Psychiatric medication reviews are getting more attention. That is a good thing, with one important caveat: a review can become another administrative ritual.
In February 2026, an American Society of Clinical Psychopharmacology task force published consensus recommendations calling for periodic reassessment of psychotropic medications, at least annually. In May, federal agencies also emphasized medication review and, when appropriate, deprescribing as part of individualized mental health care.
That sounds straightforward. It isn’t.
A medication review can mean anything from confirming a pharmacy and renewing a prescription to doing the harder work of reconsidering whether a medication still has a favorable place in someone’s life. Those are not the same activity.
A review can confirm. Reconsideration has to investigate.
It is entirely reasonable for a medication to remain part of treatment over time. It may still be reducing symptoms, preventing recurrence, supporting function, or making other parts of life more manageable.
But continued prescribing should not be mistaken for continuing evidence of benefit. A prescription can persist because it is helpful. It can also persist because no one has paused long enough to reconstruct the original decision.
I keep coming back to a basic question: What, specifically, is this medication doing now that makes continued exposure worth it?
That question is not anti-medication. It is the same question we should ask of any long-term intervention.
A useful reconsideration looks beyond whether someone feels generally “fine.” It examines what was happening before the medication began, why it was started, what changed after starting, what changed with later dose adjustments, and what may have changed during long-term use.
It also asks whether the original target is still the target. A medication started during a severe episode, an acute crisis, a major disruption in sleep, or a difficult life period may still be useful years later. Or the context may have changed substantially. Neither conclusion should be assumed.
“It isn’t working” can mean several things
The recent consensus statement makes a point that is easy to overlook: before labeling a medication ineffective, adherence and adequacy of the trial deserve examination.
That matters because an apparent lack of benefit can have several explanations. The medication may not be taken consistently. It may have been taken inconsistently because of adverse effects, cost, access problems, ambivalence, or a schedule that does not fit real life. The original symptom may have been misunderstood. The dose or duration may not have allowed a fair assessment. Or the medication may simply not be delivering enough benefit.
Those explanations lead to different decisions. Collapsing them into “it failed” makes the next decision less reliable.
The same discipline applies in the other direction. “It works” can mean that a person notices a clear and sustained improvement linked to the medication. It can also mean they fear what might happen without it, have not had an opportunity to examine its effects, or have experienced symptoms after a prior change that were automatically classified as relapse.
Fear is not evidence of ineffectiveness. It is also not evidence that a medication remains necessary indefinitely.
Do not let withdrawal answer a question it cannot answer
This is where medication decisions often lose precision.
Physical dependence can develop with long-term exposure to some psychiatric medications. That describes physiological adaptation. It does not, by itself, mean addiction, compulsive drug-seeking, or that the original prescription was inappropriate.
It does mean that symptoms following a reduction or discontinuation may be difficult to interpret. Withdrawal can include symptoms that resemble the original reason for treatment, including anxiety, insomnia, low mood, agitation, or panic. Rebound symptoms, return of previous symptoms, recurrence of an underlying condition, medication effects, and life circumstances can also overlap.
So what do we actually know in a given situation? Often less than people want to believe.
Timing matters. Symptoms that emerge soon after a medication change may raise one set of questions. Symptoms that appear much later, follow a familiar prior pattern, or occur alongside major life changes may raise others. But timing alone does not prove the cause. Nor should uncertainty become a reason to dismiss a plausible temporal pattern.
Some people report symptoms that persist well beyond the period when a medication has left the bloodstream. The emerging literature on prolonged antidepressant withdrawal makes clear that this possibility deserves study, not reflexive dismissal. At the same time, persistent symptoms alone cannot establish protracted withdrawal. Competing explanations remain important.
The point is not to force every experience into one category. The point is to stop pretending the categories are interchangeable.
Benefit and burden both need names
A vague risk-benefit discussion is not much of a discussion. “Benefits outweigh risks” is a conclusion, not an analysis.
A more useful conversation names the actual benefit: fewer panic episodes, more stable sleep, fewer severe mood episodes, better concentration, less distress, or more reliable daily function.
It also names the actual burden: sexual effects, emotional blunting, weight or metabolic changes, sedation, disrupted sleep, cognitive effects, movement symptoms, interactions with other medications, financial cost, or the practical burden of monitoring.
Not every unwanted effect is an adverse effect caused by a medication. Not every change after starting a medication is unrelated, either. This is why a timeline matters more than a checklist alone.
There is another burden worth naming: complexity. As medication lists grow, it becomes harder to know which medication is helping, which is contributing to a problem, and how one medication may alter the effects of another. A regimen can be clinically justified and still deserve periodic simplification questions.
The question is not “keep or stop?”
The most useful current shift is not toward more medication or less medication. It is toward less automatic medication.
Reconsideration asks for a working account of the medication’s present role, the costs of continued exposure, the costs of changing course, the uncertainty around both, and the person’s own priorities.
That is slower than a refill. It is also more honest.
When a medication is reviewed, the question worth leaving the room with is not simply, “What happens next?” It is: What evidence would tell us that this treatment is still serving its intended purpose, and what evidence would tell us to reconsider it?
Written for Dr. Teralyn Sell, PhD
Psychology · Brain Health · Human Behavior
Sources worth knowing
- Recommendations for the Deprescribing of Psychotropic Medications: A Consensus Statement From the American Society of Clinical Psychopharmacology Task Force
- Dear Colleague Letter: Psychiatric Medication and Non-Pharmacological Treatments
- FAQs: Billing for Supervised Medication Deprescribing Services Under Care Management Codes on the Medicare Physician Fee Schedule
- Post-acute withdrawal syndrome (PAWS) after stopping antidepressants: a systematic review with meta-narrative synthesis