Canalith
Canalith
Almost everyone told to "keep a diary" writes too much for the first fortnight, then stops. A useful diary is not a longer diary. It is a short, fixed set of facts, captured in the minute after an episode, in the same shape every time — because a clinician reads the pattern across weeks, not the prose of any single entry.
Here is what to record, why each item earns its place, and how to turn eight weeks of entries into one page that is worth the appointment.
The date and the clock time the episode began — not the time you got round to writing it up. This is the field people are laxest about and the one that carries the most structure. Timestamps are what let anyone see that your episodes cluster in the first hour after waking, or arrive in runs of three days with a fortnight of quiet between, or have quietly halved in frequency since a change in treatment. A note that says "a few times last week" contains none of that.
If you can only do one thing, do this one. A bare timestamp with nothing attached is still a data point.
You do not need a stopwatch. A band is enough, and the same bands each time: seconds, a few minutes, an hour or more, all day. What matters is that you use the same vocabulary, because duration is one of the sharpest dividing lines between the causes of vertigo. Brief spells triggered by movement look quite different, on paper, from spells that last hours and come with hearing changes, or from vertigo that is present continuously for days. A clinician sorting through possibilities will reach for duration long before they reach for severity.
Estimate from the start of the spinning to the point where the room stopped moving — not to the point where you felt entirely normal again, which is often much later and worth recording separately if it stands out.
Name the movement, as concretely as you can: rolled over in bed, lay down, sat up from lying, tipped the head back at the sink, bent forward to the bottom drawer, looked up at a high shelf, turned quickly while walking — or nothing at all, it began while I was sitting still.
"Nothing at all" is a real and important answer. Do not leave the field blank because there was no trigger; write that there wasn't one. Whether a change of head position reliably provokes the episode is among the most informative single facts you can bring to an appointment, and it is only visible if the non-triggered episodes are recorded with equal care.
If you rolled or tipped, which direction: left, right, or unsure. "Unsure" is a legitimate entry and better than a guess, but try to catch it, because a consistent side across many entries is a pattern that can be tested directly during an examination. Do not attempt to conclude anything from it yourself — which ear and which canal are involved is determined by a clinician performing a positional test and watching your eyes, not by counting entries in a notebook.
One to ten, the same scale every time. Nobody else can calibrate your six, and that is fine: the value of a consistent personal scale is not comparability between people but readability over time. If your worst episodes were nines in June and are sevens in September, that is a change worth noticing, and it only exists if you were rating the same way in both months.
Rate it as soon as the episode is over. Severity remembered a day later drifts towards whatever the last bad one felt like.
Keep a short fixed checklist and tick it rather than writing prose: nausea, vomiting, headache, sensitivity to light or sound, fullness or pressure in one ear, ringing, any change in hearing, visual disturbance, sweating, palpitations, faintness.
Accompanying symptoms are often what separates one working diagnosis from another, and they are precisely what memory drops first. A checklist takes five seconds and returns something a paragraph rarely does: reliable negatives. Knowing that thirty episodes came with no hearing change at all is genuinely useful information.
Record any medication you took and whether it appeared to help. Note unusually short sleep, a missed meal, an illness, a long drive, an unusual stretch of screen work. Some people also track the weather, because falling barometric pressure is a commonly reported trigger for vestibular migraine and is often mentioned by people with Meniere's — recording the pressure at onset is one way to find out whether it holds for you rather than assuming either way.
Two cautions. Keep this to observations, not theories; a diary that records what happened is evidence, and one that records what you concluded is an argument. And resist adding fields. Every extra column is a reason to skip an entry at two in the morning, and the entries you skip are usually the bad ones.
If a clinician has prescribed a repositioning or habituation exercise, log every time you performed it: date, which side, how many repetitions, and what happened afterwards. Whether you have been able to do it as prescribed, and what followed, is one of the first things you will be asked at a follow-up.
Log the exercise only if it was prescribed to you. Which procedure is appropriate depends on a diagnosis of which ear and which canal is involved; a general explainer on how the named manoeuvres differ is not a reason to start one.
Bring a summary, with the full log behind it. Most consultations do not have time to read eight weeks of entries, but they have a minute for a page that says: how many episodes over what period; the typical and longest duration; the usual trigger and the usual side; the average and worst severity; what usually accompanies them; what you have tried and what followed. The raw entries then answer whatever the page raises.
A few habits make the log worth summarising. Log the episode, not the day, so twelve short spells in one morning read as twelve rather than as "a bad Tuesday". Log the absent days too, or at least keep a calendar view, since a run of clear weeks is as meaningful as a run of bad ones. Do not backfill from memory more than a few hours later; an honest gap is better than an invented entry. And keep going for at least six to eight weeks — patterns in vertigo rarely resolve inside a fortnight.
All of this is achievable with a notebook by the bed, and if that is what you will actually use, use it. The reason we built Canalith is the moment of capture: writing anything at all is hard in the thirty seconds after a spinning episode, which is exactly when the details are still accurate. In the app, severity is a single drag on a dial, the symptoms are a checklist of taps, the time is taken for you, and the barometric pressure and weather at that instant are attached without you doing anything. It keeps the diary and exports the one-page summary described above. It records and it guides an exercise you were already prescribed; it does not diagnose or treat anything. It is not on the App Store yet — there is a waitlist on the home page, and any questions can go to support.
This is a guide to record-keeping, not medical advice, and a diary is not a substitute for an examination. See a clinician about new or recurring vertigo. Treat it as urgent — emergency care, the same day — if vertigo comes on suddenly together with any of: double vision, weakness or numbness on one side, trouble speaking or swallowing, a severe or unusual headache, difficulty walking or standing, or new hearing loss in one ear.