Staying Well Through Pregnancy and the Postpartum Weeks

A practical, evidence-based guide for people with bipolar disorder, schizophrenia, or schizoaffective disorder.

For some people, the weeks after birth carry the highest relapse risk of their lives. That is clearest for people with bipolar I disorder or a previous postpartum psychosis. Across 37 studies of people with bipolar disorder, a history of postpartum psychosis, or both, about 35 in 100 deliveries were followed by a relapse. Studies followed people for different lengths of time, from four weeks to a year, most often three to six months, so that figure describes a range of windows rather than one. Among the bipolar groups, pooled relapse was 66 in 100 when people had been medication-free during pregnancy, versus 23 in 100 when preventive medication was used. Those numbers do not prove that medication alone caused the difference, because the studies were not randomized. They do show why a postpartum plan should be made before labor, not after symptoms arrive.[1] Evidence is thinner when bipolar II disorder or schizoaffective disorder is studied on its own.

Postpartum psychosis is uncommon. ACOG describes it as occurring in 1 to 2 per 1,000 pregnancies, and notes that most people hospitalized with it had no known psychiatric history.[7,33] The risk is much higher after bipolar I disorder or a previous postpartum psychosis. Onset is usually sudden: an expert consensus review reports that most people develop symptoms within the first two weeks after delivery, with a median around day 8 to 10.[2] Warning signs can include almost no need for sleep, rapidly increasing energy, agitation, racing or disorganized thoughts, confusion, paranoia, hallucinations, or beliefs that feel unquestionably real but are out of character. Untreated postpartum psychosis raises the risk of suicide and, rarely, infanticide, but most affected parents do not harm anyone. This is a medical emergency, not a parenting failure.

If these changes appear, treat them as urgent. ACOG recommends that clinicians provide immediate medical attention for postpartum psychosis.[33] Call the prescriber or obstetric clinician the same day and name what you are seeing. Postpartum Support International advises that someone stay with the parent and the baby to keep everyone safe while help is arranged.[34] If there are thoughts of suicide or of harming the baby, call or text 988. If anyone is in immediate danger, call 911.[34]

Start with the background risk

No pregnancy begins at zero risk. A current U.S. medicine label gives estimated general-population background ranges of 2 to 4 major birth defects and 15 to 20 miscarriages per 100 clinically recognized pregnancies.[3] Those are context ranges, not a matched no-medication comparison for any particular study. A medicine study should compare the same outcome, during the same time window, in similar groups. An “exposed risk” without a valid comparator can be frightening but misleading.

Medication decisions also have two patients and two kinds of risk: possible effects of treatment, and effects of an untreated or relapsing illness. Severe mania, depression, psychosis, poor sleep, inability to eat or attend prenatal care, substance use, suicide risk, and hospitalization can all affect the pregnant person, the fetus, the infant, and the family. A sound plan is rarely “take everything” or “stop everything.” It weighs the pregnant person’s health together with fetal and newborn outcomes, and aims to preserve stability while reducing overall risk.

Relapse risk comes before a medication list

In a prospective study of 89 pregnant people with bipolar disorder who began pregnancy well, 71 in 100 had a mood recurrence. Those who discontinued a mood stabilizer had earlier and more frequent recurrences, and abrupt or rapid discontinuation was linked to a much shorter time to illness than gradual discontinuation.[4] This was a small, nonrandomized specialist cohort, so it cannot predict one person’s outcome. It does support a firm safety rule: do not abruptly stop lithium, an antipsychotic, lamotrigine, valproate, or another psychiatric medicine because of a positive pregnancy test. Call the prescriber promptly and make a supervised plan.

Evidence is also accumulating for psychotic disorders. In a 2026 Danish and Swedish study, people with a primary psychotic disorder who stopped antipsychotics during pregnancy had more severe relapse from pregnancy through three months postpartum than those who continued, with an adjusted hazard ratio of 1.60 and a 95% confidence interval of 1.01 to 2.54. The bipolar estimate in the same study was 1.00, with a confidence interval from 0.48 to 2.10, too imprecise to show a difference.[5] A Korean study of 3,026 women with schizophrenia found a smaller postpartum rise in hospital admission among those who continued antipsychotics than among those who stopped.[6] These are observational studies built on prescription records and hospital events, not proof that continuation is best for everyone. They do reinforce the importance of prior relapse history, medication response, adherence, and illness severity.

