23. May 2026

When Asking Creates Risk: A Reflection on Suicide-Related Internet Use in Assessment

During a recent suicide prevention training session, a participant posed a question that has stayed with me:

“If we ask people about suicide-related internet use, could we actually be introducing the idea—especially if they hadn’t thought about it before?”

It is a question that sits in quiet tension with current guidance—and one that warrants careful, disciplined reflection.

NICE guideline NG225 (Self-harm: assessment, management and preventing recurrence) is explicit about the scope and intent of psychosocial assessment. It requires exploration of “the functions of self-harm… the person’s needs… and factors that may increase or reduce the risk of further self-harm,” alongside “social, psychological and environmental factors… including access to means.” In contemporary practice, those environmental factors necessarily include the digital landscape in which people seek information, connection, and, at times, methods.

The guideline is equally clear on method. Practitioners are advised to use “open, non-judgemental questions” and to “build a trusting, supportive relationship” that enables disclosure of distressing thoughts. Assessment, in this framing, is not a procedural task but a therapeutic intervention in its own right.

This position is reinforced by NHSE suicide prevention guidance, which highlights the need to understand exposure to harmful online content and digital environments that may reinforce suicidal thinking. The NCISH has further strengthened this by identifying suicide-related internet use as an emerging factor in suicide deaths—present in a minority (~8%), but often associated with method access, planning, and cognitive reinforcement.

Empirical data adds precision to this picture. A large-scale study of patients presenting to hospital following self-harm in England found that approximately 8% of adults had engaged in suicide-related internet use, rising to around 26% in younger people, with such use associated with higher suicidal intent (Mars et al., 2015; Daine et al., 2013). These findings suggest that internet use is not a background variable, but a potential marker of escalating risk and behavioural organisation.

Qualitative research further indicates that internet use evolves alongside risk. In earlier stages, it may be incidental or exploratory; as intent strengthens, it often becomes more deliberate, focused, and action-oriented (Daine et al., 2013). This distinction is clinically significant. It positions internet use not merely as a static risk factor, but as part of a trajectory toward action.

Sector guidance aligns with this interpretation. The Samaritans’ practitioner guidance on internet use and suicide emphasises that online environments can both mitigate and exacerbate distress, and recommends that practitioners actively explore how individuals are engaging with digital content. This includes identifying exposure to material that may encourage, maintain, or normalise suicidal behaviour, as well as opportunities for support and connection (Samaritans, Talking about suicide and self-harm: internet use).

Taken together, the direction across NG225, NHSE, NCISH, and wider research is coherent:

Exploring suicide-related internet use is a legitimate and, at times, essential component of comprehensive assessment.

And yet, the participant’s question introduces a necessary counterweight.

NG225 is unequivocal in advising practitioners to avoid “sharing detailed information about methods… that could increase risk.” The mechanism—suggestion, modelling, and increased capability—is well established.

The question, then, is whether asking about suicide-related internet use might operate as a more subtle variant of the same mechanism. Not by providing methods directly, but by signalling that such searches exist, framing them as a pathway, or introducing a behavioural option that had not previously been considered.

The current evidence base offers partial reassurance. There is no indication that asking about suicide introduces suicidal ideation. However, this concern sits adjacent to that evidence. It relates not to the emergence of thoughts, but to the potential shaping of behavioural pathways—particularly those linked to access, planning, and progression toward action.

On this point, the evidence remains limited.

What is clear is that suicide-related internet use tends to cluster with higher levels of intent and organisation. What is less clear is whether enquiry itself can influence its emergence. This is precisely the space where clinical judgement must carry the weight that evidence does not yet fully resolve.

NG225 provides the framework for this judgement. It does not advocate for indiscriminate or checklist-driven questioning. Instead, it calls for individualised, context-sensitive assessment, guided by formulation rather than routine.

Within that framework, asking about internet use is best understood not as a universal requirement, but as a clinically indicated line of enquiry—most relevant where there is already evidence of suicidal thinking, escalation, or movement toward action.

Equally critical is how the question is framed. A direct enquiry such as, “Have you been searching online for ways to end your life?” introduces specificity that may be unnecessary and, in some contexts, unhelpful. A more open formulation—“When things feel this difficult, people sometimes use the internet in different ways—information, distraction, or support. Has that been part of your experience?”—is more consistent with NG225’s emphasis on neutrality, collaboration, and understanding the individual’s perspective.

This is not a matter of style. It is a matter of risk calibration.

It also reflects a broader principle that is often implicit in guidance but critical in practice:

Assessment is not neutral. It is an active component of care.

The questions we ask shape the space we are working in. They influence how distress is understood, what options are perceived as available, and how individuals make sense of their own experience. The SPARK programme reinforces this by highlighting the importance of understanding the behaviours and influences that underpin suicide risk, rather than focusing solely on outcomes.

The participant’s question, therefore, does not challenge current guidance. It refines its application. It draws attention to the fact that good practice is not defined solely by inclusion—by whether a topic is “covered”—but by precision in timing, framing, and intent.

For practitioners, the implication is clear. Exploring suicide-related internet use is often necessary, particularly where risk is evident or evolving. However, it should not be approached as a routine or mechanistic enquiry. It requires deliberate timing, careful phrasing, and an awareness of its potential to both reveal and shape risk.

In this sense, NG225 does not ask us to ask more questions. It asks us to ask better ones.

And at times, the most clinically responsible decision is not simply knowing what to ask—but recognising when even the right question needs to be asked with particular care.