There’s a hidden placebo response to watch for when patients stop meds on their own.

Publication Date: 06/23/2025

Duration: 13 minutes, 43 seconds

Transcript:

CHRIS AIKEN: Last week, we covered a hidden placebo effect that can amplify responses to any med that is started on the first visit. Today, we’re going to flip that on its head. I call it the “placebo flip,” and the anti-psychiatrists are going to hate me for this one. It’s when patients paradoxically get better after stopping a medication.

KELLIE NEWSOME: Welcome to the Carlat Psychiatry Podcast, keeping psychiatry honest since 2003. 

CHRIS AIKEN: I’m Chris Aiken, the editor-in-chief of the Carlat Report.

KELLIE NEWSOME: And I’m Kellie Newsome, a psychiatric NP and a dedicated reader of every issue. Imagine you have a chronic illness like bipolar, schizophrenia, or recurrent depression. You’ve been to the doctor for a few episodes, but each time, you managed to convince yourself that it was just a temporary reaction to stress. After the fourth episode, though, you start to accept something a bit more dreary – this is your life. You’re not like the other people. You have a chronic mental illness, and you’re going to have to stay on a medication to remain well. It’s a tough pill to swallow, but to make things worse, that feeling of “wellness” never really comes. Yes, you are no longer in the hospital or in bed all day, and you can hold a job – though not at your educational level. But your mind is slowed, you get anxious for no reason, and you are so fatigued sometimes that it’s an effort to move. It’s been like this for a few years, and your doctor says you are in recovery, but you start to wonder if your medication is causing some of these problems. Your doctor is nice but a bit too optimistic and maybe isn’t telling you the whole story. So you stop your med. The first day, a little withdrawal. No big deal – you can handle it. After 5 days, that goes away, and surprise, you aren’t falling apart. You can make it on your own like everybody else. That tale you’d been telling yourself – that there is something fundamentally wrong with your brain – is not true. You didn’t need that med. Imagine what else you can do! And you do. With this newfound confidence, you reconnect with old friends. You put in an application for a promotion. Life has finally opened its doors. Three months later, you start to feel off. Your wife starts to complain that you are detached, that you act like she’s not there. Disagreements turn to arguments, and soon, you’re sleeping in separate beds. That’s when you notice the depression coming back, and you call up your doctor. 

CHRIS AIKEN: When that patient comes in, he’s not likely to ask to go back on the antidepressant he stopped three months ago. “That med did nothing for me – I felt better after stopping it – I’m only depressed now because my marriage is on the rocks.” Maybe the patient is right, and maybe it was just causing side effects all along, but keep in mind another possibility: The placebo flip. Just as patients feel better when they finally have a helpful clinician to lean on at the first visit, they can also feel better when they realize they can do it on their own when they stop a preventative medication. Most of the time, the new episode doesn’t come on right away – it takes a few months – and often stress tips the balance. That stress may have been caused by early signs of the episode – like the avoidance and disconnection with his wife in Kellie’s case – or it may have just come on. Most of my patients go through some kind of stress a couples times a year, as do most people. So there is always stress to point the finger at.

Let’s pause for a preview of the CME quiz for this episode. Earn CME for each episode through the link in the show notes.  

1. Which antidepressant augmentation strategy reliably prevents recurrent, unipolar major depression in controlled trials? 

A. Lithium 

B. Second-generation antipsychotics 

C. Lamotrigine 

D. Tricyclic antidepressants 

KELLIE NEWSOME: You need to be a little skeptical when patients say they felt better off their med. Especially if you carefully reviewed side effects beforehand and didn’t find any, and especially if they relapse within a few months of stopping it. One medication where I see this a lot is Silexan – lavender oil extract – which is used as a prescription for anxiety in Europe but available over the counter in the US. Unlike most supplements, this one has a large effect size – around 0.8 in generalized anxiety – much more effective than paroxetine in a large head-to-head trial. A lot of our readers have started recommending Silexan after reading about it in The Carlat Report, and if you are using it, here are two tips. 

• First, the most effective dose is 160 mg, which is two capsules. A lot of patients make the mistake of only taking one. You can give it all at night or space it out twice a day. 
• Second, you’ll see a lot of “placebo flips” on this one because patients are more likely to stop over-the-counter treatments than the ones we prescribe. Silexan takes about a month to work, but it might take longer for the benefits to wear off, as I often see people come in with higher anxiety about 3 months after stopping it. When I talk about going back on, they are skeptical, saying they felt just fine after coming off. So, I’ll show them their own anxiety rating scales, which usually show a pattern of improvement on it, then worsening off. 

