He Said He Was Fine, and I Wrote It Down

I want to describe a note I wrote about nine years ago, because I still think about it and because I do not think what went wrong was unusual.

The note said, in effect: patient denies low mood, denies suicidal ideation, no active plans, engaging appropriately with the ward team, no indication for ongoing psychiatric input at this time.

Every clause in that note was accurate. I asked those questions and those were the answers. If you audited me against the documentation standard, I would pass.

It was also, as a description of that man's state, close to worthless, and I knew that at the time in a way I chose not to look at directly.

The circumstances of the conversation

Let me describe how it was obtained.

It was a Friday afternoon. He was in a four-bedded bay with the curtain pulled round, which conceals nothing acoustically and which everybody in this profession knows conceals nothing acoustically. The man in the next bed had his television on. There was a drug round happening. I had been asked to see him because a nurse had noticed he was flat, and I had six other referrals and a handover at five.

I sat on the edge of a chair — not all the way back in it, the edge, which is a posture that communicates one thing very clearly — and I went through it.

Have you been feeling low in mood? No.

Any thoughts of harming yourself? No.

Sleeping alright? Not great, but you know, hospital.

Eleven minutes. I wrote it up and I moved on, and I felt the small unease you feel when you have completed a task rather than done a job, and I overrode it, because there were six other referrals.

I asked a man whether he wanted to die, in a room where four strangers could hear him, and I recorded his answer as a finding.

What I should have known

Everything I got wrong there, I could have told you was wrong if you had asked me in the abstract. I had taught it.

Closed questions produce the answer the patient believes is expected. "Have you been feeling low?" invites a no from anyone who has ever been socialised into not making a fuss, which is most people over sixty and most men of any age. "How has your mood been?" is barely better in a shared bay. The only questions that work are open, and they take longer, and they require the person to believe you have time.

And there is a specific failure I want to name, because it is mine and I see it in trainees constantly. I asked about suicidal ideation as a screening item rather than as a conversation. It came seventh in a list. It got the tone of the six questions before it, which was administrative, and so it got an administrative answer.

The disclosure that matters almost never arrives in the first fifteen minutes. It arrives once someone has tested whether you are going to stay. I knew that. I have said it out loud in teaching sessions. On that Friday I had six referrals and I let a form do my thinking.

What happened

He was readmitted some months later in significantly worse condition. I am deliberately not giving details.

I do not know — and this is important, because the tidy version of this story would claim otherwise — whether a better conversation on that Friday would have changed anything. It might not have. Depression in the medically unwell is not cured by one good consultation and I am suspicious of stories where it is.

What I do know is that my note was read by other clinicians in the intervening months, and that "no indication for ongoing psychiatric input" from a named consultant is a heavy document. It closed a door. Someone who might otherwise have re-referred him had a reason not to, and that reason was me, being efficient on a Friday afternoon.

What I changed

I do not ask about self-harm in a shared bay. If I cannot get a private space I write down that I could not, and that the assessment is therefore provisional. That sentence in a note is worth more than any answer I would have obtained.

I sit all the way back in the chair. This sounds like nothing. It is not nothing. People read your body for permission to take up your time, and they read it accurately.

I write what I asked, not only what I was told. "Denies low mood" is a claim about the patient. "Denies low mood in response to direct questioning, in a shared bay, at the end of a long day" is a claim about the evidence, and it lets the next clinician weigh it properly.

I stopped treating "no" as an endpoint. If a nurse who has spent three shifts with someone thinks they are flat, and eleven minutes of my questions say otherwise, the most likely explanation is not that the nurse is wrong.

Why I write this down

The version of this I hear most often from trainees is "I did not have time". I had that thought myself and I want to be precise about why it is not true.

I did not have time to do it properly and also to see six referrals and also to leave at five. That was a real constraint. What I did with that constraint was produce a document that looked like an assessment and functioned as a closed door, which is worse than having produced nothing, because nothing would have left the question open.

If there is one thing I would want a junior colleague to take from this: when you have not been able to do the assessment properly, the most valuable thing you can write is that you have not been able to do the assessment properly.

No patient is identifiable. Age, sex, presenting condition, timeline, setting and outcome details have been changed or omitted, and elements are composited from more than one episode.