What to Say When a Patient Wants an Unnecessary Test or Prescription They Found Online

A patient pulls out their phone in the first few minutes and reads you a diagnosis. They have spent hours on a forum, landed on a full-body scan for intermittent shoulder pain, and they want you to order it. You are already running behind, and the request is not indicated. The reflex is to correct the medicine. The work is to recognize that the test is not the actual subject of the conversation.

What the demand is carrying

When a patient arrives with a self-diagnosis and a specific ask, they are handing you a solution to their fear. The hours online were an attempt to get control over a frightening, uncertain sensation in their own body. They have assembled a story that holds together. I have this symptom, this article explains it, this test will confirm it. That story is the only thing giving them a floor to stand on, and your differential is about to take a probability and put a hole in it.

The system you are both sitting inside hardens the dynamic. The patient has unlimited time for one condition. You have fifteen minutes for three of them and a full waiting room. Their corner of the internet trades in definitive answers, often wrong ones. Yours trades in risk and watchful waiting. So when you say let us see how it looks in six weeks, the patient hears that you are not taking them seriously, and they push harder for the thing they walked in with. The loop is simple and stable. Anxiety produces a demand, the refusal reads as dismissal, the dismissal feeds the anxiety, the demand gets louder.

The three moves that feed the loop

These feel like sound practice right up to the moment they backfire, and most of us were trained into them.

The data dump. It sounds like, the evidence for that scan in this presentation is weak, the guidelines say start with physical therapy. The move tries to win on facts. It fails because the room was never about facts. It is an unspoken negotiation about whose fear and whose expertise gets to lead, and the recitation lands as dismissal while it positions you as the opponent.

The premature reassurance. It sounds like, I hear that you are worried, but I am not concerned this is anything serious. The move tries to soothe the anxiety. It fails because you have invalidated the fear before you acknowledged the effort. The patient does not hear you are safe. They hear your feelings are wrong.

The hard no. It sounds like, I am sorry, I will not be ordering that test, it is not indicated. The move sets a clean boundary, and it ends the immediate debate. It also corrodes the relationship. The patient leaves unheard and goes shopping for someone who will say yes, and they may get the wrong test for the wrong reasons from whoever does.

The shift that changes the room

The aim is to stop competing over the test and join the patient in the work they have already started. They came in having done something active about their fear. Your job is to come alongside as the consultant with the missing dataset, rather than the gate they have to clear. The frame moves from my expertise against your search to the two of us against a confusing symptom.

This is positional, the same kind of turn you would make with a client who is fighting you. You validate the work, the researching, the worry, the wish for an answer, and you hold off on validating the conclusion. From there the conversation stops being a referendum on one scan and opens into the larger and more useful questions. What is the fear underneath. What is the real goal. What are all the routes to getting there. You are telling the patient you see a capable, frightened person who took action, and now you are going to take action with them.

Language that fits the new position

Give these to yourself as illustrations of the position, then say them in your own words. Each one does a specific job.

Validate the process. “It makes sense you landed there. Tell me what you read or saw that made it click.” This honors the research without endorsing the verdict, and it turns an argument into a session where the patient teaches you and lowers their guard in the process.

Define your role as additive. “You have put real time into figuring this out, and that helps me. My job now is to add my clinical experience to what you have already done.” You become a partner rather than a corrector, and the effort gets named out loud.

Align around the feared outcome. “You are worried about [the specific serious thing they fear]. So am I. The thing that matters most is that we do not miss it, and in my experience the way to be sure of that is to start with [your first step]. Can I walk you through why?” You name the fear so the patient knows you take it seriously, then you present your plan as the better route to the goal you now share.

Game out their solution. “Say we run the test and it comes back clean. Where does that leave the pain you feel every morning, and what do we do next?” The hypothetical exposes the limit of the requested answer and moves the talk from yes-or-no on one test to a multi-step plan the two of you build together.

What to listen for in the next visit

Notice whether the patient softened once you stopped pushing. A patient whose demand was fear in disguise eases when the fear gets met. The forum diagnosis loosens, the questions get more open, and they start asking what you think rather than insisting on the scan.

Watch your own pull toward the data dump when they restate the request. That pull is the argument trying to restart. The patient does not need you to be more right. They need to feel that the two of you are working the same problem.

Listen for the moment the goal becomes shared out loud. A line like, so the real thing is making sure it is not the bad outcome, means the frame has moved off the test and onto the fear, which is the only ground where you can actually plan.

When the request is the right one

Sometimes the patient is reading the situation accurately and you are not. The research points at something your differential skipped, and the insistence is a signal about your formulation. The tell is whether the patient keeps pointing, steadily, at the same gap after you have stopped pushing and gotten curious. A frightened patient relaxes. A patient with a real mismatch holds the line. Treat the second one as data and order the test, or revise the plan.

And some demands sit on top of something the brief consultation cannot hold. Health anxiety that drives the same patient back week after week with a new catastrophe, a trauma history that makes any uncertainty unbearable, a conviction that no reassurance can touch, these need their own work before the pattern in the room will move. Most requests are not that. Most are a person who got scared, did the only thing they knew how to do about it, and needs you to meet the fear before you meet the medicine.