Ketamine Guidelines: Two Expert Panels, Opposite Conclusions
An international panel of 21 academic psychiatrists spent two Delphi rounds on 105 questions about ketamine and esketamine. They agreed on 35 of them. The statement was published November 2 in the Journal of Clinical Psychiatry. It puts the consensus rate at 33 percent.
Five months earlier another expert panel covered much of the same clinical ground. That one agreed on 97 percent of its items. Two of the same people sat on both panels.
Most of what follows costs nothing to fix. The premium is the part you negotiate. HealingMaps’ GPO, the first built for the ketamine industry, gets member clinics 15 to 20 percent off malpractice and liability premiums, 40 percent or more off medical supplies, discounted LegitScript certification, and 20 to 30 percent off HIPAA compliant payment processing. Joining is free with no obligation. See what your clinic qualifies for →
| Key Takeaway | Detail |
|---|---|
| The split | The ASCP panel agreed on 35 of 105 items. The ASKP3 panel agreed on 73 of 75 |
| Why they differ | Different scope and a different question set, not rival camps |
| The overlap | Sandhya Prashad and Rupert McShane served on both panels |
| The evidence base | 61 of the 73 practitioner recommendations are graded expert opinion |
| Still unsettled | Ketamine assisted psychotherapy, high dose infusions, novel formulations, use beyond depression |
| The staffing question | The practitioner panel could not agree that psychiatric training is unnecessary |
| The regulatory gap | Esketamine runs under a REMS program. Racemic ketamine has no equivalent |
Two panels, two question sets, two answers
The November statement comes from a task force of the American Society of Clinical Psychopharmacology. Gerard Sanacora of Yale led it.
The author list reads like a roster of the field. It includes Roger McIntyre, Sanjay Mathew, James Murrough, Samuel Wilkinson, Carlos Zarate and Allan Young.
The panel set its bar in advance. Consensus meant 75 percent agreement or disagreement on five point Likert items. Thirty five items cleared it.
The abstract groups those items into broad domains. They include diagnostic conditions, contraindications and maximum dosing. Others cover racemic versus enantiomeric preparations, safety monitoring, access and how to judge outcome.
The other panel published in May. The American Society of Ketamine Physicians, Psychotherapists and Practitioners convened an interdisciplinary committee. It ran its own modified Delphi.
Of 39 invited experts, 28 completed at least one round. Consensus was set at a median rating of 7 or higher on a nine point scale. The interquartile range had to be 2 or less. The result was agreement on 73 of 75 items.
The gap is not a fight. It is a difference in what each panel asked.
ASKP3 scoped itself tightly to intravenous racemic ketamine for depression in outpatient settings. Its committee drafted the items from an evidence review first. Only then did the drafts go to the panel.
The ASCP panel asked open questions across the whole practice space. That included the parts nobody has studied.
This is why the same two clinicians could sit on both. The field agrees about intravenous ketamine for depression. It disagrees about most of what has been built on top of that.
Only 12 of 73 recommendations rest on controlled data
The ASKP3 guidelines are open access, and every item carries an Oxford evidence grade. We counted them. The tally is lopsided.
| Oxford evidence level | Consensus items |
|---|---|
| Level 1: meta-analyses and systematic reviews of randomized trials | 10 |
| Level 2: randomized controlled trials | 1 |
| Level 3: non-randomized controlled cohorts | 1 |
| Level 4: case series or case control studies | 0 |
| Level 5: expert opinion, mechanism based reasoning or existing guidelines | 61 |
Sixty one of the 73 items are Level 5. Twelve rest on any controlled data at all.
The 97 percent consensus rate is real. It is near unanimous agreement about what experienced clinicians believe.
The pattern inside the document matters more than the total. All five items in the opening evidence section carry Level 1. Those are the questions about whether the drug works, and the trials answer them.
The Level 5 grades concentrate elsewhere. They cluster in the sections on who may provide treatment, patient evaluation, consent and documentation.
So the evidence is strongest on the pharmacology and thinnest on the clinic. That is the ordinary condition of a young specialty. It is also why a clinic cannot outsource its standard of care to a guideline and stop there.
