EARLIER this year I was in hospital for urgent surgery after it was found I had an ectopic pregnancy. While waiting in the anaesthetic bay at around 10 pm, the anaesthetist came in to ask me what I assume are all the usual preoperative questions. The nurse had already asked a few and now the doctor was asking those again as well as more. The anaesthetist asked me “do you have any heart problems?” to which I answered “no”.

The anaesthetist was later surprised, and concerned, when during the operation the cardiac monitor showed frequent ventricular ectopic beats.

On my way back to the ward after midnight, a nurse mentioned to me, in my post-anaesthetic haze, that I would need to see a cardiologist. The next day the anaesthetist called me to tell me his concern at what the electrocardiogram had shown, given I had no heart problems. To which I sheepishly replied that I do in fact have a heart problem if one considers benign ectopic beats a problem.

I’d know about this for 4 years at least. So why didn’t I mention it? It would be easy to dismiss it. Perhaps I was emotional and distracted given the reason for the surgery. Maybe it was because of the late hour of the surgery and I was too tired to remember. But when I look back on it, I can see an interactional reason.

To understand medical questioning, we must look at the multiple factors that influence how those questions are designed. As a clinician, you probably don’t even realise you’re making this multitude of small decisions all throughout the consultation. Not just what information you need, but the decisions you make in terms of how you will ask it.

Questions have preferences – both grammatical and social. I’ve previously mentioned the influence grammatical preference can have on patient response. You can see that the question I was asked, “do you have any heart problems,” is more likely to receive a “no” response because of the word “any”. Beyond this grammatical preference, Heritage and Clayman describe both congruent and cross-cutting social preferences that influence question design in medicine.

Question design in medical consultations — that is, how you ask a question — is guided by three key principles: optimisation, problem attentiveness, and recipient design.

Optimisation refers to how questions can be designed to assume a favourable health or social response from a patient. In “do you have any heart problems,” no is the preferred answer because the favourable health response is that I do not ,and there was no reason to necessarily believe that there was a “problem”.

Problem attentiveness refers to the moments when optimisation is not appropriate, usually in regard to the presenting concerns or knowledge of another existing health concern. If I’d been seeing a cardiologist for palpitations, on the other hand, and they’d asked “do you have any heart problems?” I might think it odd because that would be why I am there.

Recipient design refers to asking a question with the addressee or recipient’s circumstances in mind. This is a more general principle that guides conversation. Perhaps if I had been older or male, the recipient design principle would have prompted the anaesthetist to ask it in a different way based on the assumptions related to that demographic information.

These principles might seem unnecessarily technical, but they provide insight into why you design questions in particular ways, why a patient might answer with a “no” to you but with information about a problem to another clinician, and why sometimes it can seem tricky to ask a question when these principles are cross-cutting.

“Do you have any heart problems” is not inherently wrong, it just didn’t work for me. With the “any” as well as the use of the word “problem” when the ectopic beats had previously been considered not a problem, I was not prompted to disclose relevant information – it didn’t even cross my mind. What could have the anaesthetist asked me? If the question had been “have you seen your GP or a cardiologist about your heart?”, I would have been more likely to say “yes”.

Dr Sarah J White is a Senior Lecturer in the Faculty of Medicine, Health and Human Sciences at Macquarie University. Dr White is the current Australian National Representative for the International Association for Communication in Healthcare.




The statements or opinions expressed in this article reflect the views of the authors and do not represent the official policy of the AMA, the MJA or InSight+ unless so stated.

6 thoughts on “Asking the right question for the patient in front of you

  1. Peter McLaren says:

    As an anaesthetist, I find that by the time I get to the patient they are all answered out. The hospital’s website, the secretary at the admissions desk, the admitting nurse, the anaesthetic nurse and the scrub sister have all interrogated them. However, the magistrate at an inquest will not expect me to receive my information secondhand. This elaborate process is perpetrated by lawyers, has no research validity and is like a motherhood statement: ‘If asking the question once is safe, then asking it five times must be safer’. The admission questionnaire which was once called a ‘preanaesthetic questionnaire’ is now deemed a ‘pre-admission health questionnaire’ and we are no longer consulted when it is updated.
    The other factor, probably not in this case, in patients neglecting to give their anaesthetist a full history is when they deem them a rather minor player in their health outcome.
    I was still rather amused that the author accepted no blame for the mistake she outlines. I would have thought the question; ‘Do you have any heart problems?’ covered most things.

