Stopping Bipolar Medication: Withdrawal, Relapse, and Safer Planning
Author: Ukrainian Psychological Hub · Published: October 9, 2026 · Editorial Policy
Stopping a medication used to treat bipolar disorder can be an appropriate clinical decision in some circumstances, including significant adverse effects, pregnancy planning, a carefully reviewed change in treatment, or a reassessment of which medicines are still necessary. The decision has consequences that extend beyond the last dose. Some people experience symptoms associated with drug withdrawal or physiological adaptation; some experience the return of mania, hypomania, depression, or mixed symptoms; and some remain well after a carefully managed change. These outcomes require different responses and cannot be predicted from a single symptom or an online checklist.
The central safety principle is to plan any change with the prescribing clinician, review the entire medication regimen, and arrange monitoring before reducing treatment. Stopping medication suddenly may increase the risk of a new mood episode, particularly after long-term lithium treatment. The National Institute of Mental Health advises people with bipolar disorder to discuss medication concerns with a health professional rather than stopping treatment independently; NICE guidance CG185 provides drug-specific discontinuation and follow-up recommendations.
For additional clinical context, see Treatment-Resistant Bipolar Disorder: Reassessment and Care Options and Electroconvulsive Therapy for Bipolar Disorder: When It Is Considered.
The short answer: what to do before stopping bipolar medication
Before a change, identify exactly which medication you take, what phase of bipolar illness it was prescribed for, whether it prevents recurrence, the reason you want to stop, and what happened during previous reductions or missed doses. Contact the prescribing clinician to discuss the concern, including intolerable adverse effects, financial or access barriers, uncertain benefit, pregnancy, or a desire to simplify treatment. The next step is an individualized written plan with an agreed taper, replacement treatment if appropriate, follow-up dates, early-warning signs, and an urgent-care route. NICE specifically recommends shared decisions about long-term treatment and close observation after stopping it.
For a wider comparison of continuing treatment and changing drug classes, read the bipolar disorder treatment guide. The present article concentrates on discontinuation decisions, withdrawal, relapse detection, and coordination with a prescriber.
A medication-related emergency follows a different pathway from an elective taper. Severe rash, possible lithium toxicity, marked confusion, fainting, seizures, or other serious adverse reactions call for prompt medical assessment, sometimes emergency services. The clinician may need to hold or stop a medicine immediately for safety. The principle of gradual withdrawal never justifies continuing a drug when its immediate harm requires urgent intervention. NICE medication guidance and the U.S. LAMICTAL prescribing information describe drug-specific serious reactions requiring urgent attention.
Why stopping treatment can change the course of bipolar disorder
Bipolar I disorder is defined by at least one manic episode; bipolar II disorder includes hypomanic and major depressive episodes, without a history of full mania. Long-term patterns vary considerably, and clinically stable periods may be sustained by treatment. The appearance of wellness alone does not establish that protective medication has become unnecessary. Bipolar II disorder should not be assumed to be a milder condition, particularly when depression is frequent or disabling. The CANMAT/ISBD 2023 evidence update and the 2023 VA/DoD clinical practice guideline organize treatment by phase and recurrence risk.
Maintenance treatment aims to reduce the likelihood of future episodes, hospitalization, disruption to work and relationships, and other consequences of relapse. Its evidence base differs from the evidence for controlling an acute manic or depressive episode. For example, a drug used briefly during acute mania may later be reviewed for discontinuation while another medication continues for relapse prevention. Conversely, a medication that helped one depressive episode may have been selected partly for long-term prevention of depression. NICE calls for regular review so medicines no longer needed after the acute phase are stopped when appropriate, without abandoning relapse-prevention planning.
What research shows about relapse after medication discontinuation
A systematic review and meta-analysis by Kishi and colleagues pooled 22 randomized discontinuation-versus-maintenance studies including 5,462 participants. At six months, continued mood stabilizer or antipsychotic treatment was associated with lower recurrence of any mood episode than discontinuation or placebo substitution (risk ratio 0.61, 95% confidence interval 0.54–0.70). Results also favored maintenance for manic, hypomanic, mixed, and depressive recurrence. These are averages across several drugs and study designs, not an individual prediction. About 47% of people in discontinuation groups had no recurrence during the first six months, while others relapsed; the finding supports personalized risk assessment rather than either universal continuation or universal stopping.