The two items the practitioner panel could not pass
Two of the 75 items failed and were removed after the second round. Both deserve an operator’s attention. Both still drew 73 percent agreement.
The first proposed that ketamine treatment for depression should not be restricted to medical professionals with post-graduate training in psychiatry. It did not pass.
Panelist feedback recorded in the paper shows why. Some experts regard post-graduate psychiatric training as the appropriate standard of care. Another argued that clinicians offering several mental health options present a more balanced view of the alternatives.
Read that against the item that did pass. The panel reached very strong consensus on a parallel question. It would not restrict treatment to clinicians with anesthesia or critical care training.
So the panel was comfortable ruling out one gatekeeping standard. It could not rule out the other.
The second failed item concerned dosing in obesity. It proposed calculating starting doses from adjusted or ideal body weight rather than actual weight. One panelist preferred actual body weight. Another wanted a maximum dose instead.
Neither failure is an exotic edge case. Staffing model and starting dose are decisions every infusion clinic makes before it treats anyone.
They are also the decisions that surface in a claim. We mapped that in our review of where malpractice exposure concentrates.
What the practitioner panel did rule out
The items that passed are not all permissive. Several close off practices that are currently sold.
The panel does not recommend additive or proprietary ketamine formulations for depression. It suggests starting ketamine naive patients at 0.5 mg/kg over roughly 40 minutes. The induction series runs four to six infusions across two to four weeks.
It also found insufficient published evidence that dosing three or more times weekly beats twice weekly during induction.
On dose escalation the language is pointed. Titrating above 0.5 mg/kg may be reasonable in patients who do not respond. Above 1.0 mg/kg the panel asks for clinical justification in the record.
One item goes further. Evidence that the subjective qualities of a ketamine experience predict antidepressant response is called inconclusive.
The panel therefore counsels caution about escalating dose to reproduce an intensity of experience. Measurable clinical benefit is the standard it sets instead. That addresses a model some clinics market directly.
It also lines up with the ASCP panel’s open questions. High dose infusions and ketamine assisted psychotherapy are two of the four areas of persisting uncertainty.
The others are novel formulations such as oral or intramuscular delivery, and use in conditions other than major depression.
One molecule, two regulatory regimes
The ASKP3 authors make a structural point that explains much of the rest. Esketamine is governed by a Risk Evaluation and Mitigation Strategy. The FDA required it as a condition of approval.
It restricts the drug to approved doses, indications and settings. It mandates certified clinicians and at least two hours of post-dose monitoring.
No analogous program exists for racemic ketamine. The authors call this, to their knowledge, unique. No other pair of single enantiomers of one molecule is regulated so differently.
That gives racemic ketamine flexibility. It also leaves the standard of care to be written by the people using it.
The same open questions, nine years apart
Sanacora also led the 2017 consensus statement on ketamine for mood disorders, published through the American Psychiatric Association. That document anchored the 0.5 mg/kg over 40 minutes protocol.
It warned that long-term data were sparse. It cautioned strongly against take-home self-administration. It also said consent should cover how limited the off-label evidence was.
Its authors were explicit that it was not a standard or a guideline. Nine years later, the same lead author reports that agreement remains elusive for many key decision points.
The 2026 practitioner guidelines still call long-term outcomes a meaningful limitation of the evidence base.
What the split means for operators
The practical reading is not that the field is in disarray. The settled core is simply narrower than marketing language suggests. For everything outside it, the burden of proof sits with the clinic.
Three things follow. Any service in one of the four unsettled areas needs a documented clinical rationale, because no guideline supplies one.
Second, consent language should reflect the actual evidence grade of the treatment being offered. Third, a clinic generating its own outcome data is not only measuring quality. It is building the record guidelines currently lack.
That last point is why we treat outcomes data as a competitive asset rather than an administrative chore.
Credentialing is the companion question. The training and credentialing landscape remains unsettled for the same reason.
The author lists are worth one last note. The ASKP3 panel was drawn largely from operating clinics. Eight of the organizations represented on it appear in the HealingMaps directory, across 41 listings.
The people writing the standards are running the practices.
Patients comparing supervised options can start with our directory of verified ketamine clinics.
This article is for informational purposes and is not legal, medical or clinical advice. Consult your own counsel and clinical leadership about your practice.