Before changing treatment, ask: What happened the last time this medicine was reduced? Have there been manic, mixed, psychotic, suicidal, catatonic, or postpartum episodes? How quickly did past episodes become dangerous? Which medicine actually kept you well? A person with one mild remote episode has a different risk balance from someone with repeated hospitalizations or a previous postpartum psychosis.

If you are planning pregnancy, choose a stable time when possible. Ask whether the regimen can be simplified without losing control, how the prescriber defines the lowest effective regimen for you, and how long stability should be observed before trying to conceive. Use reliable contraception until planned changes are complete. Some antipsychotics raise prolactin and can affect ovulation, but reduced fertility is not contraception.[11,12]

How to read a medication label now

The old FDA pregnancy letters A, B, C, D, and X are obsolete. Current prescription labels use narrative sections for pregnancy risk, clinical considerations, human and animal data, lactation, pregnancy registries, contraception, and fertility.[8] “No signal found” does not mean “proved safe,” especially for newer drugs studied in small numbers. An association does not automatically mean the drug caused the outcome. The most useful evidence combines absolute numbers, an appropriate psychiatric comparison group, adjustment for illness and other factors, and a confidence interval showing the uncertainty.

Lithium: effective, monitorable, and not risk-free

Lithium deserves careful discussion because it can be an effective maintenance treatment for bipolar disorder, while pregnancy changes both its blood level and its risk profile.[11]

The strongest U.S. malformation study compared 663 first-trimester lithium-exposed live births with more than 1.3 million unexposed births. Cardiac malformations occurred in 16 of 663 lithium-exposed infants, or 2.41 in 100, and in 15,251 of 1,322,955 unexposed infants, or 1.15 in 100. After adjustment, the estimated difference was 0.95 additional cardiac malformations per 100 births, with a 95% confidence interval running from 0.22 fewer to 2.12 more per 100. The adjusted relative risk was 1.65, with an interval of 1.02 to 2.68. Any malformation was found in 5.73 in 100 exposed births versus 3.26 in 100 unexposed, adjusted relative risk 1.37, interval 1.01 to 1.87. Higher prescribed-dose groups had higher estimates, but those categories are not instructions for changing your own amount.[9]

It is worth knowing where the older, frightening numbers came from. A 1970s voluntary registry suggested a roughly 400-fold increase in Ebstein anomaly on the basis of two reported cases, and its final report described 6 Ebstein cases among 225 exposed infants. That registry had no control group and collected unusual cases preferentially. The modern cohort found a far smaller association.[9]

Lithium blood concentrations often fall during pregnancy as kidney filtration and body water change, then rise again after birth. One longitudinal study found the largest average fall in the second trimester and a rebound postpartum.[10] This is why the prescriber may request more blood tests and review kidney and thyroid function, hydration, vomiting, fever, interacting medicines, and signs of toxicity. It is also why changing the amount without a level and a clinical plan can create either relapse or toxicity.

Labor guidance is genuinely unsettled. The current U.S. label and UK NICE guidance take a conservative approach near delivery, while CANMAT describes routine withholding as controversial. A small newer cohort found no sudden maternal level rise around delivery and no detected relationship between neonatal lithium levels and neonatal outcomes.[11,12,13,35] That cohort was too small to rule out uncommon harm. Your psychiatrist, obstetric clinician, anesthesiology team, and newborn clinician should agree on a written plan before labor. This article deliberately gives no stop, restart, or dosing schedule.

Breastfeeding while taking lithium is also not a universal yes or no. The U.S. product label states that breastfeeding is not recommended with maternal lithium use, and adds that if a parent chooses to breastfeed, the infant should be closely monitored for signs of lithium toxicity.[35] UK NICE advises against breastfeeding on lithium.[12] LactMed and MotherToBaby describe carefully selected situations in which a healthy full-term infant may breastfeed with coordinated monitoring.[14,15] Infant exposure varies widely and can rise when a newborn is premature, ill, dehydrated, or feeding poorly. If lithium is the medicine that best prevents a severe relapse, discuss the feeding plan before birth with psychiatry, pediatrics, and obstetrics. Ask who will assess the infant, which symptoms require same-day review, whether infant lithium, kidney, or thyroid testing is planned, and what happens during fever, vomiting, diarrhea, dehydration, or poor intake. Formula, donor milk, combination feeding, and expressed milk are legitimate tools.

