In epilepsy trials the primary endpoint is often a count from a diary, so the diary design is the endpoint design. This template logs seizure-free days and, for each seizure, the type, duration, trigger, rescue medication and recovery.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Daily seizure diary · Day 23
Subject 002-0009 · Demo study · today
Seizures today
Seizure 1: type
Seizure 1: how long
Rescue medication taken?
At a glance
Why the diary matters
Epilepsy is episodic, so a clinic visit every few weeks says little about what happened between them. Regulators and sponsors therefore rely on prospective diaries: participants record each seizure as it happens or at the end of the day, and the trial compares seizure rates in a baseline period with those on treatment. Responder analyses, such as a 50 percent reduction from baseline, are built from the same counts. The overall neurology picture, including clinician-rated scales, is covered in EDC for CNS and neurology trials; this page goes deeper on the diary itself.
Because the count is the data, small design choices change the result. A day with no entry is missing, not seizure-free, and the analysis plan needs a rule for the minimum number of diary days per period. A diary that is tiring to complete produces missing days. And an unclear seizure definition produces counts that differ between participants. Each field below is there to protect one of those three things.
The ILAE 2017 classification groups seizures by onset (focal, generalised, unknown), awareness and whether there is motor activity. A participant cannot be expected to use that vocabulary. A practical diary offers three or four plain-language categories that the investigator maps to the classification, for example convulsive, non-convulsive with loss of awareness, non-convulsive with awareness kept, and not sure. Train participants on the categories at enrolment and map them to the protocol seizure types in the data dictionary.
Triggers are optional context, such as missed medication, poor sleep, illness, stress, alcohol or flashing lights, and should be a short multiple-choice list with an "other" option. Recovery time (how long until back to normal) and injury are useful safety signals, and a prompt for them also records the rescue medication given. Never put doses or clinical decision rules in a diary: those come from the protocol and the treating team.
For children, people with cognitive impairment, or seizures with impaired awareness, a caregiver completes the diary. Record who completed it on every entry, and see the parent and proxy ePRO template for the design points that apply.
Diary days with data
91%
Overdue today
4
Rescue medication entries this week
3
Template
| Field | Question type | Values | Why it is there |
|---|---|---|---|
| Any seizures today? | Single choice, required | None, 1, 2, 3 or more | Makes seizure-free days explicit; drives all branching |
| Seizure time (per seizure) | Time | Approximate start time | Clusters, nocturnal pattern |
| Seizure type (per seizure) | Dropdown | Three or four plain-language categories plus "not sure" | Frequency by type, mapped to the protocol classification |
| Duration (per seizure) | Numeric | Minutes, with a range check | Long seizures are a safety signal |
| Possible triggers | Multiple choice | Missed dose, poor sleep, illness, stress, alcohol, other, none known | Context for analysis, not causality |
| Rescue medication taken | Single choice | No, yes | Safety and a secondary endpoint in many protocols |
| Time to recover | Dropdown | Under 5 minutes, up to 30 minutes, over 30 minutes, not recovered | Postictal burden |
| Injury or emergency care | Single choice | No, yes | Opens a safety follow-up path |
| Completed by | Single choice | Participant, caregiver | Data provenance |
| Notes | Long text | Free text | Context; coded by the site if needed |
Per-seizure fields are repeated for seizure 1, 2 and 3 using skip logic. Cluster days of 4 or more are recorded as a count with a note.
Run it in Capture
Skip logic shows seizure 2 only when the count is 2 or more, and seizure 3 only when it is 3 or more, so a seizure-free day is one tap. A range edit check on duration raises a query when a value is implausible, and a custom-value check can flag any "yes" to rescue medication or injury for the site's attention.
Evening diary
Question 3 of 8
How would you rate your fatigue today?
0 = no fatigue, 10 = worst imaginable
Build it in the free sandbox, schedule it daily and open the participant link before you finalise the protocol.
How to build it
Write what counts as one seizure, how clusters are counted and the categories shown to participants. The diary should use exactly those words.
Add the required count question, then three groups of per-seizure questions: time, type, duration, triggers, rescue medication, recovery, injury.
Show group 2 when the count is 2 or more and group 3 when it is 3 or more. Hide all groups when the answer is None.
A range check on duration and a custom-value check on rescue medication and injury, with priorities that match your safety plan.
Daily from the start of the baseline period through the end of treatment, with a completion window and reminders. Keep the baseline diary running before randomization.
Put the emergency instruction and who to call in the diary header, and train participants and caregivers on the categories at enrolment.
Submit entries for a seizure-free day, a one-seizure day and a three-seizure day, export them and compute a 28-day rate from the file.
Before go-live
Written in the protocol and matched in the diary wording.
The first question is required, so no entry means missing, not zero.
The analysis plan states how many days with data a period needs.
Participant, caregiver or both, recorded on each entry.
In the diary header and consent form: a diary is not monitored in real time.
The protocol says who reviews rescue medication and injury entries and where follow-up is recorded, for example the adverse event form.
Seizures per 28 days is computed from the export, since Capture's calculated fields are a fixed clinical set.
At minimum whether any seizure occurred that day and, for each seizure, the time, type, duration, any suspected trigger, whether rescue medication was used and how long recovery took. The protocol defines the exact fields and seizure categories.
Usually as the number of seizures per 28 days in a baseline period and in a treatment period, using days with diary data. Capture exports the entries; the rate and any percent reduction are derived in your analysis program.
No. A required first question lets a participant record "none", so a missing entry is genuinely missing. The analysis plan should set a minimum number of diary days per period.
Yes, and many trials require it for children and for seizures with impaired awareness. Record who completed each entry. See the parent and proxy ePRO template for design points.
Not in real time. A custom-value edit check can raise a query for rescue medication or injury, but a diary is not an emergency channel. Give participants emergency instructions in the header and the consent form.
No. It is a data collection template. Clinical decisions, rescue medication instructions and seizure action plans come from the protocol and the treating team.
No. The diary opens in the phone browser from a secure link sent by email or SMS.
Keep exploring
Build the seizure diary, schedule it daily and test it on your phone. Free sandbox, no credit card.