Working With Suicidality in EMDR: Holding Safety and Reprocessing at Once

Read Time: 9 Minutes

The client you were half-afraid to reprocess with.

Every EMDR clinician meets the moment eventually. A client discloses that they have been thinking about ending their life, and the session you had planned gives way to a different and more urgent set of questions. The instinct that follows is almost universal. You slow down, you stabilize, and you set the trauma work aside until the risk has passed. That instinct is not wrong, but it is worth examining, because the way we resolve the tension between safety and reprocessing shapes whether a client ever reaches the material driving their despair.

For years the prevailing assumption was that active suicidality placed trauma processing off limits. The clients who most needed their histories addressed were often the ones held indefinitely in stabilization instead. The difficulty with that approach is circular. The unprocessed material is frequently what sustains the hopelessness, and waiting for a stability that the material itself prevents can mean waiting a very long time.

The cost of waiting

There is a quieter risk that rarely gets named, which is the risk of doing too little. Indefinite stabilization can look prudent while it slowly communicates to a client that their history is too dangerous to touch, a message that can deepen the very hopelessness it was meant to contain. The emerging evidence complicates the reflexive caution as well. Work that treats trauma and depressive symptoms in people who are actively suicidal has found reductions in suicidal ideation rather than the destabilization long feared, which suggests that withholding the work is not the neutral, safe default it can appear to be (Van Bentum et al., 2024). The point is not that every high-risk client should be reprocessing this week. It is that the decision to wait is itself a clinical intervention with its own consequences, and it deserves the same scrutiny we give the decision to proceed.

What the AIP model suggests about suicidal states

Read through the lens of adaptive information processing, suicidal ideation looks less like a fixed trait and more like a state, one that rises and falls with activation and is often bound to specific stored experiences (Shapiro, 2018). Entrapment, defeat, humiliation, and the conviction that nothing will ever change tend to sit close to the surface of a suicidal moment, and those beliefs usually have a history. When we treat the ideation only as a risk variable to be contained, we can lose sight of the fact that it is also, in AIP terms, a symptom pointing back toward material that has never been metabolized. That reframe does not lower the stakes of risk. It widens the clinical picture so that risk management and case conceptualization can happen in the same breath. Seen this way, a spike in suicidal thinking after a hard week becomes data about what remains unresolved, not only a number to be logged.

Safety and reprocessing are not opposites

The most useful shift is to stop treating safety and reprocessing as a choice between two doors. Structured risk work belongs in the treatment whether or not you are actively reprocessing, and it runs alongside the trauma work rather than in place of it. A collaboratively built safety plan gives the client something concrete to reach for in an acute moment (Stanley & Brown, 2012), and ongoing collaborative risk assessment keeps the clinician honest about a picture that shifts week to week (Jobes, 2016). None of that requires the trauma work to stop. It builds the container that makes the trauma work defensible.

The more clinically relevant picture is not suicidal ideation erupting in the middle of reprocessing. That is rare. It is the client who carries mild, intermittent ideation between sessions, a passive expression of the despair that unresolved trauma leaves behind. The client denies a plan, means, and clear intent, and does not meet criteria for a higher level of care. The ideation is not a separate condition to be cleared before the trauma work can begin. It is one of the ways the untreated trauma speaks, and the EMDRIA literature has grown steadily clearer that its presence is not a categorical rule-out but a clinical reality to be assessed, prepared for, and worked with deliberately (Spector & Kremer, 2009).

One clinician recalls a client who named this more precisely than any scale could. The client said, "Part of me just doesn't want to be here. But I'm so exhausted living with this that I don't have the energy to end my life." She was carrying the aftermath of a sexual assault, living with flashbacks and nightmares and the conviction that it had somehow been her fault. On assessment her suicidal ideation was mild, and it read as a manifestation of that unresolved trauma rather than a discrete acute risk. The clinician talked through the pros and cons of beginning EMDR with her, carefully and collaboratively, and together they built a concrete plan to check in between sessions as the work began.

