Herbal & OTC Agents for Sleep in Pregnancy
- The headlineHerbal and OTC "natural" sleep aids are the most commonly self-prescribed sleep interventions in pregnancy and have the weakest evidence base of any category in this book. No herbal sleep agent has a completed RCT in pregnancy. Muñoz Balbontín systematic review (Obstet Gynecol 2019; 74 studies, >1M women): herbal medicinal products "should be discouraged until robust evidence of safety is available."
- ValerianThe most-studied herbal hypnotic. Valente 2024 umbrella review (Eur Neuropsychopharmacol; 8 systematic reviews): "no evidence of efficacy for the treatment of insomnia" by objective measures. AAFP 2023: "insufficient information: avoid use." Animal teratology reassuring (NOAEL ≥14 g/kg) but zero human pregnancy data.
- ChamomileAAFP 2023 explicit safety concerns: ductus arteriosus closure, increased preterm birth, low birth weight. Mechanism: polyphenol-mediated prostaglandin synthesis inhibition (same as NSAIDs). Cuzzolin 2010 cohort (37 regular users): 21.6% threatening miscarriage, 21.6% preterm labor. Occasional tea OK; concentrated/regular use should be avoided especially in T3.
- PassionflowerOne small PSG-based RCT (Lee 2020, n=110): TST +23 min vs placebo, borderline p=0.049. The only human pregnancy report (Ozturk 2018, 5 women): 1 neonatal death, 2 PROM, 2 MAS, 1 PPHN. Harman alkaloids are uterotonic. Avoid in pregnancy.
- MagnesiumSafe as a nutritional supplement at 200–400 mg/day elemental (UTL 350 mg/day for supplemental Mg). Cochrane review (10 RCTs, 9,090 women): no convincing benefit for pregnancy outcomes. For sleep: SOL −17 min in 3 RCTs of older adults (Mah 2021), low-quality evidence. No pregnancy-specific sleep RCTs. Reasonable as nutrition, not as a hypnotic.
- L-theanineBulman 2025 meta-analysis (19 RCTs, 897 participants): modest improvements in sleep quality (SMD 0.43) and SOL (SMD 0.15). Mechanism is anxiolysis via alpha-wave induction, not sedation. Zero pregnancy data. GRAS for food use; not FDA-evaluated for pregnancy. Cannot recommend or strongly warn against.
- CBD & St. John's wortCBD: ACOG Clinical Consensus No. 10 (2025) — "no medical indications for cannabis use during pregnancy." ~32% pregnant women have ever used CBD; 21% of CBD products contain undeclared THC. St. John's wort: Schäfer 2021 German claims — 5.8% major malformations in 312 exposed babies (vs ~3% background); potent CYP3A4 inducer disqualifies it regardless.
- What to doCBT-I first — RCT-supported in pregnancy (Felder JAMA Psychiatry 2020 digital CBT-I d=−1.03 effect size; Manber 2019 ISI 15.4→8.0; 64% remission). If pharmacotherapy is needed, Z-drugs have the 4.3M-pregnancy safety database. Screen for herbal supplement use at the first prenatal visit. The conversation: "Natural doesn't mean studied, and studied doesn't mean safe in pregnancy."
Executive Summary
Herbal and OTC "natural" sleep aids — valerian, chamomile, passionflower, magnesium, L-theanine, CBD, and others — occupy a unique and clinically dangerous position in perinatal pharmacology. They are the most commonly self-prescribed sleep interventions in pregnancy, the least regulated, and the least studied.
- The epidemiology: approximately 29% of pregnant women worldwide use herbal supplements, with prevalence ranging from 7% to 55% depending on culture, ethnicity, and socioeconomic status. Herbal supplement use is frequently not reported to healthcare professionals, and providers often do not ask.
- The evidence base: no herbal sleep agent has a completed RCT for insomnia in pregnancy. Existing data consist of in vitro placental studies, animal teratology, small uncontrolled case series, and extrapolation from non-pregnant adult data of generally low quality.
