Say a doctor has raised a supervised treatment because the usual medications have not lifted your depression. Now you face a short list of places, or a single one, and must judge it without any of the training that would make judging easy.

The questions that separate a well-run program from a thin one work wherever you live.

Begin with the preference you probably already hold. In June 2026 we asked 443 adults spread across ten Midwest states where ketamine or esketamine therapy should take place. A clinic visit in person came first, with 44 percent. Video care at home drew 22 percent and a route that opens at the clinic and continues at home 23 percent, leaving 11 percent with no view. The results are final.

Step one: settle what is being offered

Before judging a provider, pin down the treatment. Spravato's esketamine spray holds an FDA approval limited to depression that resisted earlier treatment, and each dose happens in a certified office followed by a period of observation. IV ketamine at a clinic is the generic drug used off label, still under direct supervision. Ketamine prescribed by a telehealth service for home use is also off label, with oversight that varies by company.

Ask which one you are being offered, in those words. An answer that arrives as a slogan rather than a specific is your first data point about the operation. For a sense of what a specific answer sounds like, Brain Recovery Centers sets out how it delivers Spravato in plain language.

Step two: run the coverage question before anything else

Coverage is the filter nearly everyone applies first, spoken or not. For 85 percent of the people in our survey, insurance ranked among their two biggest provider considerations, well clear of every other factor.

Three calls settle it. Ask your plan whether this treatment is a covered benefit, whether it needs prior authorization, and whether it bills medically or through pharmacy. Ask the provider if your specific plan has it in network, not merely whether it "accepts" that plan; those are different claims. Then ask who in the office prepares prior authorizations and how long they usually take.

Uninsured or between plans? Ask the total cost per session and how many sessions the first phase involves. A per-session price without a course length is not a price.

Our survey's payer mix is a reality check on any clinic's posture. With every applicable payer counted, commercial coverage reached 39 percent and Medicaid 37, while TRICARE sat at 5, the uninsured at 9, and Medicare at 23; 2 percent declined to say. A program limited to commercial insurance excludes much of the population around it.

Step three: judge the clinical setup

Ask who is in the building: whether a physician or psychiatric prescriber is on site during treatment hours, who watches patients during observation, and what training that person has. Ask the protocol if someone becomes distressed mid-session, and which medical emergencies they are equipped to handle.

Ask what happens before the first dose. A serious program screens medical and psychiatric history, current medications, blood pressure, and substance use, and records a baseline depression score so improvement is tracked, not assumed. And ask how they decide the treatment is not working.

Step four: price the logistics honestly

A supervised session runs longer than its slot because of monitoring, and you cannot drive afterward. Get the door-to-door duration and the first month's frequency, then lay it over your real week, including your driver and whatever you must tell an employer.

This is where proximity earns its place: it was the second-ranked consideration in our survey, at 43 percent, behind coverage. Distance you can absorb once becomes the reason people quit in week three, so be realistic rather than optimistic about the drive.

Step five: check the fit that is specific to you

• Privacy: 11 percent of respondents put discretion in their top two. If your job or community makes this sensitive, ask about parking, the waiting area, appointment timing, and how the practice contacts you.
• Service experience: 10 percent ranked a veteran or first responder specialty highly. If it applies to you, ask how many patients like you the practice has treated and whether it takes TRICARE and works with your own doctor.
• Scheduling: ask when slots are open, not just that they exist. An evening or early morning opening can decide whether a course gets finished.
• Continuity: ask whether you will see the same clinician, and who answers questions between visits.

If you do not know where to start at all

That was the answer of 5 percent in our survey. If that is you, the well-worn path is simplest: 56 percent of respondents would start with primary care, and 23 percent with a psychiatrist or therapist.

Bring a written medication history: each antidepressant, rough dates, duration, and the reason it stopped. That single page moves your care further than weeks of research, because every prescriber and insurer will ask for it.

Warning signs worth walking away from

Be wary of any provider who promises results, quotes a glowing effectiveness number with no source, pushes a prepaid package before an evaluation, will not name the prescribing clinician, treats screening as a formality, or cannot explain how its treatment relates to FDA approval. None of those are close calls.

What this is and is not

This guide helps you evaluate a provider. It cannot say whether a treatment is right for you. Treat it as background rather than medical advice, and expect no promise of improvement. These medications carry real requirements and risks; a clinician familiar with your history and other conditions decides suitability.

If the search itself has worn you down to where you are unsure you want to keep going, tell someone today. Text or call 988, anytime, for the Lifeline. It is confidential and free, and reaching out while you are still deciding whether to reach out is exactly the right moment.

Methodology

The study was ours: this publisher commissioned it and covered the cost. Pollfish surveyed 443 adults aged 18 to 64 through its consumer panel before the June 23, 2026 close, and respondents lived in ten Midwest states, from Iowa and Nebraska to Indiana and Ohio. Each cleared a consent screener first. Items allowing several answers are reported per respondent, which is why those shares add past 100. All values come after the panel finished validating the sample.