Stephen Stahl and Jeffrey Strawn: Deprescribing as a core competency in psychiatry

In the evolving landscape of psychiatric care, deprescribing is emerging as a core competency. Last May at the 2025 American Psychiatric Association (APA) conference, I heard Christopher Aiken, MD; Nassir Ghaemi, MD; and a panel discussion led by Ronald Winchel, MD mention deprescribing in passing. Tetyana Rocks, Ph.D and colleagues presented a course titled Redefining mental health care: Integrating diet, lifestyle, and deprescribing in professional practice. 

Next year’s 2026 APA conference promises to offer more presentations specifically about deprescribing, by psychiatry leaders Donovan Maust, MD; John Kane, MD; Anita Clayton, MD; and others.

In early February, Stephen Stahl, MD, PhD, an internationally recognized educator in psychiatry with specific expertise in psychopharmacology, and co-author Jeffrey Strawn, MD published Stahl’s Deprescriber’s Guide. In it, they present deprescribing not as the antithesis of prescribing but a deliberate, patient-centered intervention. 

I interviewed Drs. Stahl and Strawn about their new book via email.

In 2000, he founded the Neuroscience Education Institute (NEI) to provide continuing medical education (CME) about psychopharmacology. Currently, he is NEI’s chairman and Editor-in-Chief of the journal CNS Spectrum.

Recently, Dr. Stahl estimated that his many publications have sold more than 2 million copies worldwide (the number of psychiatrists globally is less than 500,000). 

His publications and speaking have made an internationally recognized educator in psychiatry with specific expertise in psychopharmacology. He is also an Adjunct Professor of Psychiatry at the University of California, San Diego and an Honorary Fellow in Psychiatry at the University of Cambridge. 

He directs the Anxiety Disorders Research Program, where his work focuses on the treatment of anxiety and depressive disorders in children and adults.

In our correspondence, Drs. Stahl and Strawn, looking beyond psychopharmacology’s historical emphasis on initiating medications while often overlooking strategies for discontinuing them, envision deprescribing becoming normalized, with enhanced guidelines, and shared decision-making, transforming it into routine clinical practice.

They highlighted critical gaps in deprescribing research that future studies must bridge, including comparative trials of tapering methods like linear, prolonged, and hyperbolic strategies tailored to individual factors such as treatment duration and patient history. They call for longer-term investigations to better differentiate withdrawal symptoms from relapse. Ultimately, they anticipate the trajectory of deprescribing shifting from debating its merits to refining its application. 

Stahl and Strawn respond to my questions

Badre: Deprescribing is gaining significant attention, with developments of guidelines and prominent discussions at the 2025 APA conference. How would you characterize the current role of deprescribing in psychiatry, and where do you see it heading in the next few years?

Strawn & Stahl: Deprescribing is finally being recognized as a core competency, not an ideological movement. As we write in the Deprescriber’s Guide, deprescribing isn’t simply the reverse of prescribing. Instead, it’s a strategic and patient-centered process that requires as much judgment and pharmacology as starting a medication.

Looking ahead, we think deprescribing will become more normalized and less politicized. We’ll see clearer, medication-specific guidance, better integration of shared decision-making, and more realistic research that reflects how long people actually stay on medications. Ideally, deprescribing becomes what it should be: routine clinical maintenance, not a movement.

Badre: Could you share some of the core principles behind your approach to deprescribing? I am particularly interested in your thoughts on the scientific basis and practical application of hyperbolic tapering.

Strawn & Stahl: Our basic principle is that stopping a medication is a clinical intervention. It deserves planning, monitoring, and follow-up, just like starting one. In practice, that means individualizing how tapering or discontinuation is performed based on the initial condition being treated, the patient’s history, the duration of treatment, the reason for initial treatment, prior discontinuation experiences, and psychological factors.

The goal isn’t to taper as slowly as possible or taper based on one single receptor, it’s to taper as slowly as necessary for that patient.

Badre: Many clinicians are concerned about the risks of deprescribing, such as relapse or withdrawal. How do you address these concerns, and what does the evidence say about the range of patient outcomes?

Strawn & Stahl: When we talk with patients and clinicians about deprescribing, we usually start with a simple statement: two things can be true at once. Some patients experience real discontinuation symptoms. And many patients relapse when an effective medication is stopped, not because the drug harmed them, but because the underlying illness returns. Most of the confusion, and much of the polarization in this area, comes from collapsing those two phenomena into one.

