Skip to main content

Seizure Diaries: Witness Accounts and Useful Neurology Follow-Up

A seizure diary is useful when it preserves what happened before, during and after an event, who observed it and how certain the timing is. Keep event descriptions separate from the clinician's diagnosis and record the days actually observed. A blank day is not proof that no event occurred, and a diary or video must never delay emergency care or replace specialist assessment.

By Bractis · Updated · 8 min read

Indian neurology and general-practice teams reviewing suspected seizures, patient-held diaries and caregiver accounts across visits.

Start with an event account that can be checked

Ask the patient and witness to describe the sequence before selecting a diagnostic label. Record the activity immediately beforehand, the first change noticed, what the witness saw and how the person recovered. Separate the patient's remembered experience from what someone else observed.

NICE NG217 recommends a detailed history and examination after a first suspected seizure, using eyewitness accounts and available video where possible. This supports preserving evidence; it does not mean that a video or a completed diary confirms epilepsy.

A witness may arrive late, estimate duration or learn part of the story from another person. Record those limits. 'Observed movements for about one minute after entering the room' is different from 'the complete event lasted one minute'. Neither should be silently converted into an exact duration.

Give each event a date and a local record identifier so later corrections can refer to the same episode. Two family members describing the same event should not automatically create two seizures in a monthly count.

Keep urgent assessment independent of diary completion. A person who needs emergency care should receive it through the clinical pathway; obtaining a witness statement or recording video comes afterwards when appropriate.

Source: NICE NG217: Diagnosis and assessment of epilepsy

Separate onset, observed event and recovery

A diary should allow several times rather than forcing the entire episode into one duration field. Where possible, record the last point at which the person was known to be usual, the event's observed start and end, and the recovery milestones described.

In a fictional account, a witness notices the event at 08:14, reports that visible movements stopped at 08:16 and says usual conversation resumed around 08:31. The observed movement interval is two minutes; the later 15-minute interval concerns the witness's recovery description. Calling the whole 17 minutes a measured convulsion would change the evidence.

Those timestamps are illustrative and are not an emergency threshold. If a time was estimated, label it approximate. If the onset was not seen, do not set onset equal to the time a phone recording started.

Record injuries, assistance given and any prescribed rescue treatment actually used, with its source instruction and times. This article does not provide a rescue dose or a first-aid procedure. The treating service should supply an individual action plan and teach the appropriate caregiver.

Preserve other symptoms and concerns that do not fit the usual pattern. A form should not make a witness choose a familiar event type merely because there is no space for an uncertain description.

On a narrow screen, scroll the table horizontally to read all columns.

Fictional witness timeline: keep observations separate
TimeWhat was reportedLimit of the evidence
08:14Witness first notices the episodeEarlier onset is possible if not observed
08:16Visible movements reportedly stopDoes not by itself establish full recovery
08:31Usual conversation reportedly resumesApproximate recovery milestone, not a measured seizure end
Later reviewClinician assesses the accountDiagnosis and next action belong to this assessment

Count recorded events without inventing event-free days

State the intended observation period and how completely it was recorded. If a diary covers 14 calendar days but only ten have entries, keep the four missing days visible. They may represent forgetting, an unavailable caregiver or uncertainty rather than an absence of symptoms.

In a fictional diary, three events are recorded over ten completed days within that 14-day period. Report those facts. Do not label the month 'three seizures', turn the four blank days into zeros or extrapolate a daily rate as though observation were continuous.

Distinguish a completed entry saying no event was noticed from an uncompleted entry. Even a completed day reflects the limits of the observer and the person's awareness; it is not an assurance that every event would have been recognised.

If the clinic uses a frequency measure, define the observation period, counting rules and missing-data treatment. A change from caregiver-recorded events to patient-only recording can change the apparent count without establishing a treatment effect.

Bring the actual event descriptions into the review. A total can hide a new event pattern, different recovery or injury that matters more than the arithmetic.

Connect events to actual use without assuming a cause

Record the current authorised medicine plan and what the patient reports taking around the event. Include changed products, uncertain doses, interruptions and other relevant medicines. A filled prescription is evidence of supply, not complete exposure.

Collect context such as sleep, illness, work routine and other circumstances the clinician wants reviewed. Keep observed facts separate from a claimed trigger. Two episodes after a late shift are a reason to discuss the pattern, not proof that the shift caused them.

Do not turn a diary entry into an automatic dose change. The treating clinician needs the history, examination and other relevant evidence. Any new instruction should identify its author, effective date and which previous instruction it replaces.

Ask what prevents the patient from maintaining the diary. A short paper form used consistently may preserve more useful evidence than a complex application that is rarely completed. With the patient's agreement, identify whether a caregiver can add observations while preserving who wrote each entry.

Handle video as sensitive evidence with a defined purpose

Recording must not interfere with helping the person or obtaining urgent care. If video already exists and the patient agrees to its use, retain the relevant source and consent context through the clinic's approved record-handling process.

Keep the recording's timestamp, who supplied it and what part of the episode it shows. A clip beginning after onset should not be described as a complete event. A cropped copy or edited file should remain distinguishable from the original supplied evidence.

Do not forward the clip through an unrestricted staff group or reuse it as teaching material because it was sent for clinical review. A family member's possession of a video does not by itself establish permission for every other use.

The review note should identify which account or clip the clinician considered, what remains uncertain and the next assessment or follow-up. If additional evidence is requested, name the question it is meant to answer rather than asking the family to collect more recordings without a purpose.

Two messages describe one episode

A fictional patient sends a brief diary entry and a caregiver later sends a fuller account with a video. Staff link both to the same event rather than increasing the monthly total to two. The clinician can compare the descriptions, identify that the clip starts late and document the interpretation. The four uncompleted days elsewhere in the diary remain missing. The record improves because its evidence and denominator are clearer, not because the count looks precise.

Seizure-event review worksheet

Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.

Event identity
Are multiple accounts linked to the same episode with their sources preserved?
Timing
Which intervals were actually observed and which were estimated or unseen?
Coverage
How many days were completed, explicitly event-free as reported, or missing?
Clinical review
Who reviewed the evidence, what remains uncertain and what action plan applies?

Questions doctors ask

Can a seizure diary establish epilepsy?

No. It supplies history for the clinician's assessment. Keep the event account separate from the diagnosis and the investigations or specialist review considered necessary.

Should blank diary days be entered as zero events?

No. Identify missing days separately from completed entries reporting no observed events.

Is recovery time the same as seizure duration?

Record the observed episode and later recovery milestones separately. The clinician interprets them; staff should not combine different intervals into a single precise seizure duration.

Should a caregiver prioritise recording an episode?

No. Helping the person and obtaining necessary urgent care take priority. Available recordings may assist assessment when handled with appropriate consent and record safeguards.

Does a repeated diary association prove a trigger?

No. Preserve the pattern as a question for review and avoid changing medicines or labelling a cause automatically.

Sources and scope

Updated on . This guide structures evidence and follow-up. It is not a seizure-diagnosis tool, rescue-treatment protocol or medicine-adjustment rule. No independent clinical or legal review is claimed.

Try the workflow
with Bractis.

Review two fictional witness accounts of one event and a partially completed diary. Inspect event linking, uncertainty and the clinician's follow-up.

Request a clinic demo