Valproate deserves its own warning

Valproate, including Depakote, divalproex, valproic acid, and sodium valproate, carries substantially greater pregnancy concern than most alternatives. Current U.S. labeling says it generally should not be used for bipolar disorder during pregnancy or pregnancy planning unless other options have failed or are unacceptable. It also calls for effective contraception in people who can become pregnant. EU and UK rules are more restrictive and require formal pregnancy-prevention measures for many patients.[16,20,21]

In the 2024 EURAP epilepsy registry, major congenital malformations were found in 153 of 1,549 valproate monotherapy pregnancies, or 9.9 in 100. Risk rose with dose in the registry, and current U.S. labeling states that a threshold dose below which no risk exists cannot be established.[16,17] These are epilepsy-registry numbers, not a bipolar-specific prediction, but they are among the clearest prospective data available. UK regulators tell patients that around 1 in 9 babies exposed to valproate in pregnancy will have a birth defect, and that about 3 or 4 children in every 10 may have problems with early childhood development.[21]

Neurodevelopment is a separate outcome. In the prospective NEAD epilepsy cohort, adjusted mean IQ at age six was 97 after valproate exposure, compared with 105 to 108 after the other studied monotherapies.[18] A Danish population cohort estimated autism-spectrum diagnoses in 4.42 in 100 valproate-exposed children versus 1.53 in 100 in the population overall, with attenuation in an epilepsy-restricted analysis.[19] Neither observational study can remove all effects of epilepsy, genetics, dose, seizures, or other family factors. Together with other evidence, they support strong avoidance and prevention policies. Folic acid is recommended generally in pregnancy, but it is not known to eliminate valproate-specific malformation or developmental risk.[16]

Do not stop valproate on your own if you are pregnant. Abrupt destabilization can be dangerous, and a safer replacement may require planning. Contact the prescriber and obstetric team urgently for an individualized transition discussion. If you can become pregnant but are not planning pregnancy, ask specifically about reliable contraception and drug interactions.

Paternal valproate guidance is evolving. In June 2026, the European Medicines Agency concluded that studies of neurodevelopment after paternal use were inconsistent and that the causal role of valproate is uncertain. It nevertheless kept its 2024 precautions in place and recommended product-information updates.[22] This is a precautionary EU position, not proof that paternal exposure causes harm, and not a U.S. rule. Anyone taking valproate who plans to father a pregnancy should ask a clinician for current local advice rather than stopping treatment.

Valproate usually produces low levels in milk, and LactMed does not treat valproate monotherapy as a reason by itself to stop breastfeeding. That lactation point does not cancel the danger of a new pregnancy while taking it. Watch a breastfed infant for jaundice or unusual bruising or bleeding, and keep contraception and future-pregnancy planning explicit.[32]

Lamotrigine, carbamazepine, and oxcarbazepine

Lamotrigine, often known as Lamictal, is commonly considered when preventing bipolar depression matters most. In the EURAP epilepsy registry, major malformations were observed in 110 of 3,584 lamotrigine monotherapy pregnancies, or 3.1 in 100.[17] That estimate is reassuring relative to valproate within the registry, but it comes from epilepsy, not bipolar disorder, and it does not prove zero risk.

Lamotrigine clearance can rise markedly during pregnancy and reverse quickly after delivery. A small bipolar series also found rapid postpartum concentration increases.[37,38] Blood levels, symptoms, and pre-pregnancy response help a clinician interpret these changes. Do not copy another person’s monitoring schedule. During breastfeeding, lamotrigine reaches milk and infant blood. LactMed advises careful monitoring of the breastfed infant for effects such as apnea, rash, drowsiness, or poor sucking, including serum levels if there is a concern.[23]

In the same registry, major malformations occurred in 5.4 in 100 carbamazepine and 2.9 in 100 oxcarbazepine monotherapy pregnancies. The oxcarbazepine group was much smaller, so its estimate was less precise.[17] UK NICE does not recommend carbamazepine for a mental health problem when planning pregnancy, during pregnancy, or while considering breastfeeding.[12] U.S. labels also warn that carbamazepine and oxcarbazepine can make some hormonal contraceptives less effective, and advise additional or alternative non-hormonal contraception.[24] Breastfeeding data are more reassuring than pregnancy data for both, but infant drowsiness, feeding, weight gain, jaundice, and development still warrant observation.[32]

