Pediatric Epilepsy
The 20 absence seizures your patient had during math class won't appear in a seizure diary.
Pediatric epilepsy generates a constant stream of clinical signals — seizures at school, medication side effects that look like behavior problems, developmental changes that might be the disease or the treatment — yet caregivers are asked to reconstruct weeks of events from memory at quarterly visits. In rural settings, the nearest pediatric neurologist is hours away, and the daily data that would transform a telehealth call goes uncaptured. Forma captures this signal.
Forecast
Based on prior Topics logging
Topics
How it works
15 seconds of speech. 25 data points.
Forma turns each caregiver into a continuous data source. By voice or text, parents log seizure events with type and context, medication doses and side effects, school attendance and in-school seizures, developmental changes, sleep patterns, and healthcare access barriers in under a minute. No forms, no recall bias — and voice is essential because these parents are logging at 2am after a seizure, or in the car after a school nurse call. The result is a structured, longitudinal dataset that pairs seizure frequency with medication timing, school impact with seizure type, and developmental trajectories with treatment changes — visible to the neurologist before the next telehealth call.
“”
Forecast
Based on prior Topics logging
Topics
Structured, coded, longitudinal
- Timing
- Seizure time:6:00 AMSeizure duration:45 secRecovery time:20 minMissed dose time:Previous evening
- Location
- Injury location:Tongue
- Severity
- Injury:Tongue bitePost-ictal symptom:Confusion
- Actions
- Adherence:Missed doseMedication:Keppra
- Environmental
- Barometric pressure:29.82 inHg (falling)Moon phase:Waxing gibbous (87%)
From data to insights
What the data reveals
Daily voice logs compound into actionable insights — for patients managing their condition and for researchers running studies.
For Patients
Are these staring spells actually seizures?
Eli had 47 staring episodes this week — average 8 seconds each, most during screen time or transitions. His teacher reported 12 during class. Pattern matches absence seizures, not inattention. Since adding Ethosuximide, episodes dropped from 20/day to 6/day.
Daily absences
Is the Keppra causing his behavior problems?
Eli's irritability scores spiked 4 days after the Keppra dose increase and stayed elevated for 9 days. Meltdowns averaged 3.2/day during that period vs. 0.4/day before. His mood improved within 48 hours of the dose reduction. This is a known Keppra pattern.
Irritability trend
How is his school performance doing?
Eli attended 14 of 20 school days this month. On seizure-free days, teachers report full assignment completion. On days with 5+ absences, he finishes less than half. Reading level is recovering — up one level since switching off Topiramate 6 weeks ago.
School days attended
Summarize this month for the neurologist.
Tonic-clonics: 2 (days 1 and 13, both nocturnal, one required ER). Absence seizures: down from 20/day to 6/day on Ethosuximide. Keppra irritability prompted dose reduction. EEG day 18 showed fewer spike-waves. Developmental win: spontaneous two-word sentence. Total school days missed: 6.
For Researchers
How reliable is caregiver seizure reporting?
Voice-logged seizure counts correlate r=0.91 with EEG-confirmed events in our 42-patient cohort. Absence seizures are undercounted by 34% in paper diaries vs. voice logs. Caregivers capture 2.3x more seizure detail when logging by voice within 30 minutes of an event.
Reporting accuracy
Trend
What predicts medication side effects vs. disease?
Behavioral changes within 3–7 days of dose changes are medication-related in 82% of cases (validated against clinician adjudication). Keppra rage onset: median 4 days post-increase. Topiramate cognitive effects: median 2 weeks. Daily logging disambiguates what quarterly visits cannot.
Side effect timing
Trend
What is the real rural access burden?
Rural families average 4.2 hours round-trip per in-person neurology visit. 38% report local pharmacies that don't stock rescue medications. Telehealth visits with daily voice data last 22 minutes vs. 14 minutes without — and result in 2.1x more medication adjustments.
Travel burden (hrs)
Trend
How does seizure frequency correlate with school outcomes?
Days with >5 absence seizures predict 58% reduction in assignment completion (p<0.001). Tonic-clonic events cause 2.3 average missed school days. Children with daily voice monitoring have 41% fewer unrecognized school seizures — teachers report when prompted by data.
Seizures vs. school
Trend
Eli had a big tonic-clonic at 7am — lasted 3 minutes, fell off chair, turned blue. Gave Diastat. Kept home from school.
20+ absence seizures while watching TV. Zoning out, unresponsive to name, ~10 seconds each. Post-ictal from yesterday's big one.
Back at school. Nurse called at 11 — staring spell in class witnessed by aide. Sent home early. Teacher worried about his focus.
Drove 2.5 hours to children's hospital for neurology follow-up. Dr. Chen increased Depakote to 750mg BID. Added Ethosuximide for the absences.
Really irritable all day — snapping at his sister, crying over nothing. Maya suspects it's the Keppra. Terrible night — up three times.
Full day at school — no seizures! Teacher said he did great, finished all his work. First good school day in two weeks.
Telehealth with neurologist. Discussed the irritability — may switch from Keppra. EEG scheduled for next week. Maya stressed but managing.
Really bad seizure at 2am — asleep, monitor alarm, turned blue, incontinent, 5 minutes. Had to call 911. ER admission overnight.
Home from ER. Exhausted — Maya cancelled two work calls, hadn't slept. Local pharmacy doesn't stock Diastat, mail ordering. Stress overwhelming.
EEG showed improvement — fewer spike-wave discharges. Dr. Chen cautiously optimistic. Eli said a two-word sentence spontaneously — huge milestone.
Seizure-free day! He was interactive and happy — playing with cousins for hours. Really felt like himself. First genuinely normal day in months.
Key insight
Over three weeks, Maya's daily logs reveal the patterns a quarterly visit compresses into 'things have been rough': a cluster of 20+ absence seizures on day 2 that the school initially dismissed as inattention, a Keppra-related irritability spike on day 7 that triggered a medication discussion, and a 5-minute nocturnal seizure on day 13 that required 911 from a town 3 hours from the nearest children's hospital. The treatment response — Ethosuximide added on day 5, EEG improvement by day 18, and a spontaneous speech milestone the same day — is visible in the daily data weeks before the next scheduled visit. The rural access burden is equally clear: a 5-hour round trip for one appointment, a local pharmacy that doesn't stock Diastat, and a caregiver who missed work to be present for all of it.
See Pediatric Epilepsy configured for your study
See Forma configured for Pediatric Epilepsy