Discontinuation trials may reflect both the removal of an effective preventive intervention and drug-specific effects of withdrawal. Some trials enroll people who responded to a medication and then randomly continue or replace that treatment, which can increase apparent differences between groups and does not reproduce every real-world taper. A 2024 cross-class evidence synthesis discusses this problem directly. Nevertheless, the overall direction of controlled maintenance evidence makes planned monitoring and continuity of care especially important.
Withdrawal, rebound, and relapse: three different problems
Discontinuation or withdrawal symptoms
Withdrawal describes symptoms that arise after the body has adapted to a medication and its exposure is reduced. Depending on the drug, symptoms can include nausea, dizziness, insomnia, restlessness, sweating, anxiety, irritability, or changes in movement. Timing is often related to the last dose and drug half-life, although it varies widely. Withdrawal is not proof of addiction. Many psychiatric medications can produce physiological adaptation even when used correctly under medical supervision. A clinical review of psychotropic discontinuation and the 2024 synthesis describe the variability of withdrawal reactions.
Rebound or early clinical worsening
Rebound is used when an increase in symptoms follows treatment removal and may be more abrupt or pronounced than the patient's ordinary course would predict. For lithium, unusually early recurrence after rapid discontinuation has been repeatedly documented, although distinguishing pharmacological rebound from recurrence caused by loss of preventive treatment can be difficult. This distinction matters for interpreting studies, but it does not change the clinical need to evaluate significant symptoms promptly. See the prospective Faedda and colleagues lithium study and the 2024 synthesis.
A new bipolar mood episode
Relapse or recurrence means a clinically significant new mood episode or return of active illness. Mania can involve persistently increased energy and activity, reduced need for sleep, pressured speech, expansive or irritable mood, impaired judgment, and sometimes psychotic symptoms. Hypomania has a distinct duration and impairment profile; severe depression involves sustained symptoms and functional change. Mixed features combine symptoms of opposing mood poles within an episode. A few hours of mood variability are insufficient to establish any of these diagnoses. Assessment considers duration, baseline, functioning, substances, medication history, sleep, and clinical context. See NIMH and VA/DoD 2023.
Why timing alone cannot make the diagnosis
Symptoms appearing soon after a dose reduction may be withdrawal effects, emerging mood symptoms, a medical problem, substance effects, an interaction, or several processes simultaneously. Insomnia is particularly difficult to interpret: it can occur during medication withdrawal, during stress, or as an early sign of mania. The clinician examines onset and pattern, changes in energy and mood, return of the person's characteristic episode features, impairment, and the medication's pharmacology. Do not restart an interrupted prescription, double a missed dose, or adjust a taper merely to test a hypothesis without drug-specific guidance. NICE recommends individualized monitoring and relapse planning.
Why abrupt discontinuation can be risky
Rapid removal of a protective drug can leave a person without adequate relapse prevention while exposing them to discontinuation effects. In a prospective study of 64 people previously stable on lithium, Faedda and colleagues found substantially earlier and more frequent recurrence following rapid versus more gradual discontinuation. The study is older and observational in its taper comparisons; it establishes an important safety signal, not a reliable numerical estimate for every modern medication regimen. A separate review by Baldessarini and colleagues summarizes evidence that sudden discontinuation of lithium, antidepressants, and antipsychotics can be followed by earlier clinical worsening.
Patients may wish to stop suddenly because of fatigue, weight change, cognitive dulling, tremor, sexual side effects, laboratory abnormalities, pregnancy concerns, or frustration after a symptom-free period. These concerns warrant clinical action, and sometimes a different treatment is substantially more suitable. The safest response is to address the specific burden rather than assuming the options are simply to tolerate it or remove treatment without a replacement plan. The NICE long-term treatment recommendations emphasize collaborative review of benefits, adverse effects, previous episode severity, and alternatives.
