Questions and answers

I spend quite a lot of time in my work observing conversations between doctors and patients, and also the ones that doctors have with each other, generally in the context of educational and clinical supervision. I would say that ‘axe handle’ conversations are extremely common in all these kinds of encounter. I also believe that they are evidence of a wider problem: as doctors we are better at asking questions than in hearing the answers and responding to important cues. We seem to spend a disproportionate amount of time in our own heads, trying to work out what to say next, and too little time trying to get inside our patients' heads and working out what is going on there instead.

I suspect there are all kinds of reasons for this. For example, when we first learn how to take a history at medical school, we do so by memorising a set of questions in a rather one sided way, without considering how we might need to adapt them uniquely on each occasion, according to the answers we receive. Later on in our training we may have some teaching in communication skills, but this is unlikely to include any specific instruction on how to recognise an important cue, let alone how to craft a question that exactly fits the cue from the patient's point of view. As a result, many exchanges between doctors and patients have the kind of quality demonstrated in this classic example from the USA:

Patient: It's one spot right here. It's real sore. But then there's like pains in it. You know how…I don't know what it is.

Doctor: OK…fevers or chills?

Patient: No

Doctor: OK. Have you been sick to your stomach, or anything like that?

Patient: (Sniffles, crying) I don't know what's going on.2

Here, as so often, one has the impression that the conversation is entirely doctor led. When the patient uses phrases like “I don't know what it is” and “I don't know what's going on”, or when she sniffles and cries, the doctor doesn't even recognise these as cues. He responds as if these are technical data only, not important information about her state of mind, emotions or train of thought.

Joining the cue

In an attempt to understand this problem a bit more, I recently carried out an exercise with a group of clinicians who are all experienced in case based discussion and peer supervision, and have all had a fair amount of specialised training in this. I asked them first of all to come up with a list of what they considered to be the characteristics of a good cue in a medical conversation. Initially, they came up with some fairly obvious answers—for example, when the other person keeps repeating something, or says it with a particular emphasis, or with palpable emotion. However, as the discussion progressed something interesting emerged. Working in pairs, and independently from each other, they all came to the conclusion that they recognised a good cue not just by what the other person said or did, but also by noticing their own reaction to it. In other words, they had learnt how to become attuned to what mattered to the patient through paying attention to their own response—for example, in their increased curiosity, anxiety, or sense of tension in their own bodies.

All their ideas seemed to point towards one simple fact. Holding an effective medical conversation depends on a noticing moments of difference, discomfort or puzzlement in oneself. Assuming this description may hold true for all good supervising or consulting, it points towards an important area for training. Learning to spot a good cue in a conversation may not depend on ‘just listening’ but on developing one's own somatic self awareness, internal dialogue and intuition.

The importance of context

Following this exercise, the group watched a video recording of some live supervision between a senior general practitioner and a younger one who was having a problem with her registrar. It showed a conversation that many professionals might regard as quite a competent example of supervision, but the members of this group were mostly critical of it. They noticed how many of the senior trainer's questions were driven not so much by the words and emotions of his colleague but more by his own internal set of preoccupations. These included such things as his wish to know more about the practice where she worked, and also his concern that her registrar might have a serious performance problem. In terms of responsiveness to cues, this audience felt the supervisor was not being very sensitive. In their view, he was showing the same kind of single minded and self absorbed behaviour demonstrated in the consultation example above.

I was struck by their response. I knew enough about the circumstances surrounding the conversation they were watching to believe that the supervisor's questions were in many cases justified, even though it was hard to see this from the content of the conversation alone and without knowing anything about the context. I also knew (from having spoken to both parties after the original conversation) that the two general practitioners had been fully aware of a range of issues that were never explicitly set out in the conversation itself. They had also regarded it as an effective and successful piece of supervision. Setting this knowledge against the rather negative response of this skilled and experienced group of observers was an interesting challenge.

Getting the balance right

What the contrast in views showed, I believe, is that medical conversations generally require two things. One of these is an attentiveness to the cues that matter most to the other person. The other is an appropriate sense of what you need to import into the conversation from other perspectives—training needs, perhaps, or organisational pressures, or issues to do with patient safety and clinical risk. A conversation that only pays attention to the first of these may be a helpful and comforting experience for the patient or trainee who brings a problem for discussion, but it may not protect them from harm, or stop them from getting into trouble. Conversely, conversations where their own verbal and emotional cues are silenced by an oppressive barrage of questions about external matters are unlikely to produce much change in knowledge or understanding. The crucial question for all such conversations may be: when must a cue be recognised and honoured, and when can it be safely ignored?

Finding the right balance may not be easy. My guess is that many counsellors and mentors, for example, might be tempted to follow their clients' cues religiously and with quite a high level of precision, while being relatively inattentive to considerations of what is going on outside the room. Doctors and clinical teachers, by contrast, might be so concerned about worst case scenarios that they might often just plough on with their own litany of prepared and stereotyped enquiries, regardless of what the other person is saying. If that is the case, what we may need to develop in future training is an art or science that covers both aspects of a good conversation: when to be able to pick up a cue, and when to ignore it.