Zoom

Antipsychotics: think by drug, not just by class

Antipsychotics may be essential for schizophrenia, schizoaffective disorder, mania, psychotic depression, or relapse prevention. Large adjusted data from Nordic countries and the United States do not suggest a meaningful increase in major congenital malformations for antipsychotics as a class. Individual-drug certainty is uneven, and a few specific signals remain too imprecise to dismiss or confirm.[25]

Metabolic effects matter. Among U.S. Medicaid patients already using an atypical antipsychotic, standardized gestational-diabetes risk was 11.7 in 100 for those who continued olanzapine versus 7.3 in 100 for those who stopped, an adjusted difference of 4.4 per 100 with a 95% confidence interval of 0.8 to 8.1. For quetiapine, risks were 7.1 versus 5.6 per 100, an adjusted difference of 1.6 per 100 with an interval of 0.0 to 3.1.[26] The quetiapine interval reaches zero, so that estimate is compatible with no increase. These comparisons reduce some confounding but do not prove the drugs caused every extra case. Ask how weight, nutrition, family history, and glucose will be monitored. A metabolic concern is usually a reason for surveillance and shared decision-making, not abrupt discontinuation of an effective medicine.

Current FDA labeling warns that third-trimester antipsychotic exposure can be followed by newborn agitation, unusual muscle tone, tremor, sleepiness, breathing distress, or feeding difficulty. These reactions have been reported rarely. Some infants recover quickly; others need longer observation or hospitalization.[3] Tell the delivery and pediatric teams about every medicine and the timing of the last prescribed dose. This supports newborn assessment and is not, by itself, a reason to stop treatment before birth.

Data for some individual medicines remain limited. A prospective registry reported major malformations in 2.19 in 100 lurasidone pregnancies, 1.85 in 100 quetiapine pregnancies, and 1.77 in 100 psychiatric comparison pregnancies, but the lurasidone group included only 134 pregnancies and its confidence interval was wide.[27] Corrected multinational data were reassuring for the neurodevelopmental and learning outcomes studied after prenatal antipsychotic exposure, but smaller effects, later diagnoses, and drug-specific effects cannot be excluded.[28]

Long-acting injectable antipsychotics can protect people whose illness worsens when doses are missed. Pregnancy evidence is thin: a 2024 review found only 74 women and 77 pregnancies, mostly in case reports and case series.[29] Switching a stable person off an injection solely because pregnancy has begun may trade an uncertain fetal benefit for a real adherence or relapse risk. UK NICE advises avoiding depot treatment unless the person responds well to it and has a history of not taking oral medication reliably.[12] This decision needs the treating psychiatrist, not a blanket rule.

Breastfeeding questions are drug-specific

Milk transfer is only one part of a feeding decision. Infant age, prematurity, kidney and liver function, combination treatment, relapse risk, sleep, milk supply, and the family’s ability to monitor all matter.

These summaries come from LactMed and do not rank a drug that has never worked for you above one that has maintained years of stability.[30] A newborn who becomes hard to wake, feeds poorly, has fewer wet diapers, seems floppy or unusually stiff, trembles, has abnormal movements, or struggles to breathe needs prompt pediatric assessment.

Other treatments and diagnostic safeguards

Psychotherapy, family education, practical support, regular meals, substance-use treatment, and sleep protection can strengthen a medication plan. They do not replace an effective maintenance treatment for someone at high risk of mania or psychosis. Antidepressants are not a substitute for a mood-stabilizing plan in bipolar disorder and, when used alone, can destabilize some people.[7,11]

Electroconvulsive therapy may be considered by a specialist team when severe depression, mania, catatonia, or psychosis needs a rapid response or has not responded to medication. Both CANMAT and NICE address its use in this setting.[11,12] The supporting literature is limited: a 2024 systematic review of ECT in pregnancy found 130 pregnant patients across 45 case reports and case series, with many improvements reported, but no controlled comparison and substantial publication bias.[31] ECT requires psychiatric, obstetric, anesthesia, and newborn planning.

Your delivery and postpartum checklist

• Put one written plan in the obstetric, psychiatric, and hospital records. Name the maintenance medicine, what happens around labor, who can change it, and when follow-up occurs.

• Identify your early warning signs. Include reduced need for sleep, racing thoughts, irritability, unusual confidence, spending or risk-taking, feeling watched, hearing or seeing things, severe hopelessness, and inability to function.