Drug-by-drug considerations when treatment changes
Lithium: recurrence risk, gradual reduction, and toxicity
Lithium has strong evidence for long-term relapse prevention in bipolar disorder and requires ongoing measurement of blood lithium levels and renal and thyroid function. During ordinary planned discontinuation, NICE recommends reducing the dose over at least four weeks and preferably up to three months, even when another antimanic medication has already begun. NICE also recommends close monitoring during reduction and for three months afterward, with ongoing longer-term monitoring after all treatment stops. This is a professional guideline framework, not a self-directed schedule; treatment duration, kidney function, serum levels, comedications, and prior relapse history change clinical planning.
A person taking lithium needs separate instructions about dehydration, illnesses with vomiting or diarrhea, certain diuretics, nonsteroidal anti-inflammatory drugs, and interacting medicines, because toxicity can develop even in someone taking a familiar prescription. Severe tremor, vomiting, diarrhea, confusion, poor coordination, and altered consciousness require urgent evaluation for possible lithium toxicity. These circumstances may require immediate medical decisions rather than following a routine slow taper. NICE medication and monitoring recommendations provide the clinical framework.
Lamotrigine: interrupted treatment and the restart problem
Lamotrigine is commonly used in bipolar maintenance treatment and has greater relevance to depressive-episode prevention than treatment of acute mania. NICE recommends a gradual reduction over at least four weeks when stopping it to minimize relapse risk. When treatment has been interrupted, restarting may require a new slow initiation schedule because severe rash, including Stevens–Johnson syndrome, is a recognized risk. The official U.S. LAMICTAL labeling directs prescribers to consider restarting the initial regimen after an interruption and generally to follow initial dosing recommendations after more than five drug half-lives; other medications change those half-lives. A person who missed multiple days should ask their prescriber or pharmacist before resuming an old full dose.
A new rash during lamotrigine treatment, particularly with blisters, mouth sores, fever, or systemic symptoms, needs immediate medical assessment. Prior lamotrigine-associated rash makes restarting a separate specialist decision under the product labeling. Interactions with valproate affect lamotrigine exposure and titration. These safety issues cannot be reduced to a generic 'cut tablets in half' instruction. DailyMed prescribing information.
Valproate or divalproex: relapse planning and reproductive safety
Valproate is used for particular bipolar presentations but has major reproductive, liver, hematologic, and other safety considerations. NICE recommends tapering valproate over at least four weeks when changing to an alternative medication to minimize relapse risk; it also calls for careful review of serious adverse reactions and pregnancy-related restrictions. The exact regulatory rules differ by jurisdiction. When pregnancy is possible or planned, a clinician should assess whether an alternative is appropriate before conception where feasible. Because valproate exposure in pregnancy can harm fetal development, prompt specialist review is important; abruptly removing effective treatment can also create substantial psychiatric risk. WHO's maintenance guidance strongly advises against valproate for women and girls of childbearing potential with bipolar disorder in remission.
Valproate may also be prescribed for epilepsy, and the same individual may have more than one condition. A change made solely for bipolar symptoms may affect seizure control. Liver disease signs, unexplained bleeding or bruising, severe abdominal pain, or other serious reactions demand urgent professional assessment. Decisions to stop valproate because of major safety concerns may be faster than standard elective tapering and should be directed by an appropriate specialist. NICE's valproate safety recommendations.
Antipsychotics: recurrence and discontinuation effects
Several second-generation antipsychotics are used during acute mania, bipolar depression, or maintenance, but individual agents have different evidence and approvals. Stopping a medication used for short-term acute control while retaining effective maintenance treatment is clinically different from withdrawing the sole relapse-prevention drug. NICE recommends reducing antipsychotics gradually over at least four weeks when stopping, partly to reduce relapse risk. Individual plans may require longer or more complex changes depending on prior dose, treatment duration, clinical state, and adverse effects; a universal taper rate is not established for all antipsychotics in bipolar disorder.
Depending on the agent, a reduction may lead to insomnia, agitation, nausea, movement symptoms, or other discontinuation effects, which can be mistaken for returning illness. Rebound psychosis or mania also requires careful evaluation. Continuing an antipsychotic can involve metabolic, cardiovascular, sedation, and movement-disorder burdens, so its ongoing indication should be reassessed with attention to alternatives. NICE provides monitoring recommendations and the 2024 cross-class review examines withdrawal effects.
