Estimated reading time at 200 wpm: 22 minutes

This article addresses the concept of planned temporary, prolonged or permanent discontinuation of psychiatric medication. This is colloquially termed ‘drug holidays’, within the specific context of United Kingdom psychiatric practice. The primary focus is directed towards the significant medico-legal considerations inherent in such decisions, acknowledging the particular sensitivities and heightened risk environment encountered within psychiatry ¹. The aim is to provide a comprehensive analysis for consultant psychiatrists navigating these complex clinical and legal intersections.

The planned discontinuation of psychiatric medication requires careful distinction between three key approaches: drug holidays, tapering, and deprescribing. Each serves a different clinical purpose and carries distinct risks, particularly in the management of chronic mental health conditions.

• Drug holidays refer to temporary breaks from medication, most commonly in ADHD treatment (e.g., pausing stimulants on weekends). Outside this context, the term is controversial, as abrupt pauses in antipsychotics or antidepressants risk relapse or severe withdrawal.
• Tapering involves gradual dose reduction with the intent to stop medication temporarily or permanently, minimising withdrawal effects. This is standard for antidepressants and antipsychotics.
• Deprescribing is a structured process to reduce or eliminate unnecessary, ineffective, or harmful medications, often in polypharmacy cases. Unlike drug holidays, it is not temporary.

Clinical rationale and motivations for discontinuation

Patient-initiated requests

Common reasons patients seek to stop psychiatric medication include:

• Intolerable side effects (e.g., weight gain, emotional blunting, sexual dysfunction).
• Perceived lack of need (e.g., feeling stable and believing medication is no longer necessary).
• Concerns about long-term health risks (e.g., metabolic effects of antipsychotics).

Sources: Mind UK, Royal College of Psychiatrists.

Clinician-initiated considerations

Doctors may recommend discontinuation in cases such as:

• Assessing ongoing need (e.g., reviewing long-term antipsychotic use after years of stability).
• Managing side effects if there is polypharmacy (e.g., reducing doses to alleviate sedation or metabolic impacts).
• Deprescribing initiatives that end in drug holidays (e.g., NHS England’s STOMP campaign to reduce inappropriate psychotropic use in intellectual disability).

Sources: NICE NG215, STOMP Programme.

Diagnostic clarification

Stopping medication is not part of recognised diagnostic processes. Stopping medication to confirm a diagnosis is highly risky and not routinely recommended. The author sees such activity as hocus-pocus and invitational of lawsuits. This will be explored later on in more detail but for the while consider a proposal by a doctor to stop a patient’s digoxin to confirm heart failure, or stopping treatment for COPD if uncertain to confirm COPD. Then consider the whether the courts will accept that as logical as per Bolitho even if a majority of your colleagues might do the same (as per Bolam defence). Read up.

Note the following:

• Withdrawal symptoms can mimic relapse, leading to misdiagnosis.
• Sudden cessation may unmask an underlying condition, complicating assessment.

Sources: BJPsych Advances, Maudsley Prescribing Guidelines.

Risks Associated with Planned Discontinuation

Relapse risk (Antipsychotics, Antidepressants)

Planned discontinuation of psychotropic medications, particularly antipsychotics and antidepressants, carries a clinically significant risk of relapse. The extent of this risk depends on diagnosis, duration of prior treatment, and the underlying stability of the patient’s mental state. In patients with schizophrenia or recurrent depressive disorders, abrupt cessation has been associated with marked increases in relapse rates. NICE guidelines on psychosis and schizophrenia (NG222), which advocate for continued treatment in patients with a history of severe illness, unless there is clear rationale and structured planning for withdrawal.

Principal messages distilled from Horowitz & Moncrieff (2024)

Implications for practice

1. Rate of change, not merely absolute dose, dictates physiological stability.
2. Hyperbolic* tapers maintain near-constant receptor-occupancy decrements, reducing neurochemical disequilibrium.
3. Early symptom emergence post-reduction warrants observation and supportive measures before reinstating higher doses.

[*A hyperbolic taper is a dose-reduction schedule in which each successive step becomes a smaller absolute amount of drug than the one before it. The sequence follows the same curve that describes receptor occupancy (a rectangular hyperbola), so that every cut removes roughly the same percentage-points of target-receptor blockade instead of the same milligrams.]

