury and stroke, and possibly environmental exposures. It's a health equity issue that doesn't get enough attention.
The prevalence question is really two stories. One is the relatively flat line in wealthy countries where most cases are managed. The other is the massive undertreated burden in places where epilepsy is still a hidden epidemic.
There's a third story — the changing causes over time. In older adults, epilepsy is increasingly a consequence of stroke and neurodegenerative disease. As populations age, you see more late-onset epilepsy. In children, some historical causes — birth asphyxia, CNS infections — have declined where healthcare has improved. But new causes have emerged, like autoimmune epilepsies. There's a whole category of autoimmune encephalitis — anti-NMDA receptor encephalitis being the most famous — that can present with seizures and was completely unknown twenty years ago.
Anti-NMDA receptor encephalitis. That's the one misdiagnosed as schizophrenia in some famous cases.
Yes, Susannah Cahalan's "Brain on Fire." Young women particularly, presenting with psychosis, seizures, autonomic instability. It's caused by antibodies attacking NMDA receptors. And it's treatable — immunotherapy, tumor removal if there's an associated teratoma. But it requires recognition. Some cases that would have been called "idiopathic epilepsy" or "schizophrenia" twenty years ago are now correctly diagnosed as autoimmune disease.
Which means the prevalence numbers might be stable partly because we're shuffling cases between categories. Some things that used to be called epilepsy are now called autoimmune encephalitis. Some things that used to be called "fainting spells" are now recognized as focal seizures.
Diagnostic substitution is a real phenomenon in epidemiology. Better EEG technology and longer monitoring catch subtle seizures that would have been missed before. At the same time, you're reclassifying some historical "epilepsy" cases into other categories. The net effect on prevalence is hard to disentangle.
Let me circle back to something. The prompt's friend had one seizure, depression, no diagnosis, no meds. If you were talking to him — not as a doctor, just as someone who knows the landscape — what would you want him to understand?
First, a single seizure in adulthood is not rare. The lifetime risk of having at least one seizure is about eight to ten percent. The risk of developing epilepsy is about three percent. So most people who have a seizure do not go on to have epilepsy. That's reassuring.
Eight to ten percent lifetime risk for any seizure. That's one in twelve people. Much higher than I'd have guessed.
Most people are surprised by that number. It includes febrile seizures in childhood, but even for adults, the risk of a first unprovoked seizure is higher than people think. Second thing — the depression connection matters and shouldn't be treated as separate. If he's seeing a psychiatrist or therapist for depression, they should know about the seizure. If he ever has another neurological symptom — even something subtle like a brief episode of confusion or an odd sensory experience — he shouldn't dismiss it.
The third thing?
Third, not having a diagnosis after one seizure is actually standard care, not a failure of the system. But it means he should know what to watch for. If he ever has a second event, the diagnosis shifts immediately — two unprovoked seizures more than twenty-four hours apart is epilepsy by definition. And treatment becomes strongly recommended, because the recurrence risk after two seizures is upwards of seventy to eighty percent.
One seizure is a yellow flag. Two seizures is a red flag.
That's the clinical heuristic. And between one and two, you're in this ambiguous space where the right answer depends on the details. An EEG can help — if it shows epileptiform discharges, the recurrence risk is higher. An MRI can help — if it shows a structural lesion like hippocampal sclerosis or a cortical dysplasia, the risk is higher. But if both are normal, and the seizure was unremarkable, you might reasonably do nothing except be aware.
What about the IQ piece in the other direction? You said the genius link is mostly myth. But is there any cognitive profile actually associated with epilepsy? Not IQ specifically, but patterns of strengths and weaknesses?
There's a literature on this, and it's nuanced. People with epilepsy, as a group, score slightly lower on some cognitive measures — processing speed, attention, memory. But that's a group average that obscures enormous individual variation. Many people with epilepsy have completely normal cognition. The factors that predict cognitive difficulties are early age of onset, frequent seizures, structural brain abnormalities, and medication side effects. Control the seizures, and cognition often improves.
The temporal lobe epilepsy personality traits you mentioned — Geschwind syndrome. Is that still taken seriously?