• Protect sleep. Decide who covers a reliable block overnight and how feeding can be shared. Pumping, formula, donor milk, or combination feeding can be part of treatment, not evidence of failure.[11]

• Schedule psychiatric contact early after birth, not only the routine six-week obstetric visit. People with a previous postpartum psychosis or a severe bipolar I course may need postpartum prevention planned before delivery.[1,7,11]

• Tell pediatrics and the delivery hospital about late-pregnancy exposure. Ask what newborn observation is expected and which symptoms require urgent review.

• List every prescription, over-the-counter medicine, supplement, nicotine product, and alcohol or cannabis exposure. Lithium interactions and enzyme-inducing drugs can change risk.

• Name a support person who can notice changes, protect sleep, hold medication and appointment information, and take over infant care if symptoms emerge.

• Save emergency and crisis routes in both phones. Know which emergency department can provide psychiatric and obstetric care. Support lines help with navigation and are not emergency services.[34]

Questions to bring to your clinicians

1. Based on my history, what is my relapse risk during pregnancy and in the first two weeks and three months after birth?
2. Which past treatment prevented my most dangerous episodes, and what happened during previous reductions or missed doses?
3. What does the best human evidence say about this exact drug, trimester, and outcome, using absolute numbers when available?
4. Which blood tests, metabolic checks, ultrasounds, or fetal assessments are indicated, and which are optional?
5. What is our written labor plan, and does the hospital pharmacy stock my medicine?
6. If I breastfeed, who owns the infant-monitoring plan: pediatrics, psychiatry, or primary care? What symptoms or illnesses should pause feeding and trigger same-day testing?
7. How will we protect sleep, and who will take over infant care if I become sleepless, confused, manic, depressed, or paranoid?
8. If pregnancy was unplanned, whom do I call today so that I do not stop treatment alone?

Frequently asked questions

Is there one safest mood stabilizer for pregnancy?

No. Valproate has the clearest serious fetal risk and is generally avoided for bipolar disorder in pregnancy, while lamotrigine and lithium carry different tradeoffs. Lithium may be the most effective relapse-prevention medicine for a particular person. Diagnosis, past response, episode severity, trimester, co-medications, and monitoring capacity all matter.[9,16,17]

Should I stop my medicine as soon as a pregnancy test is positive?

Usually not on your own. Abrupt stopping can precipitate relapse or withdrawal. Contact the prescriber and obstetric team promptly. Valproate exposure warrants urgent review, but even then the transition should be clinically supervised.[4,16]

Does a normal ultrasound prove there was no medication effect?

No. Ultrasound can evaluate many structural outcomes, but not every defect and not later neurodevelopment. It also cannot predict whether a parent will relapse. Ask what the test can and cannot answer.[7,9,16]

Can I breastfeed while taking lithium?

Authorities disagree. The U.S. label says breastfeeding is not recommended, and UK NICE advises against it. LactMed and MotherToBaby describe selected healthy full-term infants who may breastfeed with a coordinated monitoring and illness plan. Prematurity, dehydration, poor feeding, fever, kidney problems, and inability to obtain prompt testing can all shift the balance away from breastfeeding.[12,14,15,35]

Is Seroquel safer than every other antipsychotic?

No universal ranking is supported. Quetiapine has substantial clinical experience and generally low milk transfer, but it can cause maternal metabolic effects and infant sleepiness. The best choice weighs the drug that keeps you well against pregnancy, newborn, lactation, and monitoring evidence.[26,30]

Does postpartum psychosis only happen to people with bipolar disorder?

No. Bipolar I disorder and prior postpartum psychosis are major risk factors, but first episodes occur in people with no known psychiatric diagnosis, and ACOG notes that most people hospitalized with postpartum psychosis had no psychiatric history on record. Sudden inability to sleep, confusion, mania, paranoia, hallucinations, or bizarre beliefs after birth need immediate medical attention.[2,33]

Where can I contribute pregnancy data?

The National Pregnancy Registry for Psychiatric Medications enrolls exposed and psychiatric comparison pregnancies at 1-866-961-2388, and is open to pregnant people 45 and younger with a history of psychiatric illness. The North American Antiepileptic Drug Pregnancy Registry enrolls eligible antiseizure-medication pregnancies at 1-888-233-2334. Registries collect evidence. They are not emergency or prescribing services.[16,36]