Antidepressants: depressive relapse, discontinuation syndrome, and switching risk
Antidepressants are not interchangeable with mood stabilizers. Some people with bipolar depression receive an antidepressant as an adjunct under specific clinical circumstances, whereas antidepressant monotherapy can be problematic, particularly in bipolar I disorder. Abrupt antidepressant cessation may produce dizziness, sensory disturbances, flu-like symptoms, anxiety, or insomnia; taper planning must consider the particular medicine and its half-life. At the same time, the clinician weighs the risk of a return of depression and possible treatment-emergent hypomania or mania. CANMAT/ISBD 2023 describes indications and cautions, while the psychotropic discontinuation review distinguishes withdrawal from later recurrence.
A decision to discontinue an antidepressant after mania emerges differs from an elective taper during stable remission. The priorities change when a patient is acutely manic, suicidal, or severely depressed; prompt clinician assessment can outweigh the usual timetable. Do not extrapolate withdrawal rules for antidepressants to lithium or lamotrigine, or vice versa. NICE's acute mania and long-term medication sections distinguish these clinical circumstances.
Several medications, long-acting injections, or other conditions
Many people are prescribed more than one agent because manic and depressive episodes respond differently, and because a regimen may change after hospitalization. Before reducing anything, the clinician should reconstruct the complete list of prescription drugs, as-needed medicines, over-the-counter products, supplements, alcohol and other substances. Long-acting injectable antipsychotics require a different pharmacokinetic timeline from daily tablets, and epilepsy, cardiac disease, kidney disease, or thyroid disease can constrain changes. Avoid altering several medications at once without an explicit supervised rationale because it becomes difficult to identify which change caused a problem. VA/DoD 2023 and NICE support coordinated phase-specific management.
A practical safer-planning framework for a clinician appointment
1. Establish the reason and the exact treatment goal
State the problem clearly: 'I have severe fatigue,' 'My blood tests changed,' 'I am considering pregnancy,' 'I cannot afford this prescription,' or 'I do not understand whether I still need two medications.' Document how often the problem occurs, what it prevents you from doing, and what remedies you have tried. The objective may be lowering a burdensome dose, switching agents, stopping a short-term medication, or changing an entire maintenance strategy. These choices have different levels of risk. Shared decision-making is part of the NICE guideline and the VA/DoD care framework.
2. Review the longitudinal illness history
Bring a timeline of past manic, hypomanic, depressive, and mixed episodes, including approximate dates, hospitalizations, psychotic features, changes in functioning, suicide risk, sleep disruption, substance use, postpartum episodes, and prior changes in treatment. The number, severity, timing, and polarity of previous episodes help inform risk. Record how long remission has lasted and whether residual symptoms remain. Stability during effective treatment is meaningful, but it cannot by itself prove that medication is no longer protective. NICE long-term care recommendations.
3. Map the medicines and medical monitoring
Write down the exact prescription name, formulation, prescribed strength, actual use, last dose, prescriber, indication, notable adverse effects, and any recently missed doses. Ask whether blood levels, kidney or thyroid tests, liver tests, complete blood counts, weight, glucose, lipids, blood pressure, or electrocardiography are relevant to the drug being continued or changed. If a drug was prescribed for seizures or another condition as well as bipolar disorder, tell every relevant clinician. NICE's monitoring recommendations identify drug-specific assessments.
4. Agree on a supervised transition, not a self-made schedule
The written plan should name the medication to change, the clinical rationale, whether another treatment is being introduced or maintained, who will authorize each adjustment, what to do after missed doses, and who to contact about withdrawal symptoms. Its timing must be individualized. The NICE four-week minimum for certain medicines is a guideline recommendation, not permission for patients to calculate their own reductions. Complex histories, high relapse risk, drug interactions, previous withdrawal difficulties, pregnancy and urgent adverse effects can require a substantially different approach. NICE CG185.
5. Plan follow-up before the first reduction
Arrange review appointments and decide how the clinical team will monitor sleep, mood, energy, activity, judgment, depressive symptoms, adverse effects, safety, and functioning. Ask whether family members or trusted supporters may notice early changes and what the patient consents to share. NICE advises close follow-up during lithium dose reduction and for three months after it stops, and continued monitoring of mood and mental state for two years after all pharmacological treatment ends. These durations are guidance benchmarks rather than guarantees that risk ends afterward.
