Ketamine at Home: IV, IM, Subq, or Sublingual? A Physician's Honest Take

If you have been researching ketamine therapy for depression, anxiety, PTSD, or chronic pain, you have probably typed some variation of "ketamine at home" into Google. That phrase covers a lot of ground, and the route of administration is the single most important variable that determines what kind of clinical setup a legitimate practice can actually offer.

I am Dr. Ben Soffer, board-certified in internal medicine, licensed in Florida, New Jersey, and California. I prescribe ketamine via telehealth, and I have strong opinions about which routes belong at home and which do not. This post is the explainer I wish more clinics would write.

TL;DR

• Sublingual ketamine (troches or rapid-dissolve tablets). Yes, and this is the default I prescribe for nearly every patient through at-home ketamine therapy. Gentle onset, widest safety margin, lowest burden.
• Rectal ketamine. An alternative I use when taste, nausea, or absorption make sublingual hard to tolerate. Similar absorption profile, same gentle curve.
• Compounded intranasal ketamine. A selective option I do prescribe in specific cases where a faster onset is clinically preferred. Not the same as Spravato, and not my default, for the pacing reasons below.
• Subcutaneous (subq) ketamine. Evidence-backed, and offered case-by-case at my discretion only for carefully screened patients I judge appropriate to self-inject. Not a marketed, self-serve service.
• Topical ketamine. A different job entirely: localized nerve pain, not mood. No dissociative session.
• IV ketamine. No. Belongs in a clinical setting with continuous monitoring.
• IM ketamine. I do not prescribe it for home use, and the reasons are mostly about the parenteral workflow, not the timing.

The rest of this post is the why behind each of those answers, in plain language.

Why the route matters

Ketamine is a dissociative anesthetic. At sub-anesthetic doses, the kind used for depression and anxiety, it produces a brief altered state of consciousness lasting roughly 45 to 90 minutes depending on the route, with sensory shifts, time distortion, and a softening of the usual ego boundaries that many patients describe as therapeutically useful.

Two things make the route matter:

1. The onset and peak of effect. IV ketamine peaks within minutes. IM peaks at roughly 5 to 15 minutes. Subq is somewhere in between. Intranasal peaks at 10 to 20 minutes. Sublingual climbs slowly over 20 to 40 minutes and peaks at roughly an hour.

For sublingual, that slower curve means a patient takes the dose, settles their environment, sends a "starting now" message to their sitter, and is lying down well before any meaningful effect appears. The format quietly enforces a preparation window.

2. Dosing precision and bioavailability. Sublingual ketamine has a bioavailability of roughly 25 to 30 percent, with patient-to-patient variability. IM is much higher. Intranasal sits between the two depending on formulation. High bioavailability sounds like a feature, until you realize the small absorption buffer in sublingual administration is exactly what gives the route its margin. Small dosing errors sublingually produce small systemic effects. The same proportional error on a parenteral route lands the full magnification straight into the bloodstream.

This is why sublingual is the format engineered for safe at-home use, and the others are not. The pharmacokinetics are not a side note. They are the whole reason routes get assigned to settings.

IV ketamine at home: why I do not prescribe it

Intravenous ketamine has the strongest evidence base for treatment-resistant depression. It is also the most monitoring-intensive route.

In a properly run clinic, IV sessions involve continuous blood pressure monitoring (ketamine reliably raises BP and heart rate), pulse oximetry, an IV line that stays in for the duration, a nurse or physician in the room or immediately available, and resuscitation equipment within reach. Patients are screened for cardiovascular risk before the first infusion.

None of that is possible in a patient's home. No patient can safely self-cannulate, monitor their own blood pressure during dissociation, or respond appropriately to a hypertensive episode while in the middle of a ketamine experience.

A small number of telehealth practices in some jurisdictions have offered IV at home with a visiting nurse. Even in that model, which is rare and expensive, the safety floor is far higher than what any unaccompanied home setup can replicate. For self-administered home use without a clinician in the room, IV is not on the table.

