My recent experience with a woman in her early 60s illustrates the complexities of deprescribing. She had been taking fluoxetine (Prozac) for over 35 years and bupropion (Wellbutrin) for more than a decade. Despite two previous unsuccessful attempts to reduce fluoxetine, each resulting in a return of depression, we successfully managed a gradual taper. This involved reducing her fluoxetine by 5 milligrams every three months, with an additional month’s hold if she experienced instability, alongside mood tracking, regular therapy, and close monitoring. The fluoxetine taper alone spanned 18 months, followed by a three-month pause before beginning a gradual, half-dose reduction of bupropion.
While the fluoxetine taper went smoothly, halving her bupropion dose during a period of significant professional stress led to a spike in anxiety and a sharp rise in her previously well-controlled blood sugar. Bupropion has minor glucose-lowering effects, and reducing the dose likely removed this protective factor. We temporarily reinstated her previous dose, resuming the taper only after the stressor subsided. This lengthy process underscores that deprescribing is not inherently dangerous but rather a dynamic, individualized endeavor demanding continuous clinical judgment, not a singular decision.
Robert F. Kennedy Jr.’s initiative to assist Americans in discontinuing antidepressants addresses a legitimate clinical concern: deprescribing remains understudied, undertaught, and inadequately reimbursed. For four decades, pharmaceutical companies have invested in trials demonstrating the efficacy of their medications, yet very few have funded research into when and how to safely cease their use. However, Kennedy’s approach unfortunately intertwines this valid clinical need with claims that lack evidence and some that are outright hazardous.
Recognizing this gap, the American Society of Clinical Psychopharmacology (ASCP), a prominent organization of psychiatrists and researchers, convened a summit in 2025. Forty-five international experts on antidepressant deprescribing reached consensus on 44 of 50 statements. Their valuable recommendations suggest considering deprescribing when medications’ functions overlap, improvement has been minimal, a drug is no longer effective, or side effects outweigh benefits, always making only one change at a time. Crucially, the task force also determined that deprescribing is not universally suitable; for patients with three or more lifetime depressive episodes, the consensus recommends indefinite medication maintenance. While these principles are sound and evidence-based, their safe implementation demands individualized clinical judgment, gradual tapering, psychological support, and accessible alternatives—components that current federal initiatives have not delivered at scale.
Clinically responsible patient redirection from medication necessitates accessible alternatives, which are currently lacking. A 2025 JAMA Psychiatry study revealed that recent increases in psychotherapy utilization primarily benefited younger, wealthier, college-educated, urban adults with private insurance, while older, less educated, uninsured, or rural adults experienced no significant rise. This disparity partly explains why approximately one in six American adults currently take an SSRI; a prescription offers an accessibility that weekly therapy appointments often do not. Encouraging clinicians to reduce prescribing without simultaneously investing in therapy infrastructure will only exacerbate this access gap.
Regarding discontinuation symptoms, the most extensive meta-analysis to date indicates that patients stopping antidepressants experienced, on average, one additional symptom compared to placebo, falling below the threshold for clinically significant discontinuation syndrome. A related Lancet Psychiatry analysis found that about 15% experience symptoms attributable to cessation, with severe symptoms affecting roughly 3%. While these figures represent genuine experiences, they do not substantiate Kennedy’s assertion that SSRIs are more difficult to quit than heroin. Stanford addiction researcher Keith Humphreys aptly noted that antidepressants and heroin occupy “different universes” concerning addiction risk. The DSM-5 clarifies that antidepressant discontinuation syndrome lacks drug craving and is unrelated to the reinforcing effects seen in substance dependence. Furthermore, Kennedy’s claim linking SSRIs to school shootings is entirely unsupported by research literature. Such statements, made from a position of federal authority, stigmatize depression treatment and deter individuals from seeking necessary care.
The ASCP consensus also highlights that long-half-life SSRIs, such as fluoxetine, rarely cause significant discontinuation symptoms, a finding supported by pharmacokinetic data. Nevertheless, the FDA label still advises gradual reduction over abrupt cessation. A 2026 Lancet Psychiatry network meta-analysis of 76 trials further demonstrated that abrupt discontinuation of any antidepressant significantly increased the risk of relapse compared to slow tapering with psychological support. While fluoxetine’s extended half-life mitigates discontinuation symptoms, it does not eliminate the risk of depression relapse. These are distinct clinical endpoints, and their conflation poses a dangerous misrepresentation. Although stopping antidepressants roughly doubles the rate of recurring depression symptoms, a 2021 Cochrane review noted that many trials failed to differentiate true relapse from withdrawal symptoms misclassified as relapse. This crucial distinction reinforces the necessity of structured deprescribing: slow tapering accompanied by psychological support was found to be as effective as continued medication in preventing relapse, whereas slow tapering alone did not significantly outperform abrupt cessation. The taper is essential, and psychological support is indispensable.
Given the ASCP’s recommendation for indefinite maintenance therapy in patients with three or more lifetime depressive episodes, a federal deprescribing initiative that lacks individualization for this demographic carries substantial clinical risk. In stark contrast, Kennedy’s event showcased speakers who promoted ideological, rather than evidence-based, ideas like eliminating school-based mental health screenings and implementing cigarette-style warnings on antidepressant packaging. However, Kennedy’s initiative is not entirely without merit. Its most evidence-aligned proposal is the CMS reimbursement mechanism for deprescribing time, addressing inadequate reimbursement and misaligned quality-measure incentives—recognized structural barriers to deprescribing adoption. Yet, even this valuable aspect is effective only if clinicians are equipped with evidence-based protocols and patients have access to psychological support.
A more profound need lies with the FDA. While general tapering strategies exist, there is a critical absence of drug-specific tapering protocols derived from randomized discontinuation trials for each approved medication in this class—trials the FDA has never mandated manufacturers to conduct. Furthermore, the discussion often overlooks the workforce challenge. Most antidepressant prescriptions in the U.S. are issued by primary care physicians, not psychiatrists. These clinicians manage high patient volumes, short visits, and possess limited psychiatric training. The meticulous, months-long tapering process demands specialized skills and time commitments that most primary care settings are not designed to accommodate. For deprescribing to occur at scale, it would necessitate either a significant expansion of psychiatric consultation capacity or the creation of entirely new clinical infrastructure—neither of which Kennedy’s initiative addresses. The FDA’s role warrants further examination. While some contend the agency has never required manufacturers to study medication cessation, it is crucial to determine if this is truly unprecedented or if a regulatory mechanism exists but has simply not been invoked for this drug class. Should historical precedent within the FDA or international regulatory frameworks exist, the obstacle would be institutional, not legal. Clinical pharmacologists and regulatory historians could offer valuable insights on this seldom-explored aspect. After four decades of widespread antidepressant prescribing, with one in six adults currently on these drugs, we still lack optimal cessation strategies for the full spectrum of patients. The ASCP task force is well-positioned to formally demand such research, and psychiatrists should urge them to do so.
Effective deprescribing demands gradual tapering, comprehensive psychological support, meticulous risk stratification, and readily available alternatives—none of which Kennedy’s initiative offers at scale. Patients taking antidepressants deserve both an honest assessment of their ongoing need for medication and a robust system capable of safely guiding them through cessation if it is no longer required. Presently, we possess neither the necessary data nor the infrastructure to provide this.
Source link
