Beyond Deprescribing: A Formula for Excellent Clinical Care

At the American Society for Psychopharmacology (ASCP) Annual Meeting in Miami Beach, Florida, 8 senior psychopharmacologists sat down to address a question the field has largely left implicit: What does it mean to actively review each medication a patient is taking and determine if it still belongs in the regimen? Their answer amounts to a clinical framework and a recommitment to the kind of deliberate prescribing that defines the specialty.

The roundtable, filmed for the Brain Trust: Conversations in Clinical Psychopharmacology series and moderated by Joseph F. Goldberg, MD, of the Icahn School of Medicine at Mount Sinai, featured members of the ASCP Deprescribing Task Force in Psychopharmacology: Anita H. Clayton, MD; Leslie Citrome, MD, MPH; David W. Goodman, MD; Rajnish Mago, MD; Roger S. McIntyre, MD, FRCPC; Holly A. Swartz, MD; and Mauricio Tohen, MD, DrPH, MBA. Their message was affirmative and specialty-led: Systematic medication review is not a reaction to outside pressures. It is what great psychiatrists have always done and what the field is now formalizing as an explicit standard of care.

What the Debate Gets Wrong

In categorically rejecting the appropriation of the term deprescribing by voices arguing that psychiatric medications are broadly harmful and should be eliminated, the panel also rejected complacency in the other direction. Instead, the panel focused on the concept of appropriate prescribing. At every patient encounter, a well-trained clinician should be asking whether each medication in a regimen is still indicated, still effective, and still the right fit for that patient at that point in their illness course, they noted.

The panel was equally clear that the term deprescribing should not be misread as a directive to reduce medication burden universally. In bipolar disorder and schizophrenia, the panel was unambiguous: Long-term pharmacotherapy is typically indicated, and premature discontinuation carries real risk. In bipolar disorder specifically, abrupt medication cessation has been associated with outcomes worse than no treatment at all, a finding with direct implications for how clinicians frame conversations with patients who want to stop a medication on their own timeline. Deprescribing to achieve a medication-free status was considered categorically inappropriate in bipolar type I disorder.

Addressing the Real Issues

A convergence of structural and educational forces has made appropriate medication review more difficult and, in many cases, less likely to happen at all, the panel agreed.

The prescriber landscape has shifted dramatically over the last 2 decades. More than 90% of antidepressants and approximately 50% of stimulants are now prescribed by clinicians outside psychiatry, many of whom have limited training in psychopharmacology, limited time with patients, and limited capacity for the required longitudinal follow-up. When brief appointments are the norm, diagnostic reassessment rarely happens, medication regimens accumulate, and partial responses go unchallenged.

Layered onto this is a system that is fundamentally fragmented, they added. Multiple prescribers, multiple pharmacies, and telehealth platforms with variable evaluation standards all operate without a single coordinating clinician responsible for the whole picture. In this environment, inappropriate polypharmacy is not an anomaly; it is a predictable output of a broken infrastructure. The internet further complicates matters, functioning as an unregulated prescribing influence that sometimes operates entirely outside any clinical relationship.

Measurement-based care has not kept pace either, they lamented. Without validated metrics applied consistently, clinicians and patients alike can misread stability for adequacy. A patient who is not well but not in crisis may, on a brief visit, appear to be appropriately managed. Without systematic tracking of symptom burden over time, the gap between “stable” and “well” can persist for years while a medication that was never fully effective remains on the list.

Diagnosis itself is another fault line, with psychiatric diagnoses evolving, according to the panel. A patient who initially presents with psychotic depression may, over time, meet criteria for bipolar disorder or schizoaffective disorder, not because the original diagnosis was wrong, but because the illness had not full presented. If the medication regimen does not evolve with the diagnosis, it may no longer be appropriate, even if it once was. The panel pointed to the telehealth-mediated expansion of attention-deficit/hyperactivity disorder (ADHD) diagnosis as a specific example of this problem, notably, patients on stimulants who may have received inadequate diagnostic evaluations, whose medications should be reassessed before any second clinician assumes the diagnosis is accurate.