  2. Sue Colen says:

    I’m sorry for your loss Sarah.
    This topic is so pertinent and has recently been addressed by a Palliative Care Physician, Dr Kathryn Mannix, in a book entitled, Listen, how to find words for tender conversations. The right question can certainly open doors for factual details as Sarah points out, but also for significant emotions to the extent that the patient feels they have been heard.

  3. Dr Chris Hammersley says:

    Terrific article, Ms White!

    I remember asking for guidance as a medical student in relation to *the way to ask questions*, not just algorithmically, but also with emotional intelligence.
    My registrar looked befuddled, and said “Just go down the list, they’re all there!”. I instinctively knew this was poor guidance, but did not know what else to do to help myself.

    Because it was a dumb checklist, not an engaging approach, I knew that it would make the patient feel I wasn’t close to getting it, maybe wouldn’t even be capable of getting it, and even… that I didn’t care. It would sound like they were just a practice object. A dummy. “Why reveal anything to this person?”

    These days I find myself in the position of independent medical examiner and note, over and again, evidence of non-searching, non-specific history-taking which conforms to a pattern recognition approach, which is also means “jumping to conclusions”. And why is that? Because we are taught implicitly to hurry rather than engage.

    The thing is, the obvious 80% of stuff will get worked out without problems, even when clumsily explored. So it seems to work. But the truth is that an engaged bright school student might well be able to intuit many of these diagnoses. No craft needed.

    And it’s the 20% (certainly 10%) with erroneous or incomplete sets of diagnoses that get unnecessary tests, inappropriate treatments, missed time windows, just plain old unnecessary delays, and poorer outcomes leading to much greater costs. I’m actually under-emphasising that.

    Now I know to ask, when talking to a male suspected of drinking too much:

    “Do you get though more than three cartons a week?” He can then feel good about contradicting my wild over-estimate with the correct amount. Or that he prefers spirits. Or, just assent. (….Tailor your own preferred over-estimate.)

    “Do you get help with that?” (because maybe the quantity is shared). and, “What about on a Big Night?” etc

    Not, “How much alcohol do you drink?” which, for the problem drinker, tends to specifically trigger a pang of discomfort, leading to minimisation / underquoting.

    “Have you ever had any neck pain or injury?” *sounds* thorough and inescapable, but the wording is too inclusive, experienced as:
    “How would you like to go through all your past neck pain memory records and précis them for me?”, and so the obligate short answer is: ”No” (“Not really”).

    As neck pain is very common, I can reasonably ask: “When you get neck pain, how do you handle it? What works for you?”

    Et cetera.

  4. Anonymous says:

    I have regularly seen nursing and other staff go into patients in the ED and “ask” them “You don’t have any chest pain, do you?” I groan and smile with exasperation at the same time at their lack of insight.
    And then often when I talk to the patient, who has come in with a triage description of “chest pain,” and I ask them “Do you have any discomfort at all in the chest?” I get an answer like “Well, not pain but a tight feeling…”
    Several times I have had junior doctors asking teenage girls about sexual activity preface it by saying “I need to ask you some personal questions. Do you mind if your parents stay, or shall I ask them to leave?” And they have absolutely no idea of what they have just done. It has sometimes taken ten minutes of discussion afterwards for the doctor to understand, and some never get it.

  5. Anonymous says:

    The patient gave an accurate answer to the question: “any problem?” but the Dr was later alarmed. The question could have been about past tracings as we know some anomalies are not really a problem: “Have you ever had an ECG or Rhythm tracing, and if so why and was it OK?” Then there would have been no surprise, and the Dr may have been more prepared as we know that not all diagnoses are faultless.

  6. Anonymous says:

    I am sorry for your loss, Sarah.

Leave a Reply

Your email address will not be published. Required fields are marked *