You get fifteen minutes with the doctor. Two or three antidepressants have taken the edge off for you, and nothing more. You have read that something else exists. Now you have to say so out loud to a person with a computer open and a waiting room behind you.

That moment is where most of this decision gets made. Our survey of 443 adults asked whose say-so would push them toward ketamine or esketamine care, and the physician they already see was the answer for 74 percent. Ads moved just 2 percent; podcasters moved a mere 1. A friend or relative carried 18 percent; someone from the veteran or first responder world whom people follow online carried 4 percent.

Here is the catch. Spravato, the esketamine spray cleared by the FDA for depression that other drugs have not helped, was a total unknown to 73 percent of those same respondents. Patients wait for doctors to raise it, patients cannot name it, and the conversation sits unopened for years.

The opening line

Do not lead with a brand name. Lead with your history, because your history decides whether any of this applies to you.

"I want to talk about where we go next. I have been through two depression medications, each long enough to judge, and I am still not getting through my days the way I need to. What is left on the table, including things we have never talked about?"

That opening states how many medications you have tried, which is the core of the definition of treatment-resistant depression. It reports function rather than mood, which is harder to wave off. And it asks an open question, which keeps your clinician a clinician rather than a gatekeeper.

If you want to name the specific option

Sometimes the open question produces a shrug and a bump in your current dose. If you want to be specific, be specific carefully.

"I have been reading about esketamine, the nasal spray that won FDA approval for depression that did not lift on other medicines. I do not need a prescription today. What I want to know is whether someone like me is who it was designed for, and if I am not, what makes me a poor fit."

"I do not need a prescription today" is doing real work. It turns a request into a consultation, and most clinicians answer a patient who wants to be evaluated very differently from one who wants a drug by name.

The five questions to actually ask

Write these down first.

  • "Looking at my chart, would I count as treatment-resistant?"
  • "Does anything in my history, my blood pressure, or my other prescriptions make this unsafe or unwise for me?"
  • "If this is not my option, what is? I want the next two steps, not just this one."
  • "Who would you refer me to, and how long is the wait?"
  • "What will my insurance require, and who in this office handles that paperwork?"

That last question earns its place in the data. Insurance coverage made the top two provider considerations for 85 percent of respondents, and for 65 percent it was a deal-maker or deal-breaker, or close to one. For most people, coverage is the conversation. If you want to see what the visits and paperwork look like at a certified clinic before you go in, Brain Recovery Centers explains how its Spravato program is set up.

What to say if you get a flat no

A no can mean different things, and each calls for a different reply. Ask which one it is.

"Help me understand the no. Is it that I do not qualify medically, that you are not comfortable with this treatment, or that nobody nearby offers it?"

If it is the first, ask what would have to change for you to qualify. If it is the second, that is a fair clinical view, and a fair moment to request a psychiatrist for another opinion. If it is the third, that is logistics, and logistics have addresses and phone numbers.

Asking for a psychiatric referral is no insult to your primary care doctor. In our poll, 56 percent planned to begin with the family doctor and 23 percent with a psychiatrist. Moving from one to the other is the system working as designed.

Say it in your own words, not in medical ones

Our open-ended question gathered the literal search words of 319 respondents. Some asked for "ptsd help." Others wanted "help with depression," or typed "how to cope with my anxiety," or asked "how to overcome depression." Not one reached for clinical vocabulary.

That works in an exam room too. "I cannot get out of bed on Sundays and I have started calling in sick" tells a clinician more than any borrowed term. Bring specifics: hours of sleep, how work is going, what you have quit doing, and for how many months.

Bring a witness if you can

Depression is a poor narrator; it shrinks the account of your own week. Someone close to you will remember what you minimize.

That person may know this ground already. Among our respondents, 72 percent said depression, anxiety, or PTSD that routine medicine could not resolve had reached them or someone they love.

One thing to keep straight before you go

Spravato is esketamine, and the FDA signed off on it for depression that has shrugged off standard treatment; each dose is given in a certified office where staff observe you afterward. IV ketamine for mood and ketamine mailed home through a telehealth prescriber are separate arrangements with looser oversight. Say "ketamine" in the appointment and expect a minute of sorting out which one you mean.

This is not medical advice, and no article can judge your case from a distance. The point is narrower: the conversation is worth having, and you are allowed to start it.

Thoughts of suicide are an emergency and should not wait for a booked visit. Anywhere in the United States, a call or a text to 988 reaches the Suicide and Crisis Lifeline, open every hour of every day. Nobody is charged, nothing is shared, and a trained human answers.

Methodology

The publisher commissioned and funded this study. It was a 443-person survey of adults 18 to 64 from ten Midwest and Plains states, run by Pollfish with its consumer panel; responses stopped coming in on June 23, 2026. Every participant passed a consent screen. For check-all-that-apply items we report the share choosing each answer, so those totals exceed 100. Figures reflect final panel validation.