Ask people why they have not raised the subject with someone they are worried about, and the same answer comes back: what if I put the idea into their head?

It is the single largest obstacle to the conversation that matters most, and it rests on a belief that can be tested. It has been tested, repeatedly, and the belief is wrong.

What the evidence shows

The question was reviewed systematically in Psychological Medicine in 2014 by Dazzi, Gribble, Wessely and Fear, who examined the published research on whether asking about suicide and related behaviour induces suicidal ideation, in both general and at-risk populations.

They found no study in which asking produced a statistically significant increase in suicidal ideation. Several studies pointed the other way. In one of them, 57% of participants reported improved scores after being asked, and described the experience as useful; those whose scores worsened mostly attributed it to being made to focus on difficulties they had been avoiding, and expected the effect to be temporary. In a separate study of suicide-bereaved and non-bereaved parents, 94% found taking part valuable, and two people out of 1,043 expected any negative effect to last.

The authors concluded that the ethical caution around asking could reasonably be relaxed, because the harm it was designed to prevent had not been demonstrated.

Why the fear persists anyway

The fear assumes that suicidal thinking arrives from outside, as a suggestion a person might catch. In practice, by the time someone is at risk the thought is usually already present, often for months, and often unspoken because they cannot imagine anyone receiving it well.

A direct question does not introduce the idea. It removes the isolation around an idea that is already there. Those are different things, and it is the second one that helps.

There is also a logical problem with the fear. If asking about suicide could cause it, every clinical assessment, every crisis line call and every screening questionnaire would be dangerous. They are not. Asking about suicidal thoughts is among the most studied interactions in mental health, and the finding has been consistent for decades.

Where conversations do go wrong

When a conversation about suicide goes badly, it is rarely because the subject was raised. It is usually because of how it was raised.

  • Leading with reassurance. “You would never do anything silly, would you?” tells the person which answer is acceptable. It closes the conversation while appearing to open it.
  • Using a euphemism. “You are not thinking of doing anything stupid?” carries a judgement in the adjective and is vague enough to be answered with a shrug.
  • Treating the answer as an emergency that has to be resolved at once. You do not need a plan for someone else's life. You need to stay in the conversation and then help them reach someone who can act.
  • Promising secrecy. If someone asks you not to tell anyone, agreeing to that promise may be the most dangerous thing you do. It is better to say plainly that you will not keep it if you believe they are in danger.

The version that works is shorter and plainer than most people expect: Are you thinking about suicide? Then, having asked it, be quiet and let the answer arrive.

What follows

Asking does not create the risk. Not asking does not remove it, it only removes your ability to help with it. The concern that mentioning suicide might make things worse is, on the evidence, the belief most likely to leave someone alone with something they should not have to carry alone.

The practical version of this, including the exact words and what to do with the answer, is in what to say to someone who is struggling. There is also a printable poster built around this single point.