Suicide Risk in Clinical Practice, and Why OCD Demands Extra Vigilance
By Felix Murad, M.Ed., LPC-S, Licensed Professional Counselor-Supervisor (Texas & Washington)
A sober teaching piece on risk assessment, the Columbia, and safety planning, and the OCD blind spot most training never names.
Here is a sentence no one says at orientation, so I’ll say it: if you stay in this work long enough, you may lose someone to suicide. Surveys of the field put it bluntly: a large share of clinicians experience a client death by suicide at some point in their careers. And when it happens, everyone is stunned. The family. The team. The supervisor. The new Associate most of all, who replays every session looking for the thing they should have seen.
Some of that shock is unavoidable grief. But a meaningful part of it traces to two preventable failures: treating risk assessment as a form to clear rather than a clinical act, and carrying specific blind spots into the room, OCD chief among them. This piece is about closing both. Not because a tool guarantees a save. Because doing this like it matters is the actual standard, and it’s how you give your clients, and yourself, the best odds.
A note before we start: this is clinical education. If you or someone you’re with is in crisis, the 988 Suicide and Crisis Lifeline (call or text 988) is available 24/7.
You will sit with serious risk. Prediction is not the job.
Let’s set the standard correctly, because a wrong standard is what produces the crushing, paralyzing guilt after a loss.
No clinician and no instrument can reliably predict who will die by suicide. The research on risk prediction is humbling; categorizing patients into neat low/medium/high tiers and expecting those tiers to forecast outcomes does not hold up well. That sounds discouraging. It’s the opposite; it relocates your responsibility to where it belongs. Your job is not omniscience. Your job is competent, ongoing assessment, appropriate intervention, and honest documentation. The standard of care is a process, not a guaranteed outcome. A clinician can do everything right and still lose a client, and that is a tragedy rather than a failure. Internalize that now, before you need it.
Risk assessment is not a checkbox, and the checkbox mentality is dangerous
You asked the right question implicitly when you flagged this topic: people treat risk assessment as something done “for fun”: to satisfy the chart, to cover the agency. That mentality gets people killed, and it shows up in two classic forms.
The first is the perfunctory screen: the rushed “you’re not having any thoughts of hurting yourself, right?” delivered in a tone that tells the client the only acceptable answer is no. The second is the no-suicide contract: the “promise me you won’t” agreement that still floats around clinical culture. Be clear on this: no-suicide contracts have no evidence of preventing suicide, do not protect you legally, and can create a false sense of security that crowds out real intervention. They ask for a passive promise; they give the client nothing to do.
And while we’re killing myths: asking directly about suicide does not plant the idea or increase risk. The evidence is consistent on this, and direct inquiry is essential to safety (Dazzi et al., 2014; Posner et al., 2011). Vague questioning isn’t gentleness. It’s avoidance, usually of your discomfort, and it leaves the most important information unspoken.
The Columbia (C-SSRS): the instrument, used well
The Columbia-Suicide Severity Rating Scale is the closest thing the field has to a criterion standard, which is why it’s endorsed by the FDA, CDC, and WHO and used worldwide (Posner et al., 2011). Its power isn’t mystical; it’s that it imposes structure on a conversation clinicians otherwise tend to rush or skirt.
A few things to understand to use it well:
- It’s a clinician-administered, semi-structured interview, not a self-report checklist. You ask, you listen, you follow up.
- It maps the full spectrum. Ideation severity runs an ordinal ladder from a passive wish to be dead, through active ideation, up to active ideation with plan and intent. It separately assesses the intensity of ideation (frequency, duration, controllability) and behavior, including the subthreshold acts people miss: aborted and interrupted attempts and preparatory behavior, which are common and predictive.
- It is scored by the highest endorsed item, not a sum. There is no magic total. A “yes” to active ideation with plan and intent, or any recent preparatory behavior, demands immediate clinical attention regardless of how the rest reads.
- It tracks change over time, which makes it a monitoring tool across a high-risk period, not a one-and-done.
But hear this clearly: the C-SSRS structures the assessment; it does not replace your judgment. The tool surfaces data. You still have to formulate.
From screening to formulation
A score is not a plan. After you’ve screened, you build a risk formulation: you synthesize the picture into a coherent clinical judgment about acute versus chronic risk, weighing static risk factors (history of attempts, diagnosis, demographics), dynamic warning signs (recent loss, acute hopelessness, intoxication, a sudden ominous calm), protective factors (connectedness, reasons for living, engagement in care), and access to lethal means.
The shift that marks a maturing clinician is this: you stop assessing risk to predict and start assessing it to intervene. The question isn’t “what tier is this person?” It’s “what’s driving the risk right now, and what changes it tonight?” Then you document your reasoning: what you assessed, what you concluded, and why. Good documentation isn’t defensive paperwork; it’s the visible trace of having actually thought.