- The regulatory void is the defining problem. Unlike the FDA-regulated drugs in Chapters One through Seven, herbal supplements are marketed under the Dietary Supplement Health and Education Act (DSHEA, 1994), which requires no pre-market proof of safety, efficacy, or content accuracy. The quality-control crisis documented for melatonin (Chapter Six: content −83% to +478% of label, 26% serotonin contamination) applies to the entire supplement industry.
- The "natural = safe" fallacy is the defining patient misconception. Many herbal agents have pharmacologic activity (GABAergic, serotonergic, uterotonic, CYP-modulating) that is relevant in pregnancy. Absence of evidence of harm is not evidence of absence of harm.
- The clinical action is straightforward: screen for herbal supplement use at the first prenatal visit; counsel about the evidence gaps; transition to evidence-based alternatives (CBT-I first, Z-drugs when pharmacotherapy is needed); document the discussion.
Every drug in Chapters One through Seven is manufactured under FDA Good Manufacturing Practices with mandated content accuracy, purity standards, adverse-event reporting, and pregnancy labeling. The agents in this chapter are not. A pregnant woman buying "valerian root 500 mg" at a pharmacy has no guarantee the capsule contains 500 mg of valerian, no guarantee it is free of contaminants, and no guarantee the specific extract matches the one in any clinical study. The melatonin quality-control crisis from Chapter Six is not unique to melatonin — it reflects a systemic industry problem.
The Regulatory Void — Why This Chapter Is Different
Under the Dietary Supplement Health and Education Act (DSHEA, 1994), dietary supplements — including valerian, chamomile, passionflower, magnesium supplements, L-theanine, CBD products, and St. John's wort — do not require pre-market FDA approval for safety or efficacy. Specifically:
- No pre-market safety review: manufacturers do not have to prove the product is safe before selling it.
- No pre-market efficacy review: manufacturers do not have to prove the product works.
- No mandated content accuracy: manufacturers do not have to prove the labeled dose is in the capsule.
- No mandated pregnancy labeling.
- Voluntary adverse-event reporting only.
- The FDA can only act after evidence of harm emerges.
The Content Variability Crisis
The clearest documentation of the consequences comes from Erland & Saxena (JCSM 2017), discussed in detail in Chapter Six: 31 melatonin supplements tested for content variability showed −83% to +478% of label content, with serotonin contamination detected in 26% of products. This is not a melatonin problem — it is a supplement industry problem. The same regulatory framework governs every product in this chapter. A pregnant woman buying valerian, chamomile extract, or "sleep formula" capsules has no guarantee about content accuracy or contamination.
The Pregnancy Conversation Gap
About 80% of consumers worldwide use herbal medicines or other natural products. Among pregnant women, the percentage ranges from 7% to 55% depending on social status, ethnicity, and cultural traditions. The Muñoz Balbontín 2019 systematic review (Obstet Gynecol; 74 articles, 1,067,071 women) emphasized that "herbal medicine use is frequently not reported to healthcare professionals, and providers often do not ask." This combination — high prevalence, low disclosure, and provider non-inquiry — creates the most common herbal-pregnancy exposure scenario: months of undisclosed use before anyone thinks to ask.
This is the largest synthesis of herbal medicine use in pregnancy ever published. After excluding irrelevant and low-quality articles, 74 articles were included covering 47 herbal medicinal products. Specific adverse events identified:
- Almond oil — preterm birth OR 2.09 (1.07–4.08)
- Oral raspberry leaf — cesarean delivery AOR 3.47 (1.45–8.28)
- Heavy licorice use — early preterm birth 3.07-fold (1.17–8.05)
- African mwanaphepo — maternal morbidity AOR 1.28; neonatal death/morbidity
Authors' conclusion: "The use of herbal medicinal products during pregnancy and the postnatal period should be discouraged until robust evidence of safety is available." Notably: 14 studies reported absence of adverse events, but 4 studies reporting herb-drug interactions did not study adverse events arising from them.
Valerian (Valeriana officinalis)
Valerian root extract has been used as a sedative since ancient Greek and Roman times. Its proposed mechanism involves GABAergic neurotransmission — valerenic acid inhibits GABA reuptake and modulates GABA-A receptor activity. It is the most-studied herbal hypnotic, which makes the negative efficacy findings all the more striking.