Where to get help

• National Maternal Mental Health Hotline: call or text 1-833-852-6262 (1-833-TLC-MAMA), free and confidential, 24/7, in English and Spanish, with interpreters for 60 more languages. For TTY, use your preferred relay service or dial 711, then the number. This is not a crisis line.
• 988 Suicide and Crisis Lifeline: call, text, or chat 988, 24/7. For Spanish, press 2 or text AYUDA to 988.
• 911: if anyone is in immediate danger of harm.
• Postpartum Support International HelpLine: 1-800-944-4773, press 1 for Spanish or 2 for English; text “Help” to 800-944-4773, or 971-203-7773 in Spanish. Calls and texts are returned during business hours, 8am to 11pm Eastern. This is not a crisis hotline and does not handle emergencies.
• MotherToBaby, for medication exposure questions: 1-866-626-6847 or text 1-855-999-3525, Monday to Friday, 8am to 7pm Eastern, United States and territories.

Related reading on NP FADY

• Planning Before the Test Line
• Postpartum Depression Doesn’t Always Look Like Sadness
• Pregnant, Anxious, and Taking Medication: A Two-Patient Decision
• ADHD Does Not Take Maternity Leave
• The Level Is a Conversation, Not a Grade
• Mood Stabilizers Demystified: Lithium, Lamictal, and Beyond
• “Retrofit in Calm Weather”: Building Your Relapse Signature

References

1. Wesseloo R, Kamperman AM, Munk-Olsen T, Pop VJM, Kushner SA, Bergink V. Risk of postpartum relapse in bipolar disorder and postpartum psychosis: a systematic review and meta-analysis. American Journal of Psychiatry. 2016;173(2):117-127. PMID 26514657. doi:10.1176/appi.ajp.2015.15010124.

2. Bergink V, Akbarian S, Byatt N, et al. Postpartum psychosis and bipolar disorder: review of neurobiology and expert consensus statement on classification. Biological Psychiatry. 2026;99(9):740-747. PMID 41135771. PMCID PMC12673989. doi:10.1016/j.biopsych.2025.10.016.

3. U.S. Food and Drug Administration. Zyprexa (olanzapine) prescribing information. Revised January 2026. Current label.

4. Viguera AC, Whitfield T, Baldessarini RJ, et al. Risk of recurrence in women with bipolar disorder during pregnancy: prospective study of mood stabilizer discontinuation. American Journal of Psychiatry. 2007;164(12):1817-1824. PMID 18056236. doi:10.1176/appi.ajp.2007.06101639.

5. Liu X, Smout S, Mahjani B, et al. Risk of relapse in psychotic and bipolar disorders after prenatal antipsychotic discontinuation. JAMA Network Open. 2026;9(3):e260682. PMID 41893848. PMCID PMC13032156. doi:10.1001/jamanetworkopen.2026.0682.

6. Kang S, et al. Antipsychotic continuation during pregnancy and risk of postpartum relapse in women with schizophrenia: nationwide register-based study. British Journal of Psychiatry. 2025;227(3):614-621. PMID 39789797. PMCID PMC12425161. doi:10.1192/bjp.2024.247.

7. American College of Obstetricians and Gynecologists. Treatment and management of mental health conditions during pregnancy and postpartum: ACOG Clinical Practice Guideline No. 5. Obstetrics & Gynecology. 2023;141(6):1262-1288. PMID 37486661. doi:10.1097/AOG.0000000000005202.

8. U.S. Food and Drug Administration. Pregnancy and Lactation Labeling Resources. Content current as of March 10, 2025. FDA page.

9. Patorno E, Huybrechts KF, Bateman BT, et al. Lithium use in pregnancy and the risk of cardiac malformations. New England Journal of Medicine. 2017;376(23):2245-2254. PMID 28591541. PMCID PMC5667676. doi:10.1056/NEJMoa1612222.

10. Wesseloo R, Wierdsma AI, van Kamp IL, et al. Lithium dosing strategies during pregnancy and the postpartum period. British Journal of Psychiatry. 2017;211(1):31-36. PMID 28673946. PMCID PMC5494438. doi:10.1192/bjp.bp.116.192799.

11. Vigod SN, Frey BN, Clark CT, et al. Canadian Network for Mood and Anxiety Treatments 2024 clinical practice guideline for the management of perinatal mood, anxiety, and related disorders. Canadian Journal of Psychiatry. 2025;70(6):429-489. PMID 39936923. PMCID PMC11985483. doi:10.1177/07067437241303031.

12. National Institute for Health and Care Excellence. Antenatal and postnatal mental health: clinical management and service guidance (CG192). Published 17 December 2014; anticonvulsant recommendations updated February 2020; maintenance amendments to April 2022; exceptional surveillance May 2025, no update. Guideline.

