A symptom diary is a structured record of physical or psychological complaints over an extended period. It consists of individual entries. Each entry records what you felt and when, how strong it was, and which circumstances went with it.
A symptom diary is not a diagnostic instrument. It does not replace a medical assessment. It only improves the basis for one, by turning a vague memory into a solid series of data.
Individual bad days say little. Only the series shows structure. Do episodes cluster on particular weekdays? Are the intervals getting shorter? Does intensity rise or fall over months? Questions like these can only be answered if data from several weeks exists in a comparable form.
Memory does not store complaints neutrally. Strong and recent episodes stay present; mild and more distant ones disappear. Asked "how often did you have this in the past three months?", people therefore often answer inaccurately — sometimes far too high, sometimes far too low. A diary kept close to the event corrects this bias. It delivers a number instead of an impression.
Many people have a suspicion: a particular food, lack of sleep, stress, the weather. A diary makes that suspicion testable. You document the suspected trigger and the symptom separately from each other, then look at whether they actually occur together. Often the suspicion does not survive the test. That, too, is a result.
An entry should contain the following:
Not every entry needs every field. But the fields you do keep, you should keep consistently.
Prose describes a moment well and compares badly. "Pretty awful, barely bearable" and "really bad today" cannot be weighed against each other. A 7 and a 5 can.
The value of a fixed scale does not lie in its objectivity. Your 7 is not someone else's 7. The value lies in the comparability of your own values across weeks. That is the whole point.
The same applies to terminology. Draw up a short list of fixed labels — for example "pressure headache", "throbbing headache", "bloating", "abdominal cramps" — and stick to it. If you call the same complaint "pulling" one day, "stabbing" the next and "feeling off" the day after, a single symptom series falls apart into three useless fragments.
Free text still has its place. Use it to add anything unusual, not as a substitute for the structured fields.
Only documenting on bad days. This is the mistake with the biggest consequences. If only bad days appear in the diary, the basis for comparison is missing. Every supposed trigger then looks suspicious, because you never checked how often it was also present on symptom-free days. Correction: Record good days as well, at least with a short entry saying "no complaints" plus the most important accompanying circumstances.
Recording interpretations instead of observations. "The lactose intolerance again" is not an observation, it is a conclusion. It locks in a hypothesis before it has been tested, and colours every later entry. Correction: Note only what you perceived — symptom, intensity, time, circumstances. Suspicions belong in a separate field or a separate note.
Recording too many fields. A diary with thirty fields per entry is empty after two weeks. An abandoned diary is worth nothing. Correction: Start small. Five to eight fields you can reliably sustain. You can always add more later.
Switching terms. See above. Correction: A fixed list of terms, ideally as selection fields rather than free text.
Entering too late. Anyone who reconstructs the whole week on Sunday is documenting their memory, not their symptoms — and thereby imports exactly the bias the diary was meant to remove. Correction: Record during the episode or immediately afterwards where possible. If that does not work, then on the same day. A roughly estimated value from today beats a precise-looking value from five days ago.
There is no universal schedule. The useful rule of thumb is: long enough for the pattern you want to find to repeat several times.
Complaints that occur several times a week often reveal their structure after just a few weeks. Complaints that occur every few weeks need several months accordingly. Anything that might be tied to a monthly cycle or to the seasons needs several cycles or several months before any statement is possible at all.
More important than total duration is completeness. A continuous diary over six weeks says more than a patchy one over six months. So plan your record-keeping to be sparse enough that you can manage it on a bad day too.
This is the point at which symptom diaries are most often misread.
If your diary shows a pattern of the form whenever X then Y, you have found a correlation. A correlation is a hypothesis. It is not proof of a connection, and certainly not proof of a cause.
There are several reasons for this. A third factor can trigger both: stress often changes sleep, eating behaviour and complaints at the same time — the food then looks like the trigger, but is only a side effect. The direction can be reversed: someone who feels unwell eats differently and sleeps differently, so the symptom actually precedes the supposed trigger. And when many fields are recorded, chance matches are inevitable — the more fields, the more spurious patterns.
A practical consequence follows from this. Do not draw your own conclusions from a diary pattern and act on them. This applies to two things in particular:
The right way to handle a pattern you have found is to bring it along as a question: "I noticed that A and B often occur together. Is that plausible? How could it be tested?"
Raw data is almost worthless in an appointment. Forty printed pages cost time that is then missing from the conversation. Condense it instead onto one page with four blocks:
Take the complete data with you anyway, in case you are asked. It is an appendix, not the main document. Photos are especially valuable here, because visible findings have often just subsided by the day of the appointment.
In LogYourHealth this condensed view comes straight out of the entries: history, intensity curve, photos and status per symptom are already there in structured form and can be taken along as an overview.
In favour of paper: It is available immediately, needs no battery, raises no data protection questions and can be laid out entirely as you like. Many people write more reliably on paper, because no device stands between them and the entry.
Against paper: Analysing it is laborious. If you want to know frequencies and trends across months, you have to count by hand. Photos cannot be integrated. A lost notebook is gone for good, and loose sheets are awkward to bring to an appointment.
In favour of an app: Fixed selection fields enforce consistent terms. Trends and frequencies are generated automatically. Photos belong directly to the entry. Reminders help against gaps, and the export for the appointment is one button press.
Against an app: Health data is particularly sensitive. Before you choose, check where the data is stored, who has access to it, and whether you can export and delete it.
A symptom diary does not make a diagnosis. It rules nothing out and confirms nothing. It supplies structured observations — interpreting those observations is medical work and requires examination, history and, where relevant, test results that no diary contains.
What it does deliver is nonetheless substantial: it replaces "sometimes I feel unwell" with "19 episodes in seven weeks, average intensity 6, clustered in the mornings". That is a basis you can work with.
The demo starts instantly with realistic sample data. No account, no email address.