Efficacy — The Evidence Is Negative
The Valente et al. 2024 umbrella review (Eur Neuropsychopharmacol; 8 systematic reviews) concluded: "data suggested that valerian has a good safety profile; however, the results showed no evidence of efficacy for the treatment of insomnia." Some subjective improvements in sleep quality were reported in 3 meta-analyses and in sleep latency in 4 meta-analyses, but objective sleep outcomes (actigraphy, PSG measurements) generally did not improve. A high risk of bias was observed in all 8 systematic reviews, and methodological quality was critically low in 7. The characteristic odor of valerian may have compromised blinding in placebo trials.
The AASM (Sateia et al. 2017) reviewed two RCTs of valerian and found low-quality evidence for a possible small improvement in sleep-onset latency, but no improvement in TST, WASO, sleep efficiency, or daytime functioning. The task force determined that "the majority of patients would not be likely to use valerian compared to no treatment."
The VA/DoD 2019 guideline found "no between-group differences in the critical outcomes of daytime functioning, insomnia severity, and sleep efficiency" for valerian versus placebo. The Ell et al. 2023 umbrella review (J Sleep Res; 15 eligible studies) similarly concluded that "light exposure and valerian did not significantly improve sleep outcomes."
Pregnancy Safety — Insufficient Data, Reassuring Animal Teratology
- Animal data: Bao et al. 2024 (J Ethnopharmacol) comprehensive safety evaluation found no genotoxicity, no teratogenicity, and no maternal or fetal toxicity in rats at doses up to 14 g/kg body weight. Yao et al. 2007 found no adverse effects on fertility or fetal development at up to 65× the human dose. However, in vitro embryo culture showed toxicity at high concentrations; preparations containing valepotriates (epoxide-containing cytotoxic compounds) "could have a very different outcome."
- In vitro placental safety: The Simões-Wüst group studies (Planta Med 2022; Front Pharmacol 2022) using BeWo b30 placental cells found no cytotoxicity, genotoxicity, or effects on metabolic properties at ≤30 µg/mL. However, the specific compound valtrate showed cytotoxic and apoptotic effects at ≥10 µM and decreased glucose consumption at all concentrations — warranting "special attention" at high plasma concentrations.
- Transplacental passage (Spiess 2023, Front Pharmacol): valerenic acid equilibrates between maternal and fetal compartments — i.e., it does cross.
- Human pregnancy data: none. No observational cohorts, no case series, no registry data.
AAFP 2023 (Powers et al.): "Insomnia: possibly effective. Safety in pregnancy: insufficient information: avoid use."
Clinical Position
Valerian fails on both counts: it does not work well enough by objective measures to justify the unknown pregnancy risk, and the pregnancy risk is unknown enough to justify avoiding it. The conversation is simple: "Valerian hasn't been shown to improve sleep by any objective measure, and we have no pregnancy safety data. There are better options."
Chamomile (Matricaria chamomilla)
Chamomile tea is perhaps the most culturally embedded "sleep remedy" worldwide. Its proposed mechanism involves apigenin, a flavonoid that binds GABA-A receptors — but at concentrations far below those achieved by pharmacologic GABAergic agents.
Efficacy — No Evidence
The VA/DoD 2019 guideline found no benefit of chamomile over placebo for any critical insomnia outcome (daytime functioning, insomnia severity, sleep efficiency, SOL, TST, WASO, or sleep quality).
Pregnancy Safety — Genuine Concerns
The AAFP 2023 lists chamomile with the following pregnancy safety concerns: "Ductus arteriosus closure, increased preterm birth, low birth weight." Effectiveness assessment: "Lack of evidence of effectiveness."
The ductus arteriosus concern is biologically plausible. Chamomile contains polyphenols (including apigenin) that inhibit prostaglandin synthesis — the same mechanism by which NSAIDs cause premature ductus arteriosus constriction. Zielinsky et al. (Expert Rev Cardiovasc Ther 2010; J Perinatol 2010) demonstrated that maternal consumption of polyphenol-rich foods (>1,089 mg/day) in the third trimester was associated with significantly higher ductal velocities and right-to-left ventricular dimension ratios compared with low-polyphenol diets — consistent with ductal constriction.