13. Molenaar NM, Poels EMP, Robakis T, Wesseloo R, Bergink V. Management of lithium dosing around delivery: an observational study. Focus. 2024;22(1):120-125. PMID 38694160. PMCID PMC11058920. doi:10.1176/appi.focus.23021031.

14. National Library of Medicine. Lithium. Drugs and Lactation Database (LactMed), NBK501153. Last revised June 15, 2026. PMID 30000212. LactMed record.

15. MotherToBaby. Lithium. MotherToBaby Fact Sheets, NBK582797. March 2025. PMID 35952070. Fact sheet.

16. U.S. Food and Drug Administration. Depakote (divalproex sodium) prescribing information. Revised March 2026. Current label.

17. Battino D, Tomson T, Weber YG, et al. Risk of major congenital malformations and exposure to antiseizure medication monotherapy. JAMA Neurology. 2024;81(5):481-489. PMID 38497990. PMCID PMC10949148. doi:10.1001/jamaneurol.2024.0258.

18. Meador KJ, Baker GA, Browning N, et al. Fetal antiepileptic drug exposure and cognitive outcomes at age 6 years. Lancet Neurology. 2013;12(3):244-252. PMID 23352199. PMCID PMC3684942. doi:10.1016/S1474-4422(12)70323-X.

19. Christensen J, Grønborg TK, Sørensen MJ, et al. Prenatal valproate exposure and risk of autism spectrum disorders and childhood autism. JAMA. 2013;309(16):1696-1703. PMID 23613074. PMCID PMC4511955. doi:10.1001/jama.2013.2270.

20. European Medicines Agency. Valproate and related substances: Article 31 referral. European Commission final decision, 31 May 2018. EMA referral page.

21. UK Medicines and Healthcare products Regulatory Agency. Valproate: reproductive risks. Published 10 June 2025; last updated 23 September 2025. MHRA guidance.

22. European Medicines Agency Pharmacovigilance Risk Assessment Committee. Meeting highlights, 8 to 11 June 2026. EMA meeting highlights.

23. National Library of Medicine. Lamotrigine. Drugs and Lactation Database (LactMed), NBK501268. Last revised March 15, 2026. LactMed record.

24. DailyMed. Trileptal (oxcarbazepine) prescribing information, sections 7.3 and 8.3; revised September 2025. DailyMed. Tegretol (carbamazepine) prescribing information, Precautions and Drug Interactions; revised July 2026. Carbamazepine label. Oxcarbazepine label.

25. Huybrechts KF, Straub L, Karlsson P, et al. Association of in utero antipsychotic medication exposure with risk of congenital malformations in Nordic countries and the US. JAMA Psychiatry. 2023;80(2):156-166. PMID 36477338. PMCID PMC9856848. doi:10.1001/jamapsychiatry.2022.4109.

26. Park Y, Hernandez-Diaz S, Bateman BT, et al. Continuation of atypical antipsychotic medication during early pregnancy and the risk of gestational diabetes. American Journal of Psychiatry. 2018;175(6):564-574. PMID 29730938. PMCID PMC5988929. doi:10.1176/appi.ajp.2018.17040393.

27. Cohen LS, Viguera AC, McInerney KA, et al. Reproductive safety of lurasidone and quetiapine: update from the National Pregnancy Registry for Psychiatric Medications. Journal of Women’s Health. 2023;32(4):452-462. PMID 36716275. doi:10.1089/jwh.2022.0310.

28. Bruno C, et al. Antipsychotic use during pregnancy and risk of neurodevelopmental disorders in children. eClinicalMedicine. 2024;70:102531. PMID 38685931. PMCID PMC11056394. doi:10.1016/j.eclinm.2024.102531. Corrigendum: eClinicalMedicine. 2025;81:103139. PMID 40104086.

29. Pejčić AV, et al. Outcomes of long-acting injectable antipsychotics use in pregnancy: a literature review. World Journal of Psychiatry. 2024;14(4):582-599. PMID 38659600. PMCID PMC11036459. doi:10.5498/wjp.v14.i4.582.

30. National Library of Medicine. LactMed monographs: Quetiapine (NBK501087), revised July 15, 2026; Olanzapine (NBK501056), revised April 15, 2026; Risperidone (NBK501095), revised January 15, 2026; Aripiprazole (NBK501016), revised April 15, 2026; Haloperidol (NBK500910), revised November 15, 2025; Lurasidone (NBK501728), revised November 15, 2025; Clozapine (NBK501650), revised July 15, 2025. LactMed quetiapine record.