It's controversial. Norman Geschwind described in the 1970s a profile of hypergraphia, heightened religious or philosophical concerns, altered sexual interest, and viscosity in social interactions — a kind of stickiness, difficulty changing topics. Some of his observations hold up in clinical experience — many epileptologists will tell you they've seen patients who fit the profile. But as a systematic finding, it's been hard to replicate rigorously. The personality changes might reflect the underlying brain pathology rather than the seizures themselves, or the psychosocial experience of living with epilepsy, or a selection effect — maybe people with those traits are more likely to come to medical attention.
That's a clinical term for "won't stop talking about Dostoevsky.
It's a fascinating concept. The idea is that these patients have difficulty shifting mental sets. They get stuck on a topic and can't let go. In conversation, they're overly detailed, circumstantial. It's been observed often enough that it's part of the clinical lore, but whether it's specific to temporal lobe epilepsy or just a nonspecific effect of brain dysfunction is debated.
The friend's depression, the unexplained seizure, the whole ambiguous picture — it sits right in the middle of everything we've talked about. Not clearly epilepsy, not clearly not. Not clearly connected to the depression, but the statistics suggest it might be.
That ambiguity is actually the most honest representation of where the science is. We want clean categories — epileptic or not, depressed or not, high risk or low risk. But the biology doesn't respect those boundaries. The seizure threshold is a continuum. Everyone has one — it's just a question of how high or low it is. Sleep deprivation, alcohol withdrawal, certain medications, metabolic disturbances — all can lower the seizure threshold and produce a seizure in someone who would otherwise never have one. Your friend's one seizure might mean his threshold is naturally a bit lower than average, and the depression might be both a cause and a consequence of that neurobiological substrate.
The seizure threshold as a continuum. That's a framing most people never hear.
It's the most important concept in epilepsy that nobody talks about outside of neurology. Everyone has a certain susceptibility to seizures, and seizures happen when excitatory and inhibitory forces in the brain get out of balance. Some people are born with a very high threshold — they could go days without sleep, drink heavily, and never seize. Others have a very low threshold — a slight disruption and they seize. Most people are somewhere in between. A single seizure in adulthood often means the threshold got crossed once, due to some combination of intrinsic susceptibility and transient provocation. It doesn't necessarily mean the person has a seizure disorder.
Depression lowers that threshold.
The mechanisms aren't fully worked out, but there are several plausible pathways. Depression is associated with chronic stress, which elevates cortisol, which can be neurotoxic to the hippocampus — also the most common seizure focus in temporal lobe epilepsy. Depression involves dysregulation of serotonin and norepinephrine, which modulate neuronal excitability. And depression is associated with inflammation, and neuroinflammation is increasingly recognized as a contributor to epileptogenesis — the process by which a normal brain becomes epileptic.
That's the word for "how you get from zero to epilepsy.
Yes, and it's one of the holy grails of epilepsy research. We have lots of drugs that suppress seizures — anticonvulsants. We have zero drugs that prevent epilepsy from developing in the first place. If you have a traumatic brain injury, we know your risk of epilepsy goes up, but we have nothing to give you to prevent that. If we could crack epileptogenesis, we could prevent epilepsy after head injuries, strokes, infections. That would be a bigger deal than any seizure medication.
After all these decades of neuroscience.
It's humbling. We can stop seizures once they start, but we can't stop the process that creates them. The NINDS has epileptogenesis as a major research priority. There are promising leads — targeting inflammation, targeting blood-brain barrier disruption after injury, targeting pathways like mTOR in tuberous sclerosis. But nothing in the clinic yet.
The friend who had one seizure — if his depression is somehow contributing to a slow epileptogenic process, there's literally nothing medicine can do to stop it. Just wait and see if it happens again.
That's the uncomfortable truth. And it's why the watch-and-wait approach after a first seizure isn't just about weighing medication side effects. It's also about the fact that we don't have a disease-modifying intervention to offer. If we did, the calculus would be different. You'd treat after one seizure to prevent the second. But we can't.
Let's shift to something the prompt asked about that we haven't fully unpacked. You gave us focal versus generalized, motor versus non-motor. But within those, there are specific epilepsy syndromes, right?
Yes, and this is where epilepsy classification gets genuinely complex. The ILAE recognizes dozens of specific epilepsy syndromes, each defined by a combination of seizure type, age of onset, EEG findings, imaging findings, and genetic cause when known. Let me give you a few of the most important ones, because they illustrate how different epilepsies can be from each other.
Give me the highlights.
