Lisa Parker, DNP, APRN, NP Founder, Parker Neuroscience, LLC | Omaha, Nebraska
Most of what gets written about IV ketamine focuses on two questions: does it work, and is it medically safe. Both matter, and the answers are encouraging. A systematic review of 93 clinical trials found that among 3,756 participants who received at least one sub-anesthetic dose of ketamine, four experienced a serious medical adverse event, an incidence of roughly 0.1% of individuals (Hovda et al., Journal of Affective Disorders, 2024).
Those two questions leave out the one my patients actually ask me in consultation. Not “will my blood pressure go up.” They ask, in one form or another: what if it gets hard in there, and what happens if it does?
That deserves a real answer, and until recently there was not much data to answer it with.
New data on how often difficult moments happen
Ember Health, a multi-site IV ketamine practice in New York, recently published a chart review covering 5.5 years of care: 2,114 patients and 30,402 infusions, conducted with research collaborators at Harvard Medical School and Mass General Brigham (Ember Health, July 2026). It is single-organization data rather than a randomized trial, and it should be read that way. It is still the largest look I have seen at the psychological side of infusion safety, and I am grateful they published it.
What they found:
About 5% of infusions, roughly one in twenty, were described by the patient afterward as difficult or challenging. Clinicians observed outward signs of distress in 3.38% of infusions, about one in thirty. Across a full course of treatment, roughly one in five patients had at least one session they experienced as psychologically hard. Events meeting criteria for a serious psychological event occurred in 0.54% of infusions, about one in two hundred. Across more than thirty thousand infusions, a sedating benzodiazepine was administered for psychological distress eight times. Clinicians paused an infusion 26 times, and most of those patients recovered enough to complete treatment. Roughly 90% of the sessions patients described as difficult resolved with verbal reassurance and supportive presence alone, without stopping the medication.
Two things stand out to me.
The first is the gap between 5% and 3.38%. A meaningful share of difficult moments were never visible from the outside. The patient was lying still, breathing evenly, looking peaceful, and going through something hard internally. It did not surface until the reflection period afterward, when someone asked. If we do not ask, we do not know.
The second is the number eight. That figure is not small because the experiences were mild. It is small because skilled human support works. Attentive, trained support is the intervention.
Why this matters in the patients I see
Parker Neuroscience serves a lot of veterans, first responders, and healthcare workers. That is not incidental to this conversation.
In the Ember cohort, 73% of patients carried an anxiety disorder alongside depression, with PTSD, OCD, and substance use disorder also common. My panel looks similar and often carries more. When someone comes to me after fifteen years of holding it together on a fire scene, in a trauma bay, or on deployment, the thing they fear most is not the needle. It is losing control in front of another person.
For patients with trauma histories, entering a dissociative state without preparation and without a trusted person present is not a therapeutic opportunity. Preparation is what turns it into one.
What psychological safety looks like in my practice
I trained in ketamine-assisted psychotherapy through the Integrative Psychiatry Institute, and I spent more than two decades in emergency medicine before that. Those backgrounds converge in the treatment room. Emergency medicine taught me to read a patient’s respiratory pattern, posture, and face. KAP training taught me what to do with supporting the patient throughout their ketamine experience.
Before the infusion. We talk honestly about what the altered state can feel like, including the possibility that some of it will be uncomfortable. We set an intention. We choose music. We agree in advance on comfort measures: eye shades, a warm blanket, the lighting. And we establish consent for therapeutic touch before a single milligram is infused, because a patient in a dissociative state cannot meaningfully consent to being touched. That conversation has to happen while the patient is fully oriented.
I also send patients in with something I want them to carry: you are not a passenger in this. Unpleasant moments can surface, and if one does, you usually have more room to move than you expect. You can shift your attention. You can come back to your intention, to the music, to your breath, to an image or a memory that feels steadier. Knowing that in advance changes how a hard moment gets met. A patient who understands that they can turn toward something else tends to meet discomfort with curiosity instead of alarm, and alarm is what escalates.
I am careful not to overpromise it. The degree of volitional control varies by patient and by dose, and I do not want anyone concluding they failed because they could not steer out of something. So I frame it as an option rather than an obligation: you can move toward a more comfortable thought if you want to, you can stay with whatever is surfacing if it feels important, and either way you are not doing it alone. That same principle extends to how closely you want to be accompanied, which we decide together and which you can change between sessions.
During the infusion. Where I position myself is part of the preparation conversation, not a fixed rule. Some patients want me in the chair beside them the whole time. Others find that being observed makes the experience harder to settle into, and they ask me to monitor from just outside with the door open. Both are legitimate. What does not change is that monitoring is continuous, I am within immediate reach, and the patient knows exactly how to signal for me before the infusion begins. For patients who have spent a career being watched under pressure, being asked how they want to be accompanied is itself part of the treatment.
