How to turn a pile of medical notes into one readable page

2026-08-25 · 3 min read

medicalplanningcaregiving

A new specialist gets a short appointment and a history they have never seen. If it arrives as a story told out loud, most of it will not survive the retelling. If it arrives as one page in date order, it can be skimmed while you are still sitting down.

Building that page is a sorting job, not a memory test.

Work from paper, not from recall

Get everything into one pile first: appointment letters, results, discharge summaries, pharmacy labels, calendar entries, the diary. Each one carries a date somebody else wrote down, which is worth more than a date you reconstruct.

Fill in the anchors before the middle. The first time this came up. The first appointment about it. Anything with a date printed on it. Once the ends are pinned, the middle stops moving around.

One event per line, and four parts

The category is your own word, and useful ones are ordinary: symptom onset, appointment, test, result, medication started, medication stopped, procedure, hospital stay. Category is what turns a wall of lines into something a reader can jump around inside.

Leave partial dates partial

People genuinely know some things to the year, some to the month, and some to the day. 2015, 2016-04 and 2016-04-18 are all real answers, and each should be written at the precision you actually have.

Never round a vague memory up into a specific day to make the sorting easier. A history with a hole in it is honest. A history with an invented day in it is wrong, looks deliberate, and is exactly the sort of thing a reader skims straight past.

Avoid slash dates entirely. 03/04/2019 is April the third in most of the world and March the fourth in the United States, and guessing wrong on a medical history is not worth the keystrokes saved.

Gaps are worth showing

The longest stretches where nothing was written down are useful to see, because they show where your own record is thin and what you might still be able to dig out. A gap means nothing was recorded between two lines. It never means nothing happened, and the page should not be read as saying so.

The interpretation the tool refuses to do

Our timeline sorts, groups by year and month, counts by category, indexes the earliest line you recorded for each symptom, and measures the distance between your own dated events. It stops there. No ranking, no pattern, no severity, no cause and no link claimed between any two lines. Those are conversations for a clinician holding the actual letters.

There is one refusal in it worth spelling out. The bare words start, started, stop and stopped are deliberately not treated as medicine categories. Somebody typing started usually means whatever their own description says — started physio, started school, started a new job — and turning that into a medicine event would put a factual error about somebody's history onto a page they hand to a specialist. Those lines are kept under your own word instead. It never reads your description to work out a category for you either, for the same reason.

A line it cannot read is handed back to you verbatim with the reason, and an impossible date is rejected rather than quietly moved to the nearest real one. A timeline that silently dropped two events would be worse than no timeline.

Building yours

The medical history timeline takes that pasted list and lays out one clean chronological page you can print and hand over, along with the summary counts and the gaps. It runs entirely on your device and nothing is uploaded, which matters here more than almost anywhere — this is some of the most sensitive information a person owns.

Say plainly what it is when you hand it over: your own record, typed by you, unverified against any source. It is not a medical record and it does not replace one. Bring the underlying letters and reports; a clinician will need them. What the page saves is the fifteen minutes that would otherwise go on reconstructing dates.