IM ketamine at home: why I do not prescribe it

Intramuscular ketamine is a single injection into a muscle, typically the deltoid or the lateral thigh. It is used in clinic for procedural sedation and increasingly for psychiatric ketamine sessions because it is faster to set up than an IV. The onset gives the patient a similar preparation window to sublingual, so the "you would be impaired before you finished setting up" concern that applies to IV does not apply here. That timing argument is not why I avoid it.

The real reasons against IM at home are about the parenteral workflow and the operational reality of keeping a controlled-substance vial in a household.

First, the injection itself is not trivial. Ketamine for injection has an acidic formulation, which makes IM injection genuinely uncomfortable. Clinics often buffer it or use slow-push technique to reduce the burn. A self-injecting patient on their first session at home is being asked to overcome a real "I do not want to do this" reflex, into a muscle they have likely never injected before. Without training, technique errors can produce bruising, irritation near nerves at certain sites, or accidental injection into a vessel rather than muscle. The clinical staff who would normally handle this are not in the room.

Second, the needle handling and household-safety problem. Each session means working with a vial of a DEA Schedule III controlled substance, an appropriate needle, a sharps container for safe disposal, and a secure storage plan for the remaining supply between sessions. None of that is hard in a clinical workflow. At home it is a meaningful operational burden, with real risks. Needlestick injuries to the patient or a family member. Kids or pets finding a vial or sharps container. Partners or housemates with substance-use history finding the supply. Sublingual troches in foil packaging sidestep every part of this. They are easier to lock up, easier to track, harder to misuse, and have no sharps footprint.

Third, the dosing-precision problem. Parenteral ketamine for psychiatric use is dosed by weight, drawn from vials of varying concentration. The volume math is straightforward in a pharmacy or clinic setting and easy to fumble at home, especially on the first few sessions. Pre-measured sublingual troches eliminate the math entirely. A proportional dosing error orally is usually recoverable. The same proportional error IM hits the bloodstream within minutes and offers little opportunity to course-correct.

If you are being offered IM ketamine at home by a practice, ask them: who reviewed your cardiovascular screen, who is reachable during the session, and what is their protocol if something goes wrong. The answers will tell you whether you are working with a real clinic or a shipping operation with a medical license rented for cover.

Subcutaneous ketamine at home

Subcutaneous ketamine has a long history in palliative care and chronic pain, sometimes as a continuous infusion via a small pump for cancer pain that other approaches have not controlled. But it has also become the most heavily marketed at-home psychiatric route, and the honest picture is more nuanced than a flat yes or no.

The efficacy evidence for the route is real. A randomized, double-blind, active-controlled trial of repeated subcutaneous ketamine for treatment-resistant depression (Loo et al. 2023, British Journal of Psychiatry, the KADS study) found meaningful benefit, so I will not pretend the route is fringe or disparage the programs built around it. Subcutaneous absorption is high, approaching intravenous levels, with a fast onset.

What I will not do is offer it as a self-serve product. The tradeoffs are physical and they are honest: self-injection asks a patient to perform an injection correctly as dissociation begins; the peak is steeper than sublingual, so the experience arrives harder and with less warning; there is a sterility burden tablets do not have; and once an injected dose is in, there is no way to slow it or take it back. Those are the same concerns from the IM section, and they do not disappear because the needle is smaller.

So my position is case-by-case. For a carefully screened patient I judge appropriate and trustworthy to self-administer this way, subcutaneous can be a reasonable option, and I will consider it individually. For most people, sublingual delivers a comparable outcome with a far gentler safety profile, which is why it remains my default. This is a decision I make with a specific patient for a specific reason, not a box you check at signup.

The separate palliative context is worth knowing about: hospice patients sometimes receive subq ketamine through a home infusion service with a home health nurse and a controlled medication pump. That is a different clinical setup from outpatient psychiatric care.

Compounded intranasal ketamine: a selective option, not my default

Intranasal ketamine prepared by a compounding pharmacy is a real format that I do prescribe in selected cases. It is different from Spravato (the FDA-approved esketamine nasal spray), which is restricted under a federal program to in-clinic administration only. A compounded intranasal racemic spray can be prescribed for supervised home use, and reviews of intranasal ketamine for depression describe a genuine antidepressant signal (An et al. 2021, Frontiers in Psychology). The nasal lining absorbs quickly, so onset is faster than sublingual, which for a specific patient can be exactly what is wanted.