When distress surfaces, the first intervention is voice. The steering belongs to the patient, not to me, so what I offer is orientation rather than direction: their name, that I am here, that they are safe, that what they are feeling is okay. Then a breath cue if it helps, or a quiet reminder of the intention they set, which gives them something familiar to move toward if they want it. I avoid asking questions during the altered state, because a question requires the patient to organize an answer, and that is genuinely difficult in that headspace. Questions can create the very confusion I am trying to relieve.
If verbal support is not enough, the infusion can be paused and the effects allowed to soften. Pharmacologic sedation sits at the end of that sequence, not the beginning, in part because benzodiazepines may blunt the antidepressant effect we are working toward.
After the infusion. I ask directly. Not “how was that,” which invites a polite answer, but specific questions about whether any part of the session felt frightening or hard. This is the step that catches what the gap between 5% and 3.38% represents. A difficult session that gets named, processed, and integrated is often the one that ends up mattering most in someone’s recovery. A difficult session that goes unspoken becomes a reason the patient does not come back.
Difficult is not the same as failed
I want to be clear with patients and with referring colleagues about this, because I think the confusion drives unnecessary discontinuation.
A hard session is not evidence that ketamine is wrong for you. Many patients who go on to substantial, durable improvement in their depression pass through at least one session that was emotionally rough, and many later describe that session as the meaningful one. The goal is not to sand every difficult moment out of the experience. The goal is to make sure that when one arrives, there is someone with the training to support it, positioned the way that patient needs, so the patient can stay in treatment long enough to get well.
One in five is not a rare event. It is a predictable feature of the work, and it is worth preparing for out loud.
Conversations worth having before the first infusion
If you are a patient starting this treatment, or a provider preparing someone for a referral, these are useful things to talk through in advance. Practices are built differently and there is genuinely more than one good way to answer them. What matters is that the patient knows the answers before they are in an altered state.
Who is responsible for monitoring during the infusion, where will they be, and what preparation do they have in supporting altered states? How is consent for therapeutic touch handled, and when is that conversation held? What is the plan if the experience becomes frightening or overwhelming, and what can the patient do in the moment themselves? Is there dedicated time afterward to talk about what came up? How does preparation change for someone with PTSD or a significant trauma history?
None of these are gotcha questions. They are the questions I would want asked of me, and the ones I try to answer before anyone thinks to ask.
Where this is heading
As ketamine services expand, and they are expanding quickly, the conversation about safety needs to widen beyond pharmacology. The medication is the same everywhere. What varies is the container around it: preparation, environment, the continuous availability of a trained clinician who knows the patient, and what happens in the twenty minutes after the infusion ends.
We still need real research separating out which elements of psychological support do the most work. Intention setting, music, therapeutic touch, degree of proximity, and room design have all been described, but they have not been well isolated. We also do not yet know much about how patient preference interacts with any of it, which is a real gap given how differently people want to be accompanied. Until that evidence exists, the Ember data supports a principle I already practice by: skilled, attentive human support resolves the overwhelming majority of hard moments, and it has to be close enough to respond within seconds.
If you are a clinician with patients who have exhausted conventional options, I am always glad to talk through whether this is a reasonable next step for a specific person, including the cases where I would say no.
Lisa Parker, DNP, APRN, NP is the founder of Parker Neuroscience, LLC in Omaha, Nebraska, an interventional psychiatry practice offering IV ketamine and Spravato (esketamine) for treatment-resistant depression, PTSD, anxiety, and OCD. She is an Army and Air Force Reserve veteran with 24 years of advanced practice clinical experience and completed ketamine-assisted psychotherapy training through the Integrative Psychiatry Institute.
This article is educational and is not medical advice or a substitute for individual clinical evaluation. To discuss a referral or schedule a consultation, call 402-915-1862 or visit parkerneuroscience.com.
Advancing Clarity. Restoring Lives.
Sources
Ember Health. Psychological Safety During IV Ketamine Treatment. July 27, 2026. https://blog.emberhealth.co/posts/psychological-safety-during-iv-ketamine-treatment
Hovda N, Gerrish W, Frizzell W, Shackelford R. A systematic review of the incidence of medical serious adverse events in sub-anesthetic ketamine treatment of psychiatric disorders. Journal of Affective Disorders. 2024;345:262-271. https://pubmed.ncbi.nlm.nih.gov/37875227/
If you are in crisis, call or text 988 (Suicide & Crisis Lifeline) or go to your nearest emergency room.
Parker Neuroscience, LLC · 17940 Welch Plaza Ste 106, Omaha, NE 68135 · (402) 915-1862