Quantitative illustration (PET-based)

For haloperidol, terminating therapy at 0.25 mg lowers striatal D₂ occupancy by roughly 25 percentage points in a single step; tapering further to 0.08 mg before cessation limits the final occupancy change to approximately 10 percentage points. The smaller physiologic perturbation is presumed to translate into a lower relapse risk and fewer withdrawal manifestations.

Withdrawal symptoms (prevalence, severity, duration)

Withdrawal phenomena are often underestimated by both patients and clinicians. Symptoms may range from mild discomfort to severe functional impairment and may mimic relapse, complicating clinical interpretation. For antidepressants, the term “discontinuation syndrome” is frequently used, though some experts argue this underplays the potential severity. Notably, the Royal College of Psychiatrists acknowledges that withdrawal effects can last for weeks or months, especially with paroxetine or venlafaxine. In the case of antipsychotics, sudden cessation may provoke agitation, insomnia, rebound psychosis, or cholinergic symptoms. Tapering protocols, rather than abrupt withdrawal, are increasingly recognised as a necessary safeguard.

Diagnostic confusion (withdrawal vs. relapse)

One of the most complex challenges during drug holidays is distinguishing true clinical relapse from withdrawal phenomena. Symptoms such as mood lability, anxiety, or agitation may emerge within days of discontinuation and can be misattributed to recurrence of illness. Misdiagnosis may lead to inappropriate reinstatement of medication or escalation of treatment. This diagnostic ambiguity is particularly pronounced in forensic settings, where risk thresholds are higher and misinterpretation can carry medico-legal consequences. GMC guidance on clinical assessment and continuity of care underscores the importance of vigilance and clinical clarity in such contexts.

Medico-legal considerations

Duty of care and clinical negligence (Bolam Test, Bolitho and Montgomery UKSC)

When considering planned discontinuation of psychotropic medication, the clinician’s duty of care is engaged from the outset. This duty requires the exercise of reasonable clinical judgment, and that such judgment meets the standards of a responsible body of medical opinion. The leading authority remains the Bolam v Friern Hospital Management Committee [1957] 1 WLR 582, which established that a doctor is not negligent if acting in accordance with a practice accepted as proper by a responsible body of medical professionals. However, this is not the end of the matter.

The courts have since clarified that professional consensus alone does not insulate a decision from scrutiny. In Bolitho v City and Hackney Health Authority [1997] UKHL 46, the House of Lords ruled that any professional opinion must also be demonstrably logical and reasonable. If a court considers the rationale behind a body of medical opinion to be incoherent or internally inconsistent, it may reject that opinion altogether. This is particularly relevant when ceasing treatment in complex psychiatric cases, where risks may be high, evidence may be equivocal, and where the clinician’s rationale must be clearly articulated and capable of logical defence.

In addition, the Montgomery v Lanarkshire Health Board [2015] UKSC 11 decision profoundly reshaped informed consent by affirming patient autonomy over paternalistic models of decision-making. The judgment requires that patients be told of any material risks involved in a proposed treatment (or its withdrawal), and of reasonable alternatives. Materiality is defined with reference to what a reasonable person in the patient’s position would want to know. In this regard, clinicians must ensure that patients understand the risks of relapse, withdrawal, diagnostic confusion, and long-term implications of discontinuation—particularly when such decisions are initiated at the patient’s request. A full reading of the Montgomery judgment is available on BAILII.

Informed consent and capacity assessment (MCA 2005)

In England and Wales, the Mental Capacity Act 2005 (MCA) provides the statutory framework for assessing a person’s capacity to make decisions, including about medication changes. Capacity is decision-specific and time-specific. The patient must be able to understand, retain, weigh, and communicate information relevant to the decision. If they lack capacity, any best interests decision must account for past and present wishes, beliefs, and values.

For psychotropic drug holidays, particularly those involving high-risk medications such as antipsychotics, clinicians must avoid assuming capacity based solely on apparent lucidity. The MCA Code of Practice—available via the UK Government’s legislation site—emphasises the need for structured, recorded assessments. In psychiatric contexts, fluctuating capacity is common, and periodic reassessment may be necessary.

Even when a patient is deemed to have capacity, consent is only valid if it is informed. As the General Medical Council’s guidance on decision making and consent states, clinicians must ensure that patients are given the information they want and need in a form they can understand. Simply providing generic warnings or verbal explanations is insufficient; risks must be contextualised to the patient’s situation.

