EpilepsyPediatric Seizure Disorders

Rural Pediatric Epilepsy

Your patients' seizure diaries end when the pen hits the floor at 3 AM.

Children with epilepsy generate their most clinically valuable data between specialist visits — nocturnal seizures, medication side effects, cognitive shifts, caregiver exhaustion. For rural families separated from neurologists by hours of highway, these signals vanish into the gap. Forma captures them in the parent's own voice, structured for research.

How it works

15 seconds of speech. 25 data points.

Forma turns each family into a continuous data stream. A parent speaks for 30 seconds — 'Liam had a big tonic-clonic at 6 this morning, lasted 4 minutes, both sides shaking, I swiped the magnet twice' — and Forma extracts seizure type, duration, laterality, VNS intervention, and timestamps into structured fields across 13 condition-specific topics. The dataset integrates seizure events, medication tolerance, sleep quality, cognitive function, school impact, caregiver wellbeing, and healthcare access into a longitudinal record that bridges the months between specialist visits.

Patient speaking

😊
Today

Forecast

Based on prior Topics logging

😴
Fatigue
MTWTFSS
🤗
Mood

Topics

💧
Seizure Event
🔥
Medication
💊
Post-ictal
💧
Sleep Quality
🔥
Triggers
0
data points extracted
Research-Ready Dataset

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

Forma Insights

Are Liam's seizures getting better or worse since the Epidiolex increase?

In the 10 days before the dose increase, Liam averaged 4.2 seizures per day. In the 10 days after (excluding the illness cluster), his average dropped to 1.4 — a 67% reduction. His tonic-clonics specifically decreased from 3 per week to 1.

Daily seizure count

Does his sleep affect how many seizures he has the next day?

On nights when Liam slept fewer than 6 hours, he had an average of 3.8 seizures the following day compared to 1.6 on nights with 7+ hours. Poor sleep preceded 78% of his tonic-clonic events.

Seizures by prior night sleep

How much school has Liam missed because of epilepsy?

Over the past 4 weeks, Liam missed 6 full school days and had 3 partial days. Seizure-related absences accounted for 83% of missed time. His longest streak of full attendance was 5 consecutive days during Week 2.

School days attended per week

Can you summarize Liam's last few weeks for his neurologist?

Over the past 25 days, Liam's seizure frequency decreased from an average of 4.2 to 1.4 daily events following the Epidiolex increase to 5 mL BID on Day 3. A febrile illness on Day 14 triggered a 3-day cluster (9 tonic-clonics total) requiring Diastat once. Post-illness, seizures returned to the improved baseline. Cognitively, his teacher reports improved focus and he achieved a motor milestone (writing his full name). VNS hoarseness resolved after settings were lowered. Keto compliance is strong at 3:1 ratio. Key concern: the 4-month EEG wait remains unresolved.

For Researchers

app.formahealth.io/research/insights
Research AssistantRural Pediatric Epilepsy Monitoring Cohort

What is the correlation between illness episodes and seizure frequency in the cohort?

Among N=142 pediatric patients, febrile illness episodes were associated with a 3.2x increase in seizure frequency in the 72 hours following onset (95% CI 2.4–4.1, p<0.001). The effect was most pronounced in patients with drug-resistant epilepsy (OR 4.7) compared to well-controlled patients (OR 1.8).

Seizure rate by illness day

Trend

How does travel time to specialist care correlate with caregiver stress?

In rural families (N=89, travel >60 min), mean caregiver stress scores were 7.2/10 compared to 4.8/10 for urban families (N=53, travel <30 min), Cohen's d=1.1. Families with travel times exceeding 2 hours reported 2.3x more missed workdays per month (p<0.01).

Stress by travel time quintile

Trend

Can voice-logged cognitive observations predict EEG-confirmed subclinical seizure activity?

Caregiver-reported cognitive decline (Cognitive Clarity score drop ≥3 points over 48 hours) preceded EEG-confirmed subclinical spikes with 72% sensitivity and 68% specificity (AUC 0.74, N=64 patients with paired voice-EEG data). Word-finding and memory domains were the strongest individual predictors.

Sensitivity by cognitive domain

Trend

What proportion of nocturnal seizures are captured only through caregiver voice logs?

Of 312 suspected nocturnal seizure events logged by caregivers, only 38% were subsequently confirmed by video-EEG or witnessed. Voice logs captured 2.6x more nocturnal events than paper seizure diaries in a paired comparison (N=47 patients, 12-week crossover), with the 'Suspected but unconfirmed' classification capturing 89% of the additional events.

Events captured by method

Trend

Mon, Jun 1Day 1

Baseline day. Liam had 4 absence seizures and one focal seizure (left arm, 45 seconds). Sleep was 6 hours with 2 wakings.

Seizure Burden 2Sleep Quality 2
Wed, Jun 3Day 3
💊Epidiolex dose increase

Telehealth appointment with epileptologist. Doctor recommends increasing Epidiolex from 4 mL to 5 mL twice daily based on Liam's weight gain.

Thu, Jun 4Day 4

First day on higher Epidiolex. Liam was drowsy and nauseous. Only 2 absence seizures though — down from the usual 4-5.

Seizure Burden 1Medication Side Effects 2
Sat, Jun 6Day 6

Liam had a tonic-clonic in his sleep. Tracy heard sounds at 2 AM and found him postictal. She was up the rest of the night monitoring.

Seizure Burden 3Caregiver Sleep 3
Mon, Jun 8Day 8
First seizure-free school day

Liam made it through a full school day with no seizures. Teacher reported good focus. Tracy slept 7 hours — first full night in a week.

Seizure Burden 0Cognitive Function 1
Wed, Jun 10Day 10

Drove 2.5 hours to Winnipeg for EEG. Results showed fewer epileptiform discharges than 6 months ago.

Fri, Jun 12Day 12

Blood ketones at 3.4, glucose at 4.0. Dietitian happy with the modified Atkins compliance. Liam's nausea from the Epidiolex increase is resolving.

Medication Side Effects 1
Sun, Jun 14Day 14
Febrile illness onset

Liam came down with a cold and fever of 38.6. Tracy is bracing for breakthrough seizures — illness always lowers his threshold.

Illness Severity 2
Mon, Jun 15Day 15
🚨Illness-triggered seizure cluster

As expected — 3 tonic-clonics in 4 hours. Tracy gave Diastat after the third one went past 4 minutes. Missed school all week.

Seizure Burden 3School Impact 3
Thu, Jun 18Day 18

Fever broke overnight. Only 1 absence seizure today. Liam was foggy but improving. Tracy slept a full night for the first time since the illness started.

Seizure Burden 1Cognitive Function 2Caregiver Sleep 1
Sun, Jun 21Day 21

Back to baseline. 2 absence seizures, good energy, keto compliance solid. Tracy's mom is visiting for the week — first respite in months.

Seizure Burden 1
Thu, Jun 25Day 25
Milestone — wrote full name independently

Best cognitive day in weeks. Liam wrote his full name independently at school and the teacher reported strong engagement. VNS side effects (hoarseness) have subsided since the settings were lowered.

Cognitive Function 0Medication Side Effects 0

Key insight

Over 25 days, Tracy's logs captured a clear arc: the Epidiolex dose increase on Day 3 reduced Liam's daily seizure count from 5 to 1-2, but a febrile illness on Day 14 triggered a breakthrough cluster that took 4 days to resolve. The longitudinal pattern — medication adjustment, improvement, illness-triggered setback, recovery — is exactly the signal that 6-monthly clinic visits cannot detect.

See Rural Pediatric Epilepsy configured for your study

See Forma configured for Pediatric Epilepsy