CHRIS AIKEN: There are two things to know in this talk about the placebo flip. One is – psychologically - patients may have a boost of confidence and better mood when they are able to stop a medication. The other is that relapse is usually delayed. For someone with bipolar disorder, there’s a 95% that they will relapse after stopping a mood stabilizer, that 95% figure is over the next 5 years…they don’t relapse right away. In bipolar disorder, the average relapse happens 6-10months after coming off meds. The longer they were well before stopping, and the slower they tapered off the med, the more delayed the relapse. 

KELLIE NEWSOME: But people can go into relapse fast if they stop lithium all of the sudden, right? 

CHRIS AIKEN: Yes, but their “rapid relapse” means about 3-4 months after stopping lithium. If people come off lithium slowly, over a few months, the time to relapse is around 10 months after stopping it. Patients do not consider 3 months to be very rapid.   

KELLIE NEWSOME: What about with antidepressants? 

CHRIS AIKEN: After stopping an antidepressant, about a third will relapse in 6 months, and half will relapse after a year, so again, we’re talking about delays that are so long patients are unlikely to connect the new episode to stopping the med. I also see the same patterns when patients stop psychotherapy, exercise, or any lifestyle change that improve mood. They don’t feel worse right away – if they did, they’d never stop those things! Instead, it’s several months later that they come in more depressed, and by that time, they have trouble believe it was because they stopped going to the gym 4 months ago. 

KELLIE NEWSOME: The bottom line. When your patient stops a med, there’s a long delay before relapse – about 3-12 months. By that time, they are unlikely to attribute it to coming off the med, and instead, they may come in telling you that they felt better after stopping the med. This is one reason you need to ask directly about side effects – and document them – while they are on the med, because unless they had some awful side effect that went away off the med, their sudden improvement in mood may be the other side of the placebo: the confidence boost of the placebo-flip. 

CHRIS AIKEN: Now for a research update. This is one you can use in your practice right now. Those of you who’ve heard me talk on antipsychotics know that I’m skeptical of their use in major depression. Why? They are only approved for short-term use in depression. And that matters because the FDA was only considering their short-term risks when they gave us the green light to use them. Sure, they tried to get approval for maintenance treatment in major depression, but they failed to prevent depression in most of the industry-sponsored trials. Contrast that with Lithium – on the other hand – does prevent depression – and did so in 21 controlled trials lasting an average of 2 years. But practice is the reverse of what the FDA intended – we end up giving antipsychotics to patients with chronic or recurrent depression, and those patients stay on them long-term. As the years build up, their benefits shrink, and their risks increase – particularly metabolic and tardive dyskinesia. And we are not doing a good job at detecting those things – around half of patients on antipsychotics get metabolic screening, and around half get AIMS screening. The result is vast under-diagnosis. Take this 2022 study from the public sector – where the problem is more common – only 1% of patients on antipsychotics are diagnosed with TD, while the actual rate with long-term use is 20-30%. Yes, even with the supposedly friendlier second-generation antipsychotics, 25% of patients will develop TD on them after a decade of use. This new study hopes to change that. Anthony Sterns and colleagues – including our fellow psychiatry podcast– tested whether an AI-driven video app could detect TD as accurately as trained clinicians in three samples totaling 356 patients on antipsychotics. By the end of the study, the four raters – two PhDs and two MDs – found themselves out of a job. The video app performed better, with an area under the curve of 0.9 – which means it rarely misses the diagnosis and rarely makes a false-positive. We don’t see areas under the curve that high very often in psychiatry. During the trial, one of the lead investigators, Owen Muir, discovered that he had tardive dyskinesia from a past antipsychotic exposure for mood disorder. Dr. Muir hosts the Frontier Psychiatrist podcast and spoke about the experience in the April 2025 episode, a very good one. So, that definitely brings it close to home, which is where this app needs to be. The app –TD Screen – is free, and the study was paid for by a grant from NIMH, and there’s a link to it on my website where you’ll find other evidence-based apps for practice: https://chrisaikenmd.com/apps. Put it to use, and it will vastly increase your detection of TD and improve your charting and documentation and e In your medical billing, you can bill more 9214 if you detect side effects and chronic problems. But if you do put it to use, I don't think it is going to be straightforward and easy. Nobody likes to go looking for bad news – not us and not our patients – so you are going to need to encourage patients to use the app or do it in session. Cobenfy is a new antipsychotic, but is it safe to combine with other antipsychotics, and does it make them work better? Find out in our June/July issue of the Carlat Psychiatry Report, where you’ll also learn about a new clozapine risk that people can prevent by titrating it slowly. Get $30 off your first year’s subscription with the promo code PODCAST and help us operate free of industry support.