For patients detained under the MHA 1983 the statutory duties of care are more serious. How?

1. Quite often patients are detained for treatment of a mental disorder (as defined by S145).
2. Withdrawing a treatment on a path of some experimentation may cause relapse, increase risks (to self or others), worsen prognosis for the future and extend hospital detention.

Psychiatric injury

A particularly fraught medico-legal issue arises when discontinuation is undertaken with the aim of clarifying or confirming a diagnosis, that results in psychiatric deterioration or relapse. Not because drug holidays are ‘standard’ in ADHD patients, means that these considerations are irrelevant for those patients.

While diagnostic clarity may be clinically desirable, inducing a relapse through medication withdrawal exposes the clinician to potential liability for avoidable harm. Courts have increasingly recognised psychiatric injury as a compensable form of damage, particularly where harm results from foreseeable risk. In this context, any decision to provoke symptom re-emergence must be justified not only by clinical rationale but also by a proportionality test—balancing potential diagnostic gain against the severity and likelihood of harm. Where a less intrusive means of achieving diagnostic understanding exists, opting for a high-risk strategy may fall below the standard of care. Moreover, such practice must be considered in full with the patient, with attention to GMC expectations on safeguarding against avoidable harm and the duty to prioritise patient wellbeing. Not because a patient is of capacity and can consent to a potential relapse means that the psychiatrist is in the clear – and risk-free – to proceed.

Guidelines and professional recommendations

NICE guidelines (e.g., NG215, NG222)

The National Institute for Health and Care Excellence (NICE) provides authoritative guidance on the treatment and withdrawal of psychotropic medications across various psychiatric conditions. Notably, NICE recognises both the clinical utility of long-term pharmacotherapy and the importance of structured, patient-centred approaches to discontinuation.

In the case of antidepressant withdrawal, NICE guideline NG215 on “Medicines associated with dependence or withdrawal symptoms: safe prescribing and withdrawal management” marks a significant evolution in official recognition of discontinuation-related harms. It advises that patients be warned—before initiating treatment—about the potential for withdrawal symptoms, which may be severe and prolonged. The guidance recommends gradual tapering and close monitoring, and explicitly warns against abrupt cessation except in cases of serious adverse effects.

For patients with psychosis or schizophrenia, NICE guideline NG222 sets out clear processes for considering medication reduction or withdrawal. The document advises that antipsychotic discontinuation should only be attempted when a patient has been stable for a significant period, and even then, only as part of a shared decision-making process with robust risk planning. Clinicians are directed to offer written information about the risks of relapse and to agree a strategy for rapid reinstatement should symptoms re-emerge. The guideline also highlights the importance of involving carers, particularly where insight or adherence may be impaired.

Both guidelines stress the need for individualised tapering regimens, guided by clinical judgment, patient preference, and history of prior discontinuation attempts. They recommend reviewing patients at regular intervals, with flexibility to adjust the speed of reduction according to tolerability.

Royal College of Psychiatrists (RCPsych) position statements

The Royal College of Psychiatrists (RCPsych) has published position statements reinforcing the complexity and risks involved in stopping psychotropic medications. In its 2020 position statement on antidepressants, the College acknowledged that withdrawal symptoms are often more severe and longer-lasting than previously appreciated. It calls for services to develop protocols that support clinicians in conducting careful, patient-informed tapering.

The RCPsych also places emphasis on shared decision-making and transparency. It advises that clinicians routinely discuss withdrawal effects when prescribing, as part of a long-term care plan. Moreover, the College supports the development of specialist deprescribing clinics and cross-disciplinary approaches, particularly for individuals on complex polypharmacy or with long-term dependency concerns. This is especially pertinent in forensic and community psychiatry, where patient history, risk factors, and social contexts are highly variable.

While not statutory, the College’s positions often shape professional norms and are frequently cited in medico-legal evaluations of practice standards.

GMC standards (Prescribing and Consent)

The General Medical Council (GMC) sets out core expectations for prescribing and managing treatment changes. Its guidance on prescribing and managing medicines and devices requires doctors to ensure that patients understand the purpose, risks, and alternatives to any medication—this includes the consequences of stopping it. Where drug holidays or dose reductions are proposed, doctors must be satisfied that the patient has the capacity to make the decision, and that they have been offered sufficient information in a comprehensible format.