Especially OCD: the blind spot that costs lives
Now the part most risk training never teaches, and the reason you flagged this.
OCD carries a genuinely elevated, and historically under-recognized, suicide risk. Meta-analysis finds a moderate-to-high association between OCD and suicidality (Angelakis et al., 2015); in OCD samples roughly a third report current suicidal ideation, about half report lifetime ideation, and at least one in ten have attempted (Pellegrini et al., 2020). A large Swedish cohort found OCD to be an independent risk factor for death by suicide even after adjusting for comorbidity, at a level comparable to schizophrenia and bipolar disorder (Fernández de la Cruz et al., 2017). Risk concentrates where obsessions are severe, where the “unacceptable/taboo thoughts” dimension is prominent, and where depression is comorbid. The old belief that OCD is a low-suicide-risk condition is simply wrong, and it kills people by making clinicians complacent.
Here is the differential that defines competent OCD care, and it cuts in two directions:
Suicidal obsessions are not suicidal ideation. A substantial number of people with OCD experience intrusive suicidal thoughts that are ego-dystonic: unwanted, horrifying, and senseless to them. These live in the same family as harm, sexual, and blasphemous obsessions: the person is afraid of the thought, not drawn to it. They don’t want to die; they’re terrified by the intrusive question of whether they might. Typically you’ll see the OCD signature around it: distress, avoidance, mental checking (“do I really want to?”), and reassurance-seeking.
Error one: mistaking the obsession for intent. React reflexively: hospitalize, alarm, treat the thought as a danger to be neutralized, and you can traumatize the client, rupture the alliance, and reinforce the OCD, because you’ve just confirmed the thought is as dangerous as the disorder insists. You’ve performed a compulsion on the client’s behalf.
Error two: mistaking real ideation for an obsession. This is the deadlier one. Because OCD patients carry real, elevated suicide risk and high rates of comorbid depression, “it’s probably just an obsession” is a lethal assumption. A person can have suicidal obsessions and genuine suicidal ideation at the same time. The OCD does not immunize them.
So how do you tell them apart? You assess the function and quality of the thoughts. Genuine ideation carries some degree of desire, intent, or plan, however ambivalent; the suicidal obsession is feared, resisted, and ego-dystonic. You use the C-SSRS, but you interpret it through the OCD lens rather than reading items mechanically. And here’s the advanced tension: you must do real assessment without getting pulled into the reassurance loop, because “tell me I’m not really suicidal” can itself become a compulsion you’d be feeding. The skill is to assess thoroughly, hold both possibilities, refuse to either over-react or under-react, and, when you’re genuinely unsure, consult rather than guess. Treatment follows the formulation: a suicidal obsession, once genuine risk is carefully addressed, is treated as OCD (with ERP, not reassurance); genuine ideation is treated as genuine risk.
Safety planning that actually works
When risk is real, you don’t reach for a contract. You build a Stanley-Brown Safety Plan, a collaborative, written, prioritized list the client can actually use when a crisis escalates. The evidence is strong: in a study of 1,640 ED patients, safety planning plus structured follow-up was associated with roughly 45% fewer suicidal behaviors over six months and more than double the odds of engaging in outpatient care, compared with usual treatment (Stanley et al., 2018). It’s recognized as a best practice by the Joint Commission, and Texas HHSC offers free training in it.
The six steps move from self-reliance toward escalating support: (1) recognizing personal warning signs; (2) internal coping strategies the client can do alone; (3) people and settings that provide distraction; (4) people to ask for help; (5) professionals and agencies to contact, including 988; and (6) making the environment safer: lethal means counseling.
That last step is the one clinicians flinch from and the one that most directly saves lives. Means matter because lethality matters: firearms are involved in roughly half of US suicide deaths and are by far the most lethal method (CDC). Reducing access during a high-risk period (securing firearms outside the home, locking up medications) buys the thing suicidal crises are often missing: time. You do this collaboratively, often with a trusted family member, framed as temporary safety rather than permanent restriction.
One OCD-specific caution: a safety plan can itself be colonized by the disorder, becoming a ritual the client reviews compulsively for reassurance. Build it, then watch that it stays a tool rather than another compulsion.
When you do lose someone
Because you named it: sometimes, despite competent care, a client dies. The grief is real and particular; it braids ordinary mourning with self-doubt and, often, fear of blame. Know three things in advance. A client’s suicide is not, by itself, evidence that you were negligent; competent clinicians lose people. You should not be alone with it. Bring it to supervision and consultation immediately, lean on your own therapist, and seek out clinician-survivor support, which exists precisely because this is common and devastating. And document and follow your agency’s process while you do. Carrying it in silence is the worst option, and the one too many Associates choose out of shame.