The Cuzzolin et al. Italian cohort (Pharmacoepidemiol Drug Saf 2010; 392 pregnant women) found that among 37 regular chamomile users, the frequency of threatening miscarriage was 21.6% and preterm labor was 21.6% — higher than non-users. Two neonates of chamomile-using mothers had malformations (one cardiac defect in a Down syndrome infant, one enlarged kidney). Causation could not be established, but authors noted that "a possible influence of a regular intake of chamomile on threatening miscarriages and preterm labours could be hypothesized."
Clinical Position — Dose Matters
An occasional cup of chamomile tea is unlikely to cause harm. Regular, concentrated chamomile supplementation (capsules, extracts, or multiple daily cups of strong tea) should be avoided in pregnancy — particularly in the third trimester, when the ductus arteriosus is most vulnerable to prostaglandin-synthesis inhibition. The efficacy case is nonexistent regardless.
Passionflower (Passiflora incarnata)
Passionflower has been used as a sedative in traditional medicine across the Americas, Europe, and the Middle East. Its proposed mechanism involves GABAergic modulation, though the active ingredients and precise mechanism remain incompletely characterized. It also contains harman alkaloids with documented uterotonic properties.
Efficacy — One Positive RCT, Small
The Lee et al. RCT (Int Clin Psychopharmacol 2020; n=110 adults with insomnia disorder) was double-blind and PSG-based. Passionflower extract increased total sleep time by 23.05 min vs −0.16 min for placebo (P=0.049 — borderline significant). Sleep efficiency and WASO improved within the passionflower group but did not differ significantly from placebo. This is the only PSG-based RCT of passionflower for insomnia — a single small study with a borderline p-value.
Five pregnant women with depression and/or anxiety who used Passiflora incarnata during pregnancy (doses 700–2,100 mg/day, varying gestational timing). Outcomes:
- One neonatal death (Case 1) at week 36 — PPROM and meconium-stained amniotic fluid
- Two cases of premature rupture of membranes (Cases 1 and 5)
- Two infants with meconium aspiration syndrome (Cases 3 and 4)
- One infant with persistent pulmonary hypertension of the newborn (Case 4)
- No birth defects in live-born infants; no growth or developmental abnormalities at 6 months follow-up
The case series is uncontrolled, tiny, and cannot establish causation — but it is the only human pregnancy data that exist for passionflower, and the harman alkaloids in passionflower have documented uterotonic activity, providing a plausible mechanism for the PROM and meconium-related outcomes observed. Authors concluded: "Pregnant women exposed to Passiflora incarnata should be monitored, unless further data are available."
Clinical Position
Passionflower should be avoided in pregnancy. The efficacy evidence is marginal (one small RCT with borderline p), the pregnancy safety data are alarming (though uncontrolled), and the uterotonic mechanism is biologically plausible. There is no clinical scenario in which passionflower is the best available option for a pregnant patient.
Magnesium
Magnesium occupies a different category from the herbal agents above — it is an essential mineral with well-established physiologic roles, a defined RDA in pregnancy (350–400 mg/day elemental), and a long safety record at nutritional doses. The question is not whether magnesium is safe in pregnancy (it is), but whether it works as a sleep aid (the evidence is weak).
Efficacy for Sleep — Weak
Mah & Pitre meta-analysis (BMC Complement Med Ther 2021; 3 RCTs, 151 participants in older adults) found magnesium supplementation reduced sleep-onset latency by 17.36 minutes versus placebo (95% CI −27.27 to −7.44; P=0.0006). Total sleep time improved by 16 minutes (NS). All trials were at moderate-to-high risk of bias; outcomes supported by low to very low quality evidence.
Arab et al. systematic review (Biol Trace Elem Res 2023; 7,582 subjects from 9 studies) found that observational studies suggested an association between magnesium status and sleep quality, while RCTs showed "an uncertain association between magnesium supplementation and sleep disorders."