6. Make an individualized action plan for recurrence
Identify the patient's usual earliest signs of mania or depression, the clinician's after-hours contact, the crisis plan, and concrete decisions about driving, major spending, work, sleep, and access to support if judgment changes. Include steps for a supporter who notices prolonged reduced need for sleep, escalating impulsivity, severe depression, or psychotic symptoms. A safety plan is more useful when written while judgment is stable, revisited collaboratively, and linked to professional care. NICE relapse-prevention recommendations describe this process.
How to monitor early warning signs without treating every change as relapse
For a companion explanation of patterns that may precede another mood episode, see early warning signs of mania and bipolar relapse. This can help when building a personal monitoring plan, while clinical diagnosis still requires professional assessment.
A simple daily record can include time asleep, perceived need for sleep, changes in activity and speed of thought, irritability, anxiety, social contact, depressive symptoms, prescribed doses, and significant life events. The aim is to notice sustained departures from the individual's baseline. A single poor night of sleep, occasional frustration, or a positive mood does not establish mania. Conversely, consistently sleeping much less without fatigue, rapid acceleration in projects and spending, mounting agitation, or a return of characteristic delusions can require urgent assessment. NIMH and VA/DoD 2023 describe clinically significant episode patterns.
A mood chart can help the patient and clinician examine the sequence of dose changes and symptoms, but it cannot reliably diagnose withdrawal versus relapse. It also cannot replace clinical assessment of suicidality, psychosis, mixed symptoms, or functional impairment. The practical question is whether a notable pattern is emerging and which response the treating team agreed upon. If the pattern changes sharply or becomes unsafe, contact the treating service rather than waiting until the next scheduled review. NICE emphasizes individualized early-warning and risk-management planning.
Special circumstances that can change the decision
Pregnancy, conception, and the postpartum period
Pregnancy planning requires both reproductive medication safety and protection against severe mood recurrence. Stopping all medication suddenly after a positive pregnancy test can expose the patient to substantial risk; continuing particular drugs, especially valproate, can expose the fetus to substantial harm. The right decision depends on the exact medication, clinical history, reproductive plans, and access to specialist perinatal psychiatry. In a prospective cohort of 89 pregnant women with bipolar disorder, Viguera and colleagues found more recurrence after discontinuing mood stabilizers than after maintaining them; the observational study supports careful planning but cannot prescribe an individual choice. NICE and WHO provide related safety guidance.
The postpartum period can be a particularly vulnerable time for severe bipolar mood episodes. Planning should address sleep, infant-care support, breastfeeding and medication compatibility, and urgent assessment of manic, psychotic, or suicidal symptoms. Valproate restrictions and the relative reproductive risks of specific alternatives depend on current jurisdiction and specialist advice. For planned pregnancy, arrange a medication review before conception where feasible; for an unplanned pregnancy, seek prompt individualized assessment rather than modifying treatment alone. NICE CG185 discusses pregnancy-related treatment planning.
Children, adolescents, and older adults
Young people require specialist assessment because diagnosis, drug indications, development, and adverse-effect risks differ from those of adults. In older adults, kidney function, falls, interactions, cognitive effects, and other conditions may make an apparently familiar regimen less suitable. A caregiver should not taper a child's or older adult's psychiatric medication based on advice written for a general adult population. NICE provides distinct recommendations for children and young people and notes older-adult prescribing considerations.
Financial barriers, side effects, and trouble taking medication consistently
Missed prescriptions caused by price, pharmacy shortages, work schedules, stigma, sedation, or difficult monitoring requirements deserve a practical care plan. Ask about generic formulations, an alternative treatment with a manageable monitoring schedule, coordinated refill timing, patient assistance options, and a documented plan if access is interrupted. Clinicians should approach nonadherence as information about the treatment's real-world feasibility, not merely a personal failing. There is a clinical difference between a negotiated switch and repeated unplanned gaps in treatment. VA/DoD 2023 and NICE consider engagement, tolerability, and adherence in treatment choices.