In one session the client processed the most intrusive part of the trauma, and she arrived, on her own, at the recognition the shame had kept out of reach. She said it aloud in the room: "It's not my fault. Oh my God, I see it. It's not my fault." In the wake of that session the shame lifted, and the intrusive symptoms and the suicidal ideation lifted with it. One case is not a promise, and this pace is not the norm, since much of this work is slower and far less linear. But it shows the mechanism the AIP model would predict. When the ideation is downstream of unresolved trauma, resolving the trauma can resolve the ideation, and the distressing material that surfaces along the way is frequently the process doing exactly what it is meant to do (Shapiro, 2018).

It helps to hold a clear line between self-harm and suicidality, since the two are often conflated and call for different responses. Non-suicidal self-injury frequently functions as a way of regulating unbearable affect rather than an attempt to die, and reading it accurately changes both the risk picture and the target you choose. Risk is also dynamic. It moves within a single session and across a course of treatment, which is why one intake assessment cannot carry the weight, and why the clinician's attention to fluctuating state has to stay live throughout the work.

Where the clinical judgment actually lives

If there is a fulcrum in this work, it is preparation. The quality of Phase 2, the client's access to affect regulation, the resources you have helped install, and your read of their window of tolerance (Siegel, 1999) determine far more about a session's safety than any rule about who is or is not a candidate. The harder judgments are rarely binary. They are questions of sequence and pacing. Which target belongs at the center of this session and which belongs later. When a rising charge signals productive movement and when it signals that a client is leaving their window. Whether the work this week should extend the client's capacity or consolidate it. These are the calls that separate careful EMDR from merely cautious EMDR, and they do not reduce to a flowchart.

When the future is the target

One of the more striking developments in this area concerns suicidal intrusions, the vivid mental images of one's own death that some clients experience as involuntary and intolerable. Rather than treating those images only as risk indicators, clinicians have begun targeting them directly through the flashforward approach, which turns the anticipated catastrophe itself into the object of reprocessing (Logie & De Jongh, 2014). A multicenter randomized trial found that a dual task add-on aimed at these suicidal flashforwards reduced both the frequency and the intensity of the intrusions in patients with depressive symptoms (Van Bentum et al., 2024). The evidence in this specific application is still developing, and it will not fit every presentation, but it illustrates the larger point. Suicidality can be something we work on, not only something we work around.

The calls you should not have to make alone

What makes this work demanding is not a shortage of protocols. It is the density of judgment it requires, made in real time, usually with incomplete information, and carrying a weight that few other clinical situations match. Deciding whether to reprocess or to stabilize with a high-risk client is precisely the kind of decision that benefits from more than one clinical mind, yet is also the type of decision many of us end up making alone. Ongoing consultation as a therapist, and especially as an EMDR therapist, not only ensures you check your blind spots, but it is a means for navigating clinical complexity with skill and nimbleness. The clinicians who last in this work tend to be the ones who treat consultation as a standing feature of their practice rather than something they reach for only in a crisis.

References

_{Jobes, D. A. (2016). Managing suicidal risk: A collaborative approach (2nd ed.). Guilford Press.}

_{Logie, R., & De Jongh, A. (2014). The "flashforward procedure": Confronting the catastrophe. Journal of EMDR Practice and Research, 8(1), 25–32.}

_{Shapiro, F. (2018). Eye movement desensitization and reprocessing (EMDR) therapy: Basic principles, protocols, and procedures (3rd ed.). Guilford Press.}

_{Siegel, D. J. (1999). The developing mind: How relationships and the brain interact to shape who we are. Guilford Press.}

_{Spector, J., & Kremer, S. (2009). Clinical Q&A: Can I use EMDR with clients who report suicidal ideation? Journal of EMDR Practice and Research, 3(2), 107–108. https://doi.org/10.1891/1933-3196.3.2.107}

_{Stanley, B., & Brown, G. K. (2012). Safety planning intervention: A brief intervention to mitigate suicide risk. Cognitive and Behavioral Practice, 19(2), 256–264. https://doi.org/10.1016/j.cbpra.2011.01.001}

_{Van Bentum, J. S., Sijbrandij, M., Kerkhof, A. J. F. M., Holmes, E. A., Arntz, A., et al. (2024). Reducing intrusive suicidal mental images in patients with depressive symptoms through a dual task add-on module: Results of a multicenter randomized clinical trial. Journal of Consulting and Clinical Psychology, 92(11), 756–768.}