Keeping track of medication changes
“It didn’t work” is three completely different facts — and only one of them is true for you.
The first prescription is easy to remember. So is the first dose change, roughly. By the third medication and the fourth adjustment, most people are working from a genuinely unreliable memory: a name they are fairly sure about, a dose they would guess at, a stretch of time they would describe as last winter, maybe?
That is not carelessness. Nobody hands you a form for this, the changes happen months apart, and the periods you are being asked to recall are precisely the periods when your memory was working least well. It is a completely predictable gap, and it quietly costs you.
Why the history is worth more than you would think
Here is the thing that makes this worth five minutes of admin: “it didn't work” is not one fact. It is at least three, and they lead in different directions.
- I took it at a full dose for two months and my mood did not shift.
- I felt sick for three weeks and stopped taking it.
- I never got past the starting dose because the appointment to review it kept slipping.
All three get summarised, months later, as “I tried that one, it didn't work.” But to the person deciding what to try next, they are not remotely the same. The first says something about the medication. The second says something about tolerability, which is often solvable with timing or a different option. The third says nothing at all about the medication — it says the plan never actually happened.
If the record has collapsed into “didn't work,” an option that might genuinely have suited you can get quietly crossed off a list for the rest of your life.
The four things worth writing down
You do not need a spreadsheet. You need four fields, filled in at the moment of change while you still know the answer.
What — the name, spelled properly. “One of the sertraline-type ones” is where most histories start losing resolution.
How much, and when it changed. The dose matters, but the dates matter more than people expect, because almost everything you will be asked later is really a question about timing — did the side effect arrive with the increase, did the improvement follow it, how long were you actually at that dose.
How long you were on it. Not “a few months.” Start date, stop date. Six weeks and six months are very different pieces of evidence.
Why it changed — in your own words, on the day. Not the tidy version you will reconstruct later. “Stopped because I couldn't sleep” is worth ten times “didn't get on with it.”
Record the boring bits too
There is a natural bias toward logging only bad days, because bad days are what prompt you to reach for your phone. It gives you a record that reads worse than the period it describes, and — more practically — it leaves out the two things a prescriber will actually reach for: the dates, and the stretches where nothing much happened.
An uneventful fortnight is real information. It is the thing that turns “I have been up and down” into “I was steady for two weeks after the increase, then it slipped.”
What this buys you in the room
Review appointments are short and they open with a question you cannot answer well from memory: so, how have you been getting on with it?
Arriving with the answer already written changes what the appointment is for. Instead of spending most of it reconstructing a timeline out of fragments — which is a bad use of a scarce ten minutes, and produces a shaky timeline anyway — you start from an agreed set of facts and spend the time on the decision.
This is what Nagi's medication diary and doctor report are built to do: keep the names, doses and dates as you go, and put them on one page when you need them. But the format is not the point. A note in your phone, updated on the day something changes, does most of this. The value is in recording it at the time, not in the tool.
One line that is not optional
None of this is a basis for changing anything yourself. Keeping a good record makes the conversation with your prescriber better; it does not replace it, and it is not a licence to adjust a dose, skip days, or come off something because the chart looks fine. The NHS puts it without qualification: “Do not stop taking an antidepressant suddenly or without talking to your doctor first.”1 Withdrawal effects can start within days and run for weeks, and they can be mistaken for the depression itself returning.
The record is evidence for the appointment. The decision stays in the appointment.
The short version
Write down four things every time something changes — what, how much and when, how long, and why. Do it on the day. In two years, when someone asks what you have tried, you will have an answer that opens doors instead of closing them.
— Richard MacLean
Related: is my antidepressant working? and keeping a side-effect diary.
Sources
- NHS, Overview — Antidepressants. nhs.uk — accessed 10 August 2026.
Nagi is a mental health companion, not a medical device, and nothing here is medical advice. Never start, stop, or change the dose of a prescribed medication without talking to your prescriber.