Then there is the human side of the equation. Psychological as well as physiological attachment to medications is a genuine and underappreciated barrier, the panel agreed. Patients, families, and clinicians all develop resistance to change, even when a regimen is demonstrably suboptimal. Adverse effect tolerability is another driver that clinicians systematically underestimate, with weight gain, cognitive dulling, and tremors cited as the leading reasons patients stop their medications unilaterally. The clinical literature may classify these effects as manageable, but patients often disagree and act on that disagreement without telling their prescriber.

Additionally, the panel pushed back on any framing that positions deprescribing as a one-directional goal. The evidence on undertreatment is unambiguous. Appropriate prescribing means recognizing both failure modes: the patient who is on something they no longer need, and the patient who is not on something they do.

What to Do at the Next Patient Encounter

The goal is not to resolve a complex polypharmacy question in a single visit; rather, the goal is to initiate a process (Table).

Begin with a single opening question: Ask the patient whether they have any concerns about their medications or the number of medications they are taking. The answer immediately reveals how much room there is to work with, they said. The clinical work will continue to unfold over multiple visits— not a single decisive conversation, but an iterative relationship.

Before any medication is changed, revisit the diagnosis, the panel said. Ask yourself: Is the diagnosis the patient carries still accurate? Has the illness evolved? Are validated screening tools being used consistently? A patient who has not responded to multiple antidepressants may warrant screening for bipolar disorder, borderline personality disorder, or ADHD before attempting another antidepressant trial.

When reviewing medications, recognize that not every concern can be addressed at once. Identify the highest-stakes issue—ie, a drug combination producing orthostatic hypotension, a pattern of anticholinergic burden contributing to cognitive impairment, an agent with no documented response after an adequate trial—and address that first. Pharmacological expertise is precisely what allows a psychiatric clinician to see what a primary care provider cannot: which medications are working against each other, which adverse effects are modifiable, and which regimens can be simplified without compromising therapeutic gains.

See highlights from the panel here:

• Joseph F. Goldberg, MD (Moderator) Clinical Professor of Psychiatry, Icahn School of Medicine at Mount Sinai, New York: “It’s not that medicines are bad; it’s that you may be taking something conflictual. It also maybe something conflictual with a nonpsychotropic medication. So a good clinician can potentially step in and say, ‘I can make you better.’”
• Roger S. McIntyre, MD, FRCPC Professor of Psychiatry and Pharmacology, University of Toronto: “It’s a relationship you build with a person. And I think most times, most patients will have trust in someone who has their best interest at heart, and just lay out what you’re doing. That fits into a 10-minute office visit for me.”
• Rajnish Mago, MD Clinical Professor of Psychiatry, SUNY Upstate Medical University: “It’s a dedication to acknowledging that people may be on medications that are interacting, some of which may be unnecessary or can be simplified.”
• Anita H. Clayton, MD Professor of Psychiatry and Neurobehavioral Sciences; Professor of Clinical Obstetrics and Gynecology, University of Virginia: “What we need to be doing is finding the right treatment for them, whatever it is. And if we’ve got them on something that isn’t working, we need to go to the next step, and sometimes that means adding a medicine if they’ve had a partial response. And sometimes that means switching, if they’ve had a poor response.”
• Holly A. Swartz, MD Professor of Psychiatry, University of Pittsburgh: “The risk with some of the dialogue around prescribing and deprescribing is
  that the psychosocial components of this are not well integrated into our care delivery, into our discussions around alternatives, around personalizing care.”
• David W. Goodman, MD Clinical Associate Professor, SUNY Upstate Medical University; Faculty, Johns Hopkins Health System; Adult ADHD Specialist: “I ask, ‘Do you have any concerns about your medications, or the number of medications you’re on?’ That gives me an opening. If they say ‘no,’ then I know my road is going to be a bit uphill. If they say ‘yes,’ I know that I have an in, and I can then address which of the medications I would like to prioritize.”
• Leslie Citrome, MD Clinical Professor of Psychiatry and Behavioral Sciences, New York Medical College; Adjunct Professor, Northwestern Feinberg School of Medicine: “If medicines were prescribed in a way that they ought to be, then they should help
  and not harm. And if they don’t help, they should be replaced by something else for that individual person. But that doesn’t always take place, and people are on medicines that don’t work for long periods of time, and people get disillusioned.”
• Mauricio Tohen, MD, DrPH, MBA Chair, Department of Psychiatry and Behavioral Sciences, University of New Mexico: “Every time we see a patient, we should consider: Should we continue this medication or not? That is good clinical practice.”