When a client says they are not safe: the first thirty seconds

The fear most early-career clinicians name is not the escalating client. It is the disclosure.

And when you ask what specifically they are afraid of, it is usually not the procedure. Most people know the procedure. What they are afraid of is the thirty seconds after the sentence lands, when the room changes and their own alarm goes off, and they have to keep having a conversation that now requires more care rather than less.

What the alarm does

Your own activation is not a professional failing. It is what happens when someone tells you they may not be safe, and it happens to experienced clinicians too.

What it does is narrow. Under alarm, the question collapses to whether this person is safe and how quickly that can be established. Which is a necessary question and an insufficient one, because it produces a particular kind of interaction: rapid, closed, information-gathering, with the clinician working through a mental list.

You can conduct a technically adequate risk assessment that way and lose the person while doing it. The client experiences being processed at the moment they took the largest risk of the relationship, and the disclosure does not fully arrive because the conversation has become about the checklist.

Slowing down is the counterintuitive move

The instinct under alarm is to speed up. The thing that helps is usually the opposite.

Not slowing down the assessment itself, which needs to happen and needs to happen properly. Slowing your own pace: your speech, the gaps between your questions, the transition from what they just said to what you ask next.

Two reasons. The first is that clients track your state closely, and a clinician visibly moving into emergency mode communicates that what they said was unmanageable. The second is that your own thinking is better at a slower tempo, and you need it.

A short acknowledgment before the first question does a lot of work here. Something that marks that you heard the thing and are staying in the room with it, before you begin gathering.

Asking directly

The most common error at this point is indirectness, and it comes from good intentions.

Clinicians soften the question because they do not want to be blunt about something so serious, or because there is a lingering belief that asking directly plants an idea. It does not, and the softened version has a real cost: it signals that this is a topic requiring careful handling, which makes it harder for the client to answer plainly.

Asking directly, in ordinary language, communicates that this can be spoken about. Which is often the more useful thing you do in that moment.

What sits outside this article

Everything above is about staying steady enough to have the conversation. It is not risk assessment, and it does not replace it.

What follows the conversation depends on your scope, your supervision, your service’s procedures and your legal obligations, and those differ by discipline, setting and jurisdiction. Any account that skipped over that would be doing you a disservice.

Why practice matters more here than anywhere else

The reason this moment is disproportionately hard is that almost nobody has practised it before it was real.

You have read about it. You have possibly discussed it in supervision. But the first time you say the words out loud to a person in distress is usually the first time it counts, with your heart rate doing what it is going to do, and no prior experience of having got through it.

That is a strange feature of how we train clinicians. We rehearse a great many things and generally not this one, despite it being the moment people report as most frightening. Having said it aloud before, even in a scripted exercise with a colleague, changes what is available to you when it happens for real. It is also one of the things supervision is structurally poorly placed to build, because it happens afterward rather than during.

Where to go with it

Risk is not a separate process from the rest of what happens in a difficult session. It can arise within escalation, collapse, attachment protest or dissociation, and the state a client is in affects how the conversation goes.

The free clinical reference is a one-page tool covering four processes that look similar in the room and need different responses. Several practice principals have found it useful as a shared reference and a starting point. The February cohort of When Clients Escalate includes practising a risk conversation in a small group before you have to have one unprepared.

Applied Brain Co provides clinician-facing education and professional development. It does not replace case-specific supervision, organisational procedures or emergency protocols.