Pregnancy Outcomes — Cochrane Says No Convincing Benefit
Makrides et al. Cochrane review (2014; 10 RCTs, 9,090 women) found no significant difference in perinatal mortality, SGA, or preeclampsia with oral magnesium supplementation vs placebo. When analysis was restricted to the two high-quality trials, none of the primary outcomes differed between groups. Authors concluded: "There is not enough high quality evidence to show that dietary magnesium supplementation during pregnancy is beneficial."
Clinical Position
Magnesium supplementation at nutritional doses (200–400 mg elemental magnesium daily, typically as magnesium glycinate, citrate, or oxide) is safe in pregnancy and reasonable for women with inadequate dietary intake — which is common. It should not be prescribed or marketed as a hypnotic. The sleep-promoting effect, if real, is modest and not established in pregnant populations. Practical guidance:
- Magnesium glycinate preferred over magnesium oxide for tolerability (less GI distress).
- The IOM upper tolerable intake level for supplemental magnesium is 350 mg/day — doses above this may cause diarrhea, nausea, and abdominal cramping.
- The honest framing for patients: "Magnesium is a mineral your body needs, and most pregnant women don't get enough. It might help a little with sleep, but it's not a sleeping pill. Think of it as good nutrition, not a treatment."
L-Theanine
L-theanine (γ-glutamylethylamide) is a non-proteinogenic amino acid found in tea leaves (Camellia sinensis). Its proposed mechanism is anxiolysis via induction of alpha brain waves, modulation of GABA, serotonin, and dopamine neurotransmission, and glutamate receptor antagonism — promoting relaxation without sedation.
Efficacy — Modest, Growing Evidence Base
Bulman et al. 2025 meta-analysis (Sleep Med Rev; 19 RCTs, 897 participants) found L-theanine significantly improved:
- Subjective sleep-onset latency (SMD 0.15; 95% CI 0.01–0.29; P=0.04)
- Subjective sleep quality (SMD 0.43)
- Daytime dysfunction (SMD 0.33)
The Cotter et al. 2025 systematic review (Nutr Neurosci; 13 trials, 550 participants) concluded that "200–450 mg/day appears to be a safe and effective way to support healthy sleep in adults." Effects are modest but real in non-pregnant adults.
Pregnancy Safety — Zero Data
No published pregnancy data exist for L-theanine — no observational cohorts, no case series, no animal teratology specific to pregnancy. L-theanine has GRAS (Generally Recognized as Safe) status for use as a food additive but has not been formally evaluated by the FDA for pregnancy use. Safety beyond GRAS for food purposes cannot be assumed for supplemental doses (typically 200–400 mg).
Clinical Position
Insufficient data to recommend or strongly warn against L-theanine in pregnancy. The evidence gap is the relevant fact, not a specific safety signal. The honest framing: "L-theanine helps some people with sleep quality in adults who aren't pregnant, but we have no data on it in pregnancy. We have better-studied options."
CBD (Cannabidiol)
ACOG Clinical Consensus No. 10 (Obstet Gynecol 2025;146(4):600–611): "There are no medical indications for cannabis use during pregnancy and the postpartum period." This explicitly includes CBD as well as THC-containing products. The FDA likewise recommends avoiding CBD during pregnancy and breastfeeding.
The Mechanism & Pregnancy Concerns
- CBD crosses the placenta and may affect placental transporters.
- CYP450 inhibitor: CBD inhibits CYP2C19 and CYP3A4, creating drug-interaction risk with SSRIs, anticonvulsants, and other medications that pregnant women may be taking.
- Animal data: decreased pup weight and neurodevelopmental effects observed at supratherapeutic exposures.
- Cannabis exposure (which includes any combined CBD/THC products) is associated with small-to-moderately increased risks of preterm birth, SGA, low birth weight, NICU admission, and fetal death (Andrade J Clin Psychiatry 2024; Lo AJOG 2022).
The Product Contamination Problem
The CBD product marketplace is essentially unregulated. Published quality-control surveys have found undeclared THC in approximately 21% of CBD-labeled products. A pregnant woman taking what she believes is "CBD only" may be receiving meaningful THC exposure without her knowledge. This is the same regulatory void described in § One, with a particularly high-stakes ingredient.
