Back to Side Effects

When a Side Effect Is Not in the Leaflet

When a Side Effect Is Not in the Leaflet
Prescription Medicines Editorial 14 January 2025

Finding that your symptom is not in the leaflet is usually taken as evidence that the medicine is not responsible. It is not. Leaflets list what was seen in trials and what has been reported since, and both have limits - trials are too small to detect rare effects and too short to detect slow ones. An unlisted reaction is the single most valuable kind to report, because that is precisely how the list grows.

Why leaflets are incomplete

A medicine is licensed on trials involving perhaps a few thousand people. An effect occurring in one person in ten thousand will not appear at all.

Trials also run for months rather than years, so effects developing slowly are invisible - olmesartan enteropathy took years to recognise, and the ACE inhibitor cough can begin months after starting.

Trials exclude the people most likely to have problems: older people on several medicines, those with kidney or liver impairment, pregnant women and children.

And trials measure what they set out to measure. Effects nobody thought to ask about - taste changes, vivid dreams, sexual difficulties - are systematically under-captured.

So the leaflet reflects what has been found so far, not what exists.

See reading the patient information leaflet.

Things found after licensing

Several now-standard warnings were once nowhere in any leaflet.

Montelukast's neuropsychiatric effects. The fluoroquinolone tendon warnings. Sublingual buprenorphine and dental decay. The valproate pregnancy evidence, which took decades. SGLT2 inhibitors and euglycaemic ketoacidosis.

Each began with people reporting something that was not in the leaflet, and clinicians who did not dismiss it.

Judge it the way a clinician would

Before concluding either way, the useful questions are the same ones a prescriber asks.

Timing. Did it start within days or weeks of starting the medicine, or of a dose increase? That is the strongest single clue.

Dose. Did it worsen when the dose went up?

Stopping. Did it improve when the medicine stopped, and return if it was restarted? That sequence is close to definitive.

Plausibility. Do other medicines in the same class cause something similar? An effect shared across a class is more likely to be real.

Alternatives. Could the condition itself, another medicine, or something unrelated explain it?

Write the dates down as things happen. Reconstructing a timeline months later is unreliable in a predictable direction.

See when a side effect is actually the illness and keeping a side effect diary.

Report it anyway

The threshold for a Yellow Card report is suspicion, not proof, and this is exactly the situation the threshold was set low for.

Make clear in the report that the effect is not listed - that flags it for attention.

Include the medicine by generic name, the brand and batch number if known, when each started, the dose, everything else you take including supplements, and what happened when the medicine stopped.

Anyone can report: the person affected, a relative, a carer or a health professional.

See how the Yellow Card scheme works and what happens after you report.

Where new effects are most likely

Worth knowing, because these are where reporting is most valuable.

Black triangle medicines, marked with an inverted triangle, are under additional monitoring - and for these all suspected reactions should be reported, listed or not.

Newly licensed medicines generally.

Off-label use, which is common in children and palliative care and sits outside the evidence the leaflet reflects.

Unlicensed specials and imports, where there is no UK trial data at all.

Older people on several medicines, and children - the two groups most excluded from trials and most under-reported.

See black triangle medicines and unlicensed specials and imported medicines.

If you are dismissed

Being told "that is not a side effect of that medicine" happens, and it is sometimes wrong.

Ask for the timeline to be recorded in your notes, so the observation is not lost.

Ask whether a supervised trial off the medicine is reasonable, where stopping is safe.

Report to the MHRA yourself - you do not need anyone's agreement to do so.

Ask for a second opinion or a pharmacist review if it continues to matter. Practice pharmacists are frequently the most useful people for exactly this question.

Do not stop a medicine abruptly on your own - several cause harm if withdrawn suddenly.

Keep it in proportion

Not every symptom appearing during treatment is caused by it.

People are unwell for unrelated reasons all the time, and expectation genuinely influences what gets noticed and reported.

The right posture is neither to assume the medicine is responsible nor to assume it cannot be - it is to record what happened, report it, and let the pattern across many people settle the question.

See medication reviews.

If this was useful, Reporting on Behalf of Someone Else is worth reading next, and Side Effects collects the rest.