In parallel, the GMC’s guidance on decision making and consent builds on the Montgomery principles. Doctors are required to tailor discussions to the individual, taking account of what matters most to that person. For drug holidays, this means not merely listing potential risks but ensuring the patient has understood and weighed them in the context of their own values and goals.

Both guidance documents underscore the importance of documentation—not only of decisions made but of the process by which those decisions were reached. This aligns with the broader principles set out in Good Medical Practice, which highlight accountability, continuity, and responsiveness as essential elements of safe prescribing.

Forensic psychiatry-specific challenges

Unique patient population and risk profile

In forensic psychiatry, the implications of medication discontinuation are particularly serious, due to the high baseline risks associated with the patient population. Many individuals in secure settings have histories of violence, sexual offending, or serious threats to others, often underpinned by severe and relapsing mental disorders. The decision to implement a drug holiday—whether for diagnostic clarity, side effect management, or patient preference—cannot be divorced from public protection duties embedded in the Mental Health Act 1983, particularly where Part III and Restriction Orders (s37/41) apply.

Where psychotropic medication plays a key role in reducing risk, its removal—even temporarily—requires a full multi-agency risk formulation and approval. Structured professional judgement tools such as the HCR-20 may be adapted to capture dynamic risk during withdrawal phases. Failure to consider and document potential risk escalation could expose clinicians and organisations to civil liability, coronial scrutiny, or serious incident investigations.

The Royal College of Psychiatrists’ Faculty of Forensic Psychiatry has emphasised in its training standards that risk management must be anticipatory and not solely reactive. Discontinuation, particularly of antipsychotics, must not occur in isolation but as part of an integrated risk management and relational security plan. This includes identifying early warning signs, trigger behaviours, and contingency pathways should rapid reinstatement of treatment be required.

Enhanced risk assessment and monitoring

Given the stakes, monitoring during planned discontinuation in forensic settings must go beyond routine clinical observations. In secure hospitals, this may require increased staffing, step-up to enhanced nursing observation levels, and active surveillance for relapse indicators. Digital or physiological monitoring technologies are being piloted in some services to detect early behavioural changes during tapering protocols.

Risk to self also remains a key concern. Many forensic patients experience comorbid depression, personality disorder, or trauma histories. Discontinuing mood stabilisers or antidepressants, even under supervision, may lower threshold for self-harm, suicidal ideation, or non-compliance with other aspects of their care plans. These risks are compounded where insight is poor or trust in the clinical team is fragile.

The GMC’s guidance on maintaining trust and protecting patients from harm reminds clinicians of their duty to balance individual autonomy with the need to prevent foreseeable harm to others. This duty is particularly weighty in forensic settings, where risks to others are not abstract but historically evidenced.

Legal frameworks (MHA 1983, Restriction Orders)

Within forensic psychiatry, legal frameworks offer both constraints and protections. Under sections 41 and 49 of the Mental Health Act 1983, the Secretary of State retains authority over discharge, leave, and in some cases, significant treatment changes. This legal oversight is specifically designed to mitigate the risk of reoffending or deterioration following changes in clinical status.

Drug holidays – rare in such circumstance – must therefore be planned not only in consultation with the patient and multidisciplinary team, but also in compliance with Ministry of Justice (MoJ) notification requirements. Clinical decisions that increase risk—whether in the short or long term—may trigger enhanced MoJ scrutiny. Risk-related rationale for medication changes must be carefully recorded and communicated, particularly as such documents may later be disclosed in court, tribunals, or public inquiries.

Case reference: New Zealand and repeated NGRI findings

An instructive case from New Zealand involved a patient found not guilty by reason of insanity (NGRI) on two separate occasions, reportedly linked to non-adherence with psychiatric medication. The case drew public and legal attention due to concerns about preventability and the role of the mental health service in monitoring the patient’s compliance and condition. While full anonymised legal texts are sparse in the public domain, coverage by national news outlets and the New Zealand Law Commission reflects the increasing accountability of forensic services in managing relapse risk when medication adherence is variable. It serves as a cautionary international parallel: failure to pre-empt known patterns of relapse related to treatment discontinuation may result in preventable harm to the public and irreversible reputational damage to the treating institution.

Certainly. Below is the expanded section for:

Best practices for risk mitigation

Patient selection and assessment

Not all patients are suitable candidates for drug holidays. Effective mitigation of risk begins with rigorous patient selection, informed by diagnosis, clinical history, previous responses to medication changes, and capacity to engage in shared planning. Patients with a history of abrupt decompensation, violence, or disengagement from services require heightened scrutiny. A detailed formulation—including biological, psychological, and social risk factors—should underpin any decision to initiate withdrawal.