When to seek urgent help
Seek urgent clinical evaluation for escalating mania, prolonged sharply reduced need for sleep accompanied by marked behavioral change, psychosis, rapidly worsening depression, inability to care for basic needs, or behavior creating immediate danger. Active suicidal intent, a plan or access to means, severe confusion, seizures, suspected medication toxicity, or a serious drug reaction can warrant emergency services. If someone is in immediate danger, contact the local emergency number. In the United States, people in suicidal crisis or needing emotional support can call or text 988; the 988 Suicide & Crisis Lifeline offers 24-hour services. A crisis helpline is an adjunct to, not a substitute for, emergency medical assessment where needed.
If symptoms occur after a missed dose or reduction, tell the service precisely what changed and when. Bring medication packages or a written list if possible. Avoid driving when impaired and involve a trusted person when judgment or safety is deteriorating. If a rash or suspected toxicity occurs, do not wait for a routine psychiatry appointment. The immediate clinical priority is to assess safety and the likely cause of symptoms; only then can the treatment plan be revised. NICE safety and risk-management guidance.
What can replace a medication that is poorly tolerated?
Depending on the phase of illness and previous response, a clinician may consider another maintenance medicine, a change in combination therapy, a different formulation, dose adjustment, or psychological and recovery-oriented interventions alongside appropriate pharmacotherapy. Lithium, lamotrigine, divalproex, and specific antipsychotics do not have identical effects on acute mania, bipolar depression, and relapse prevention. Recommendations depend on bipolar I versus II disorder, symptoms, comorbidities, pregnancy potential, and individual risks. CANMAT/ISBD 2023 and VA/DoD 2023 set out phase-specific treatment options.
Psychotherapy, family interventions, routines, and attention to sleep and substance use can strengthen relapse prevention and support functioning. For people who need pharmacological maintenance, psychotherapy does not automatically replace that protection. The NIMH overview describes how psychosocial care can complement medication; WHO maintenance recommendations emphasize balancing effects, harms, and personal preference. The right goal is sustainable treatment with a burden the person understands and can participate in managing.
Frequently asked questions
Can I stop bipolar medication because I feel normal?
Feeling well is a good reason to review treatment goals and side effects, but it does not show that relapse prevention is unnecessary. In bipolar disorder, wellness can be partly sustained by the medicine itself. Review your episode history, current risks, and options with the prescribing clinician before making a change. Controlled maintenance studies show lower average recurrence with continued medication than with discontinuation, while individual outcomes vary. Kishi et al..
How long does withdrawal from bipolar medication last?
There is no single duration. Symptoms depend on the medication, how long it was taken, the nature of reduction, individual physiology, and what is actually causing the symptoms. Some discontinuation effects begin within days; a bipolar recurrence may arise later. Persistent or severe symptoms require assessment instead of simply waiting out a presumed withdrawal. Drug-specific monitoring and medication history are essential. Discontinuation evidence synthesis.
Is tapering over four weeks always enough?
No. NICE uses at least four weeks for planned discontinuation of lithium, antipsychotics, lamotrigine, and certain valproate transitions, and it prefers up to three months for lithium. Those are clinical guidelines with different contexts, not individualized minimum-safe formulas. Some patients need a slower, more closely supervised transition; serious adverse reactions may require urgent changes instead. NICE CG185.
What if I accidentally stopped or ran out of medicine?
Contact the prescribing team or pharmacist promptly, especially after several missed doses, and report the exact medicine, formulation, duration of interruption, and current symptoms. Do not double doses to catch up or automatically restart an old high dose. Lamotrigine deserves special attention because after a meaningful interruption its restart may require an initial titration schedule to reduce serious rash risk. LAMICTAL U.S. prescribing information.
Can bipolar disorder be managed without medication?
Some individuals may undergo clinically supervised medication reduction or discontinuation after a detailed review, and not everyone who stops treatment immediately relapses. However, recurrence is common in bipolar disorders, and randomized maintenance evidence generally favors continued effective medication. Psychotherapy, social rhythms, and support remain important whether or not a medication change is planned. The specific choice depends on illness severity, prior episodes, response, treatment harms, and the person's informed preferences. Kishi et al. and WHO.
If symptoms return, should I restart my previous dose?
Contact the treatment team for prompt assessment and a drug-specific plan. Symptoms may represent withdrawal, recurrence, an interaction, or a medical condition, and the previous dose may be unsafe after an interruption. Lamotrigine often requires retitration after a substantial gap; lithium may require clinical review and blood monitoring. Do not use old instructions as a universal restart rule. DailyMed and NICE.