Childhood absence epilepsy — onset usually between four and ten years old, dozens to hundreds of brief absence seizures per day, the classic three-per-second spike-and-wave on EEG. Usually resolves by adolescence, though some kids go on to develop other seizure types. Very responsive to medication — ethosuximide is the classic choice. Then there's juvenile myoclonic epilepsy — onset in adolescence, myoclonic jerks especially in the morning, plus generalized tonic-clonic seizures and sometimes absences. It's lifelong, but usually well controlled with medication, valproate being the most effective. The classic story is a teenager who starts having morning jerks, spills their coffee, thinks nothing of it, then has a tonic-clonic seizure and gets diagnosed.
Spilling your coffee turns out to be a neurological sign.
Once you know the pattern, you see it everywhere. Another big one — temporal lobe epilepsy, the most common focal epilepsy in adults. Often associated with hippocampal sclerosis — scarring and cell loss in the hippocampus visible on MRI. Can be drug-resistant in about thirty percent of cases, and those patients are often surgical candidates. Removing the sclerotic hippocampus can be curative.
Brain surgery that actually cures epilepsy.
It's one of the most effective surgeries in all of medicine. In carefully selected patients with mesial temporal lobe epilepsy and hippocampal sclerosis, about seventy to eighty percent become seizure-free after surgery. That's a cure rate that beats most cancer treatments. But the surgery is underutilized — the average patient has had epilepsy for something like twenty years before being referred for surgical evaluation.
Twenty years of seizures before anyone suggests the thing that has an eighty percent cure rate.
It's a massive failure of the referral pipeline. Part of it is that primary care doctors and even general neurologists don't think of surgery soon enough. Part of it is patient reluctance. Part of it is that the surgery is intimidating — a temporal lobectomy, removing part of the temporal lobe. There are risks, including memory decline if the surgery is on the language-dominant side. But for someone having frequent disabling seizures despite multiple medications, the risk-benefit calculation often favors surgery. And the delay costs people decades of normal life.
What about the genetic epilepsies? You mentioned Dravet syndrome.
Dravet is one of the most severe. It's a developmental and epileptic encephalopathy — meaning the epileptic activity itself contributes to cognitive and behavioral decline. Onset in infancy, typically with prolonged febrile seizures, then multiple seizure types emerge. Caused by mutations in the SCN1A gene in about eighty percent of cases. High mortality — SUDEP is a major concern. The approved treatments include cannabidiol — Epidiolex — which was a landmark approval a few years back.
CBD as an actual FDA-approved epilepsy drug. That started as parent activism and ended up changing the standard of care.
The Dravet community — parents, advocacy groups — pushed that forward. Families were moving to Colorado for access to cannabis before there was an approved pharmaceutical. Now Epidiolex is approved for Dravet, Lennox-Gastaut syndrome, and tuberous sclerosis complex. It's not a cure, but it reduces seizure frequency meaningfully in a population where nothing else works well.
That's another severe childhood epilepsy?
Yes, onset typically between three and five years old. Multiple seizure types, including tonic seizures during sleep and atonic drop attacks that cause sudden falls. Intellectual disability is common. Very drug-resistant. The EEG shows a distinctive pattern called slow spike-and-wave during wakefulness and bursts of fast activity during sleep. One of the most challenging epilepsies to manage.
On the milder end of the spectrum?
Benign epilepsy with centrotemporal spikes — also called benign rolandic epilepsy. Onset in school-age kids, focal seizures involving the face and mouth, often during sleep. Drooling, inability to speak, twitching of one side of the face. The EEG shows characteristic spikes over the centrotemporal regions. It's called benign because it almost always resolves by adolescence, and cognition is typically normal. Many kids don't even need medication.
From benign rolandic epilepsy to Dravet syndrome. That's the range we're talking about. The word "epilepsy" has to cover both.
That's exactly why the binary assumption fails. Epilepsy isn't one thing. It's dozens of different diseases that share the common feature of recurrent, unprovoked seizures. The causes range from single-gene mutations to structural brain abnormalities to autoimmune processes to completely unknown. The prognosis ranges from outgrown by puberty to lifelong and life-threatening. The treatment ranges from nothing to brain surgery. Calling it all "epilepsy" is like calling everything from a common cold to lung cancer "respiratory disease." It's technically correct and practically useless for understanding what any individual patient is dealing with.