This is especially important in the case of antipsychotic or mood stabiliser discontinuation, where the likelihood of rapid relapse is high. Where possible, structured assessments (e.g., START, HCR-20, or HoNOS) should be updated to reflect potential shifts in dynamic risk during the withdrawal process. If risk to self or others is likely to escalate, the rationale for proceeding must be robust, and contingency plans must be clearly defined and agreed upon in advance.

Shared decision-making and documentation

One of the central pillars of best practice is shared decision-making. The process must go beyond simple consent and involve a meaningful exploration of patient goals, fears, and values. NICE, RCPsych, and GMC all emphasise the ethical and legal imperative of involving patients in treatment decisions, particularly where those decisions deviate from usual practice or carry known risk.

Patients must be presented with a balanced picture—benefits of potential drug-free states weighed against relapse, withdrawal, and diagnostic uncertainty. Where patients request discontinuation, clinicians should carefully explore motivations, and where appropriate, offer alternative strategies to manage side effects or concerns. In cases where the clinician recommends withdrawal, clear explanations for the rationale and risk must be recorded.

High-quality documentation is both a clinical and legal safeguard. Records should include: rich details of discussions held, capacity assessments undertaken, informational materials provided, options considered, and the agreed plan. Where carers or advocates are involved, their views should be noted. In forensic or complex cases, this documentation may later be reviewed by courts, inquiry panels, or professional regulators.

Individualised tapering and monitoring plans

A key risk mitigation measure is the adoption of slow, personalised tapering protocols. NICE guideline NG215 highlights that a one-size-fits-all approach to withdrawal is inappropriate, particularly for medications with long half-lives or severe discontinuation effects. Dosage reduction schedules must consider pharmacokinetics, prior dependence, and tolerability.

Gradual tapering over weeks or months—with flexibility to pause or reverse at early signs of distress—is now widely endorsed. For antidepressants and benzodiazepines, reductions of 10% of the previous dose every two to four weeks are increasingly advocated in specialist literature. For antipsychotics, reductions must also account for receptor occupancy dynamics, especially at lower doses where withdrawal effects may intensify.

Monitoring during withdrawal must be active, not passive. This may include scheduled clinical reviews, symptom checklists, family observations, and—where available—digital tools that track behavioural or physiological changes. All plans should include clear criteria for early reinstatement and steps for rapid reassessment should symptoms return. In forensic settings, this may include stepped observation levels or risk management team reviews.

Summary and takeaway points

This article presents a structured and clinically-grounded exploration of psychiatric medication discontinuation, particularly within UK practice. It opens by recognising that while some patients express a desire to reduce or stop medication, the process carries significant clinical, ethical, and legal weight. There was particular emphasis on planned and clinician-led tapering, distinguishing it from non-adherence or abrupt patient-led discontinuation.

A central theme is the risk of relapse, which varies substantially depending on the speed and structure of withdrawal. Evidence from meta-analyses confirms that abrupt cessation carries the highest relapse rates—up to 77%—while gradual tapering over more than 10 weeks significantly reduces this risk to around 31%. This data underpins the argument that tapering should be slow, individualised, and closely monitored, with early reinstatement plans in place.

Medico-legal context, highlighted the obligations under the Mental Capacity Act 2005 and the Mental Health Act 1983. It explains that decisions around discontinuation require lawful consent processes in accordance with Montgomery, and that risk must be reasoned and defensible under the Bolitho test—not merely in line with professional consensus (Bolam). The risk of causing psychiatric deterioration through diagnostic withdrawal is flagged as a potential source of legal liability if not justifiable, proportionate, and clearly documented.

Guidance from professional bodies, such as NICE and the Royal College of Psychiatrists, is drawn upon to reinforce best practice.

In forensic contexts, risks extend beyond the patient to others. Medication changes in patients with histories of violence, detention, or high-risk psychosis must be approached with enhanced scrutiny. The use of structured risk tools and inter-agency communication is encouraged to pre-empt escalation.

Finally, shared decision-making, capacity assessments, and detailed documentation are encouraged. Discontinuation for drug holiday, is not a neutral act.

1. Psychiatry claims under the spotlight (MDU, Aug 2023)