Does stopping lithium make it impossible to benefit from lithium again?
Most people who resume lithium can still obtain benefit, but response varies and the clinical history matters. The evidence about a distinct phenomenon of nonresponse after discontinuation is limited and mixed. The better-established concern is an increased risk of early relapse when long-term lithium is stopped quickly. Review prior benefit and adverse effects with a psychiatrist rather than interpreting a recurrence as proof that lithium has permanently stopped working. 2024 discontinuation synthesis and Faedda et al..
Should I stop medication if I discover I am pregnant?
Seek prompt review with the prescribing clinician and an obstetric or perinatal specialist. Some medicines pose important fetal risks, particularly valproate, while abrupt removal of effective maintenance care may increase risk of serious mood recurrence. The correct decision is drug-specific and time-sensitive. Do not rely on blanket internet rules about either stopping or continuing every medication. NICE, WHO, and Viguera et al..
Making a decision that remains safe after the last dose
A sound stopping plan answers five questions in advance: why is a change needed, what protection will remain, what symptoms would prompt a review, who is responsible for monitoring, and how can care be reached quickly if circumstances change? It also leaves room for the person's preferences, prior experiences, and treatment burden. Medication decisions deserve periodic reassessment, and a decision to continue treatment can be revisited as new evidence or clinical circumstances arise. NICE recommends long-term follow-up after discontinuation; CANMAT/ISBD emphasizes phase-specific prevention and tolerability.
Related Articles
References
Canadian Network for Mood and Anxiety Treatments and International Society for Bipolar Disorders (CANMAT/ISBD). (2023). The CANMAT and ISBD Guidelines for the Treatment of Bipolar Disorder: Summary and a 2023 Update of Evidence. Focus. Journal article.
DailyMed / U.S. National Library of Medicine. (2025). LAMICTAL (lamotrigine), U.S. prescribing information. Official label.
Faedda, G. L., Tondo, L., Baldessarini, R. J., Suppes, T., & Tohen, M. (1993). Outcome after rapid vs gradual discontinuation of lithium treatment in bipolar disorders. Archives of General Psychiatry, 50(6), 448–455. DOI: 10.1001/archpsyc.1993.01820180046005.
Kishi, T., Matsuda, Y., Sakuma, K., Okuya, M., Mishima, K., & Iwata, N. (2021). Recurrence rates in stable bipolar disorder patients after drug discontinuation v. drug maintenance: A systematic review and meta-analysis. Psychological Medicine, 51(15), 2721–2729. DOI: 10.1017/S0033291720003505.
National Institute for Health and Care Excellence (NICE). (2014; updated). Bipolar disorder: Assessment and management (CG185). Guideline recommendations.
National Institute of Mental Health (NIMH). (n.d.). Bipolar disorder. Official clinical information.
Tondo, L., & Baldessarini, R. J. (2020). Discontinuing psychotropic drug treatment. BJPsych Open, 6(2), e24. PubMed and DOI recordDOI: 10.1192/bjo.2020.6 (PubMed).
U.S. Department of Veterans Affairs & U.S. Department of Defense. (2023). VA/DoD clinical practice guideline for the management of bipolar disorder. Clinical guideline portal.
Viguera, A. C., Whitfield, T., Baldessarini, R. J., et al. (2007). Risk of recurrence in women with bipolar disorder during pregnancy: Prospective study of mood stabilizer discontinuation. American Journal of Psychiatry, 164(12), 1817–1824. DOI: 10.1176/appi.ajp.2007.06101639.
Vinkers, C. H., Kupka, R. W., Penninx, B. W., et al. (2024). Discontinuation of psychotropic medication: A synthesis of evidence across medication classes. Molecular Psychiatry, 29, 2575–2586. DOI: 10.1038/s41380-024-02445-4DOI: 10.1038/s41380-024-02445-4.
World Health Organization (WHO). (2023). Antipsychotics and mood stabilizers for maintenance treatment of bipolar disorder. WHO evidence center.
World Health Organization (WHO). (2023). Duration of maintenance treatment with antipsychotics or mood stabilizers in individuals with bipolar disorder. WHO evidence center.