If I'm the prompt's friend, and I'm listening to this, what I'm hearing is — one seizure plus depression is a known risk profile, you're in good company, the watch-and-wait approach is standard, and the range of what "epilepsy" means is so vast that even if you eventually get diagnosed, it tells you almost nothing until you know the specific type, cause, and syndrome.
That's a fair summary. And I'd add — if he hasn't had an EEG and an MRI, those are reasonable to pursue even after a single seizure, especially given the depression comorbidity. Not because he needs treatment now, but because knowing whether there's an underlying substrate changes the risk calculus and the vigilance level. A normal EEG and MRI are reassuring. An abnormal one changes the conversation.
If it's abnormal — if there's something there — what does that actually mean for daily life?
Driving is the most immediate practical concern. In most US states, a single seizure triggers a driving restriction, typically three to six months. If epilepsy is diagnosed, it's usually six months to a year seizure-free before you can drive again. That's often the hardest part for patients — losing independence. Beyond driving, it's about avoiding situations where a seizure would be catastrophic — swimming alone, climbing ladders, bathing instead of showering. The risk is real but has to be balanced against living a normal life. Most people with well-controlled epilepsy live without major restrictions.
The bathing thing — that's a real recommendation?
Drowning is a significant cause of death in people with epilepsy. A shower is safer than a bath because you're not submerged. It's one of those small adjustments that feels disproportionate until you understand the risk.
SUDEP — you mentioned that with Dravet. Is that a concern for everyone with epilepsy?
SUDEP is sudden unexpected death in epilepsy — death without an identified cause, not due to injury or drowning or status epilepticus. The strongest risk factor is frequent generalized tonic-clonic seizures, especially nocturnal ones. The absolute risk for most people with epilepsy is low — roughly one in a thousand per year for adults, higher for those with drug-resistant epilepsy. But it's a real risk, and it's one of the reasons seizure control matters beyond quality of life. The mechanism isn't fully understood, but it's thought to involve respiratory depression or cardiac arrhythmia during or after a seizure.
One in a thousand per year. That's low but not negligible. For someone having frequent seizures, it's much higher.
It's one of the hardest conversations epileptologists have with patients. You don't want to terrify someone whose seizures are well-controlled. But for people with drug-resistant epilepsy, especially with nocturnal tonic-clonic seizures, SUDEP risk needs to be part of the discussion. The old paternalistic approach was to not mention it, to avoid causing anxiety. The modern approach is informed consent. Patient advocacy groups were a huge force in changing those norms.
One last thread. The prompt asked about prevalence in the general population. We gave the numbers — three point four million in the US, fifty million globally, about one point two percent. But we didn't talk much about age distribution. Is it more common in kids or adults?
It's bimodal. The highest incidence — meaning new cases — is in early childhood and in older adults, over sixty-five. In kids, it's developmental and genetic causes. In older adults, it's stroke, neurodegenerative disease, brain tumors. The prevalence — the total number living with epilepsy — is highest in adults, because most childhood-onset epilepsies are lifelong or long-lasting. But the new cases per year per hundred thousand people is highest at the extremes of age.
Like a lot of neurological conditions.
It reflects the vulnerability of the developing brain and the aging brain. Different mechanisms, same endpoint — lowered seizure threshold, aberrant electrical activity. The aging brain piece is going to become more important as populations age. Late-onset epilepsy is underrecognized — it can present subtly in older adults, with confusion or memory lapses rather than convulsions. Often misattributed to dementia.
Which brings us full circle to the depression connection and the friend in his forties. He's in between the two peaks. Not a kid, not elderly. His one seizure doesn't fit neatly into either the developmental or the neurodegenerative story.
Which is why the depression stands out as potentially the most relevant factor. In a forty-something with no other risk factors, a single seizure plus depression should prompt at least a thorough workup and probably a lower threshold for considering anticonvulsant treatment if anything else emerges. It's not an emergency. But it's not nothing.
That's probably the most useful thing we've said. It's not nothing.
It's not nothing. And now: Hilbert's daily fun fact.
Hilbert: Slovene is one of the few languages with a dual grammatical number — a form for exactly two of something — and it holds the record for the smallest number of speakers of a language with a fully preserved dual, with roughly two million speakers maintaining it since at least the nineteen eighties.
The dual number in Slovene. A grammatical form for exactly two.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. You can find us at myweirdprompts dot com or wherever you get your podcasts. If you've got a weird prompt of your own, send it our way.
