Interactions: Talking to patients

Last updated: Monday, November 15, 2021

We're going to consider three elements of your approach to advising patients about interactions: What are the potential consequences? How likely is it you’ll be affected? What are we doing about it?


Explain the interaction to your patient in a clear and meaningful way. What might happen to them because of the interaction? Avoid jargon. For example, be aware that the word ‘interaction’ itself is not widely understood by most patients. You need to use everyday words. For example: “We’ve stopped X because it made your warfarin too strong, and we were worried it might make you bleed. We'll monitor you, but you mustn’t take X again.


When talking to patients about the risk associated with an interaction, it helps if you are clear whether an adverse outcome is:

  • Theoretical – some interactions are listed in patient information leaflets or SmPCs because it is conjectured that they might occur, not because they’ve been seen in practice. For example, some drug combinations are theorised as prolonging the QT interval, but the implications for real patients are uncertain. 
  • Possible – studies suggest that only some patients are affected. Try and quantify the risk if possible. “Most patients don’t seem to have a problem taking these two drugs together, but we know that a few patients can, so we want to be careful.” 
  • Likely or certain – sometimes the interaction affects the majority. For example, rifampicin reduces the plasma levels of methadone so commonly, that long-term methadone recipients often need a methadone dose increase when rifampicin is started. 

In any event, try and contextualise the risk by relating it your patient’s personal situation. Spell out the implications for them. Telling patients that an interaction might ‘sedate them’ isn’t that helpful; it’s better to explain that you’re worried about them having a fall because of it. Sometimes the patient’s medical condition or age makes an interaction more of a concern: “You may be a bit more at risk because your kidneys aren’t working well”.


What is being done to reduce the risk of an adverse outcome for the patient? Has a drug been changed or a dose of either drug been altered? Is there any biochemical or other monitoring in place? It’s important to ensure that there’s a plan that identifies who will do the monitoring – the GP, outpatient clinic, consultant, or the patient themselves – as well as what they will do.

You should explain the nature of an interaction to the patient and describe what may happen to them if they are affected: what will the symptoms be? You should also explain if they can self-monitor, and what to do if they are affected (e.g. stop drug, see GP, self-treat etc).

Quick clinical question 
A patient asks you if she can drink alcohol when taking warfarin. What do you advise?    Consider, then click for answer.
Drinking within government safety limits is OK (up to 14 units per week spread over the week). However, don't get drunk because it might make you bleed and avoid regular heavy drinking as it may stop warfarin working.