If you’re reading this in the aftermath right now: please reach out to your supervisor or your own clinician today, and 988 is there for you too.
Frequently asked questions
If a client dies by suicide, am I liable? Liability turns on whether you met the standard of care: competent assessment, reasonable intervention, and documentation, not on the outcome itself. A death is not automatically negligence. This is general information, not legal advice; if you’re facing a specific situation, consult an attorney and your liability carrier.
Do no-suicide contracts protect me or my client? No. They lack evidence of preventing suicide and offer no real legal protection. Collaborative safety planning is the standard-of-care replacement (Stanley et al., 2018).
Won’t asking directly about suicide make it more likely? No. That’s a persistent myth. Direct, specific questioning does not increase risk and is essential to assessment (Dazzi et al., 2014).
How often should I reassess? Risk is dynamic, so assessment is ongoing, not a single intake event, especially across transitions, after a stressor, and whenever the clinical picture shifts. The C-SSRS is built to track exactly this kind of change.
Before your next high-risk session, sit with these
- When you ask about suicide, are you asking to find out, or asking in a way that’s quietly hoping for a “no” so you can move on? Your client can tell the difference.
- With an OCD client reporting suicidal thoughts, can you hold both possibilities, obsession and genuine ideation, without collapsing prematurely into either? If not, that’s a consultation, not a guess.
- If you lost a client tomorrow, would your chart show that you thought, or just that you checked a box?
Don’t carry acute risk alone
This is the one area of practice where solo improvisation is indefensible. Managing serious suicide risk, and surviving the loss of a client if it comes, is not something an Associate should do without support. If you’re holding a high-risk case and second-guessing the call, or you’re in the aftermath of a loss, bring it to consultation. That isn’t weakness; it’s the standard of care, and it’s how clinicians stay whole enough to keep doing this work.
For the diagnostic reasoning underneath the OCD section, see differential diagnosis and why the nuance changes treatment.
Crisis resources: 988 Suicide & Crisis Lifeline: call or text 988 (24/7). Crisis Text Line: text HOME to 741741.
Felix Murad, M.Ed., LPC-S, LMHC, CMHC, NCC, Licensed Professional Counselor-Supervisor. Licensed in Texas, Washington, New Hampshire, and Florida (telehealth). This article is professional education for clinicians and is not clinical, medical, or legal advice, nor a substitute for individualized assessment, supervision, or emergency care. Suicide is a sensitive subject; if this material is affecting you personally, please reach out to your own clinician or to 988.
References
Angelakis, I., Gooding, P., Tarrier, N., & Panagioti, M. (2015). Suicidality in obsessive compulsive disorder (OCD): A systematic review and meta-analysis. Clinical Psychology Review, 39, 1–15.
Dazzi, T., Gribble, R., Wessely, S., & Fear, N. T. (2014). Does asking about suicide and related behaviours induce suicidal ideation? What is the evidence? Psychological Medicine, 44(16), 3361–3363.
Fernández de la Cruz, L., Rydell, M., Runeson, B., Brander, G., Rück, C., D’Onofrio, B. M., Larsson, H., Lichtenstein, P., & Mataix-Cols, D. (2017). Suicide in obsessive-compulsive disorder: A population-based study of 36,788 Swedish patients. Molecular Psychiatry, 22(11), 1626–1632.
Pellegrini, L., Maietti, E., Rucci, P., Casadei, G., Maina, G., Fineberg, N. A., & Albert, U. (2020). Suicide attempts and suicidal ideation in patients with obsessive-compulsive disorder: A systematic review and meta-analysis. Journal of Affective Disorders, 276, 1001–1021.
Posner, K., Brown, G. K., Stanley, B., Brent, D. A., Yershova, K. V., Oquendo, M. A., Currier, G. W., Melvin, G. A., Greenhill, L., Shen, S., & Mann, J. J. (2011). The Columbia-Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults. American Journal of Psychiatry, 168(12), 1266–1277.
Stanley, B., & Brown, G. K. (2012). Safety planning intervention: A brief intervention to mitigate suicide risk. Cognitive and Behavioral Practice, 19(2), 256–264.
Stanley, B., Brown, G. K., Brenner, L. A., Galfalvy, H. C., Currier, G. W., Knox, K. L., Chaudhury, S. R., Bush, A. L., & Green, K. L. (2018). Comparison of the safety planning intervention with follow-up vs usual care of suicidal patients treated in the emergency department. JAMA Psychiatry, 75(9), 894–900.