It is a selective choice rather than my default, for two reasons worth understanding.

First, the pacing. A nasal spray takes seconds to administer and effects peak in 10 to 20 minutes. Each spray delivers a small dose. The combination of fast onset, easy administration, and the absence of a natural pause between doses can create a use pattern that, for a Schedule III controlled substance used for a mood disorder, needs more deliberate structuring than the self-pacing sublingual route. The troche takes 5 to 10 minutes to dissolve, the onset is gradual, and there is no natural impulse to take "one more" mid-session because the patient is already in the experience. When I prescribe nasal, that pacing is something I structure deliberately with the patient and the supply.

Second, for most psychiatric use it does not solve a problem sublingual cannot. Patients respond to both routes at comparable rates in the literature. Sublingual is cheaper to compound, easier to ship, and the gentler onset suits most depression and anxiety indications. So nasal earns its place when a faster onset is genuinely preferable for a particular patient, not as a blanket upgrade.

The clearest case for the faster route is chronic pain, particularly breakthrough pain. Some chronic-pain protocols use intranasal ketamine precisely because the rapid 10 to 20 minute onset is the point: a patient with a known neuropathic or cancer-pain pattern who experiences a sudden flare benefits from a route that works quickly. In that supervised setting, with a limited supply, intranasal is often the right tool.

Sublingual ketamine: why this is what I prescribe

Sublingual ketamine, also called buccal ketamine when the troche is held against the cheek, is what I prescribe to nearly every patient I see for at-home ketamine therapy. It is also well supported: a large prospective study of at-home sublingual ketamine delivered by telehealth (Hull et al. 2022, Journal of Affective Disorders) found it safe and effective for moderate to severe anxiety and depression across more than a thousand patients.

Patients receive a compounded troche or rapid-dissolve tablet from a DEA-registered compounding pharmacy. They take a single dose, holding it under the tongue or against the cheek for several minutes, then either swallow the saliva or spit (different protocols vary). Onset is gradual, 20 to 40 minutes to peak. The effect lasts roughly 60 to 90 minutes.

The reasons this format works at home:

• No needles. No self-injection, no sharps disposal, no chain-of-custody questions about parenteral controlled substances.
• Slow onset. Patient is fully oriented at the time of administration. Theysend a "starting now" message to their peer support, get into position, dim the lights, and the experience begins after they are already settled.
• Lower peak effect for a given dose. The slower absorption curve produces a gentler peak than IM or IV at equivalent total absorbed dose. Patients can still reach therapeutic dissociation, just on a smoother trajectory.
• Recoverable dosing errors. If a patient absorbs less than the prescribed dose because they swallowed early, the experience is gentler than intended, not dangerously stronger.
• Format-enforced dose pacing. The dissolution time prevents the take-another-dose pattern that nasal formats invite. The route quietly does some of the harm-reduction work the prescriber would otherwise have to do with patient education alone.
• Practical for repeat sessions. Patients can run a 4 to 6 session course over several weeks without needing a clinic visit each time.
• Troches split cleanly for fine titration. Unlike rapid-dissolve tablets, a compounded troche cuts into halves or quarters with a knife, so a dose can be adjusted down without waiting on a new prescription. Rapid-dissolve tablets break apart once they meet saliva and do not split reliably.

There is also a peer support requirement, what I call a "sitter," that is non-negotiable for every patient. The sitter is a sober adult who agrees to be physically present for the session, monitor the patient, and call for help if something goes wrong. This is the human-monitoring layer that replaces the clinical-monitoring layer you would get in a ketamine clinic.

The combination, sublingual route, peer support present, physician available by phone, screening done in advance, is the safety architecture that makes at-home ketamine medicine and not a shipping operation.