Withdrawal itself is heterogeneous. Its severity varies widely among patients and medications, and much of the controversy reflects differences in how it is measured.

That suggests that depressive symptoms emerging later in the course are more likely to reflect relapse rather than withdrawal per se. At the same time, real-world survey data tell a complementary story. Among patients with longer antidepressant exposure, a subset report more intense and more persistent symptoms during discontinuation. These studies are careful to acknowledge their limitations (e.g., selection bias, recall effects, lack of placebo control), but they remind us that withdrawal is not imaginary, nor is it uniform.

Clinically, in the Deprescriber’s Guide, we try to keep colleagues out of two equally unhelpful traps. One is the reflexive minimization of withdrawal. The other is the opposite, assuming withdrawal is inevitable, severe, and prolonged for most patients. Neither position is supported by the evidence, and neither serves patients well. 

Relapse, on the other hand, is a well-established risk in many conditions and is often misattributed to withdrawal. Discontinuation trials repeatedly show that stopping an effective medication increases relapse risk, and that risk is not static. That is to say that it depends on how long the patient has been well. As we discuss in the Deprescriber’s Guide, a double-blind discontinuation study in generalized anxiety disorder demonstrated that patients treated with venlafaxine ER for 12 months before tapering had substantially lower relapse rates than those treated for only 6 months (32% versus 54%) (Rickels et al., 2010). Findings like this address myths in our field.

What is your framework for navigating this diagnostic gray area?

Strawn & Stahl: In practice, we focus on timing, symptom pattern, and trajectory. Symptoms that emerge quickly after dose reduction, fluctuate within a day, or include features the patient never had before are more suggestive of discontinuation effects.

Symptoms that resemble the original illness and persist or worsen over time raise more concern for relapse. 

We also rely heavily on measurement-based care and functional outcomes, not just symptom descriptions. Sometimes the only way to clarify the picture is to pause the taper, slow it down, or briefly reinstate a dose and observe what happens. That’s not a setback, it’s diagnostic information.

Badre: Psychiatric care has shifted from primarily short-term interventions to frequent long-term maintenance. How does deprescribing fit into this modern context, and what role should it play in routine practice?

Strawn & Stahl: Psychiatry now treats many conditions as long-term or recurrent, and often appropriately so. Deprescribing doesn’t oppose that model. It gives clinicians a structured way to ask whether the current medication, dose, and combination still make sense for the patient in front of them.

Strawn & Stahl: Despite growing clinical interest, practice has moved faster than evidence, and the next phase of the field needs to be driven by data that reflect how patients actually discontinue medications in real-world settings.

A central unanswered question is who benefits from which tapering strategy. Linear, prolonged, and hyperbolic tapers are widely used, yet we lack head-to-head trials that consider important clinical phenomena (e.g., dose, duration of exposure, prior withdrawal history, age and comorbidity). 

Another major gap is more precise differentiation between withdrawal and relapse. Many trials assess discontinuation effects over short intervals and rely on brief symptom checklists, which may capture early physical symptoms but miss later affective or functional changes. As highlighted in a recent meta-analysis (Kalfas et al., 2025), mood worsening after antidepressant discontinuation does not consistently parallel early withdrawal signals, suggesting that relapse and withdrawal are often conflated. Studies with longer follow-up, repeated measurement, and placebo-controlled discontinuation designs are needed.

We, as a field, also lack evidence on how to communicate about deprescribing without creating nocebo effects. Transparency is essential, but how risk is framed likely influences symptom experience. 

Taken together, the future of deprescribing research is not about deciding whether deprescribing is good or bad, but about determining how to do it well for the right patient, at the right time, and in the right way.

---

Nicolas Badre, MD is a clinical and forensic psychiatrist in San Diego and an advisor to the Psychotropic Deprescribing Council. He teaches medical education, psychopharmacology, ethics in psychiatry, and correctional care at various institutions, including USD Department of Counseling & School of Law, NMCSD Psychiatry, and FHCSD Psychiatry. His psychiatry textbook, Essential Psychopathology & Its Treatment,  5th Edition, can be pre-ordered on Amazon. His website is BadreMD.com.