31. Cipolla S, et al. Safety of electroconvulsive therapy (ECT) in pregnancy: a systematic review of case reports and case series. Archives of Women’s Mental Health. 2024;27(2):157-178. PMID 37957411. PMCID PMC10933171. doi:10.1007/s00737-023-01394-1.

32. National Library of Medicine. LactMed monographs: Valproic Acid (NBK501274), revised April 15, 2026; Carbamazepine (NBK501271), revised September 15, 2024; Oxcarbazepine (NBK501243), revised September 15, 2024. LactMed valproic acid record.

33. American College of Obstetricians and Gynecologists. Screening and diagnosis of mental health conditions during pregnancy and postpartum: ACOG Clinical Practice Guideline No. 4. Obstetrics & Gynecology. 2023;141(6):1232-1261. PMID 37486660. doi:10.1097/AOG.0000000000005200.

34. U.S. Health Resources and Services Administration, National Maternal Mental Health Hotline and program FAQ; Postpartum Support International, Postpartum Psychosis and HelpLine pages; 988 Suicide & Crisis Lifeline, About and Contact pages. All accessed September 12, 2026. Hotline page.

35. DailyMed. Lithium oral solution USP (Advagen Pharma), U.S. prescribing information, sections 2.4, 8.1, 8.2, and 17. Revised January 2025; DailyMed record updated September 11, 2026. Current label.

36. Massachusetts General Hospital Center for Women’s Mental Health. National Pregnancy Registry for Psychiatric Medications, participant enrollment page. Registry page last modified 26 March 2026. Registry page.

37. Pennell PB, et al. Lamotrigine in pregnancy: clearance, therapeutic drug monitoring, and seizure frequency. Neurology. 2008;70(22 Pt 2):2130-2136. PMID 18046009. PMCID PMC3589527. doi:10.1212/01.wnl.0000289511.20864.2a.

38. Clark CT, Klein AM, Perel JM, Helsel J, Wisner KL. Lamotrigine dosing for pregnant patients with bipolar disorder. American Journal of Psychiatry. 2013;170(11):1240-1247. PMID 24185239. PMCID PMC4154145. doi:10.1176/appi.ajp.2013.13010006.

If you or someone you know is in crisis

• Call 911 or go to your nearest emergency room for any life-threatening emergency.
• 988 Suicide & Crisis Lifeline — call or text 988, or chat at 988lifeline.org, available 24/7, free and confidential. En español: marque 988 y oprima 2, o envíe AYUDA al 988. Deaf or hard of hearing: 988 videophone (ASL) or TTY 711 then 988. Veterans: 988 and press 1, or text 838255.
• Crisis Text Line — text HOME to 741741.
• The Trevor Project (crisis support for LGBTQ+ young people) — call 1-866-488-7386, or text START to 678-678.
• National Sexual Assault Hotline (RAINN) — call 1-800-656-HOPE (4673) or text HOPE to 64673; free, confidential, 24/7. Online chat at RAINN.org/hotline.
• National Domestic Violence Hotline — call 1-800-799-SAFE (7233) or text START to 88788; 24/7, help in 200+ languages. Online chat at TheHotline.org. If your phone or computer may be monitored, calling from a safer device is an option.
• Riverside County — Inland SoCal Crisis Helpline 951-686-HELP (4357), 24/7 (Inland SoCal United Way / 211+, in partnership with RUHS-BH); Community Access, Referral, Evaluation and Support (CARES) Line 800-499-3008, 24/7.
• San Bernardino County — Access Unit (Behavioral Health Helpline) 888-743-1478, 24/7; Mobile Crisis/CCRT 800-398-0018 (24/7, all ages) or text 909-420-0560. Arrowhead Regional Medical Center (ARMC) has a dedicated walk-in adolescent psychiatric ER (ages 13–17).
• Children under 13 — call 911 for immediate danger, contact your county's mobile crisis team (they respond to all ages), or go to the nearest pediatric emergency room.
• California Peer-Run Warm Line (non-crisis — someone to talk to) — call or text 1-855-600-WARM (9276); daytime and evening hours, not a 24/7 line.
• NP Fady (non-emergency) — for routine scheduling or questions, call (909) 707-6261. This line is not monitored for emergencies.