Rectal ketamine: a sublingual-equivalent alternative

Rectal ketamine, given as a suppository or a rectal solution, has bioavailability similar to sublingual, around 25 to 30 percent, with the same gradual onset. I reach for it when a patient cannot comfortably use the sublingual route.

Some people find ketamine's taste genuinely intolerable, and holding a bitter troche under the tongue becomes a barrier to consistent treatment. Others have nausea or upper-gastrointestinal issues that make sublingual absorption unreliable for them. The rectal route delivers a comparable amount of medication on a similar gentle curve while sidestepping the taste and some of those issues. It is less familiar to patients and takes a frank conversation to set up, but for the right person it preserves the gentleness that makes sublingual safe at home, minus the parts they could not tolerate. Same clinical goal, different rung, matched to the patient.

Topical ketamine: for pain, not mood

Topical ketamine, a cream or gel often compounded with other pain agents, belongs in this guide precisely so you do not confuse it with the routes above. It is used for localized neuropathic or chronic pain, and it is designed for minimal systemic absorption, so it does not produce the dissociative, antidepressant-type effect the other routes are prescribed for.

That low absorption is the feature. A topical acts on the nerves in the skin and tissue right where it is applied, calming localized pain signaling, while keeping blood levels low. There is no dosing "session," no sitter, no set-aside afternoon. The evidence base for ketamine in nerve pain is its own literature (Guimarães Pereira et al. 2022, Journal of Pain Research; Israel et al. 2021, Health Psychology Research), separate from the mood research behind the other routes. If you came here about depression or PTSD, topical is not your route. If you have localized nerve pain, it may be relevant, and it deserves an honest place on the list.

Red flags when evaluating at-home ketamine practices

A short checklist for patients evaluating any at-home ketamine offering:

• Does the practice require a peer support or sitter to be present? Anyone who waves this requirement is cutting the human-monitoring layer that makes at-home ketamine safe. Walk away.
• Did you have a real video consultation with a licensed physician before being prescribed? Not a form, not a chatbot screen. A physician who looked at your medical history and asked questions. This is the legal floor for telehealth ketamine.
• Is the route IV, IM, subq, or intranasal? If yes, the practice should have a clear answer about monitoring, dosing precision, and emergency protocols. If they are shipping parenteral or nasal kits with self-administration instructions and no on-call physician, that is the operation I would avoid.
• Is the pharmacy DEA-registered and named? You should receive a prescription from a known compounding pharmacy with a verifiable license. Generic "we ship from a pharmacy partner" without a name is concerning.
• Is the practice licensed in your state? Ketamine is prescribed under the prescriber's state license. If you are in Florida and the prescribing physician is licensed only in California, the telehealth setup is on shaky legal ground.

---

The honest summary. The at-home ketamine telehealth space is genuinely useful medicine for the right patients in the right format, and a regulatory gray zone in the wrong formats. Sublingual ketamine, physician-prescribed, with peer support present and proper screening, is where the safe-and-effective center lives. Everything else either belongs in a clinic or should not exist as an at-home option.

If you are considering ketamine therapy and you want to start with the route I actually trust, the eligibility assessment is the front door.

References

1. Hull TD, et al. At-home, sublingual ketamine telehealth is a safe and effective treatment for moderate to severe anxiety and depression. Journal of Affective Disorders. 2022. PMID: 35809678
2. Loo C, et al. Efficacy and safety of a 4-week course of repeated subcutaneous ketamine injections for treatment-resistant depression (KADS study): randomised double-blind active-controlled trial. British Journal of Psychiatry. 2023. PMID: 38108319
3. An D, et al. Intranasal Ketamine for Depression in Adults: A Systematic Review and Meta-Analysis of Randomized, Double-Blind, Placebo-Controlled Trials. Frontiers in Psychology. 2021. PMID: 34140915
4. Guimarães Pereira JE, et al. Efficacy and Safety of Ketamine in the Treatment of Neuropathic Pain: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Journal of Pain Research. 2022. PMID: 35431578
5. Israel JE, et al. Ketamine for the Treatment of Chronic Pain: A Comprehensive Review. Health Psychology Research. 2021. PMID: 34746491