The Epidemiology
Azubuike et al. 2026 (Obstet Gynecol): approximately 32% of pregnant women report ever having used CBD, with ~4% reporting use in the past 30 days during pregnancy. This means a meaningful fraction of any obstetric panel has personal experience with the product, and many will not consider it "drug use" worth disclosing.
Clinical Position
CBD is contraindicated in pregnancy. The conversation requires directness because patients perceive CBD as a wellness product, not a drug. The framing: "I know CBD is marketed as natural and safe, but the medical societies are clear on this one. It crosses the placenta, it can contain THC even when the label says it doesn't, it can interfere with other medications you're taking, and the official recommendation from ACOG is not to use it during pregnancy."
St. John's Wort (Hypericum perforatum)
St. John's wort is not typically a sleep aid but is frequently co-used by patients with depression who report sleep disturbance. Its inclusion here matters because the drug interaction profile threatens every other medication discussed in this course.
The Schäfer 2021 Claims Analysis — The Largest Pregnancy Cohort
Schäfer et al. (Reprod Toxicol 2021;102:90–97) analyzed German claims data identifying 496 pregnancies with SJW dispensation between 2006 and 2016. Key findings:
- 420 (85%) had a dispensation during the first trimester
- 21% had pre-pregnancy antidepressant use; 12% during pregnancy (suggesting many used SJW instead of conventional antidepressants)
- 11% of pregnancies ended in non-live births
- Among 312 babies linked to 305 pregnancies, major malformations were coded in 18 (5.8%) — compared with a background rate of approximately 3%
- 17 of 18 affected babies were exposed in the first trimester
The signal is hypothesis-generating and uncontrolled (no comparator arm in this analysis), but it is the largest pregnancy cohort with SJW exposure data and the only one to identify a malformation rate roughly twice background.
The Drug-Interaction Profile — Disqualifying on Its Own
St. John's wort is a potent CYP3A4 inducer via pregnane X receptor (PXR) activation, and a P-glycoprotein inducer. The clinical consequences for a patient taking concurrent medications:
| Affected Medication | Effect of Co-Administration with SJW | Clinical Consequence |
|---|---|---|
| Z-drugs (zolpidem, eszopiclone, zaleplon) | Substantially reduced levels via CYP3A4 induction | Therapeutic failure; the Z-drug doesn't work |
| Benzodiazepines (midazolam, alprazolam, triazolam) | Reduced levels | Therapeutic failure; potential rebound if abruptly added |
| Quetiapine, aripiprazole, lurasidone | Reduced levels via CYP3A4 induction | Therapeutic failure of antipsychotic; psychiatric decompensation risk |
| Oral contraceptives | Reduced ethinyl estradiol/progestin levels; breakthrough bleeding | Contraceptive failure → unintended pregnancy |
| SSRIs, SNRIs | Pharmacodynamic interaction (additive serotonergic effect) | Serotonin syndrome risk |
| Lamotrigine | Possible reduction via UGT induction | Loss of seizure control / mood stabilization |
| DORAs (suvorexant, lemborexant, daridorexant) | Reduced AUC via CYP3A4 induction (rifampin-suvorexant interaction shows AUC ↓88%) | Therapeutic failure |
| Cyclosporine, antiretrovirals, warfarin | Reduced levels | Transplant rejection / viral breakthrough / thrombosis |
Clinical Position
St. John's wort is one of the most dangerous herbal products in pregnancy. Even if it were proven safe in pregnancy (it is not — Schäfer 5.8% vs ~3% background), its CYP3A4 induction profile would disqualify it for any patient taking a CYP3A4-substrate medication — which includes essentially every drug discussed in Chapters One through Seven. Discontinue at the first prenatal visit. Transition to an SSRI with established pregnancy safety data if antidepressant treatment is needed.
Other Agents — Briefly
| Agent | Mechanism / Adult Evidence | Pregnancy Position |
|---|---|---|
| Hops (Humulus lupulus) | Often combined with valerian; weak adult sleep data; in vitro BeWo placental cells safe at ≤30 µg/mL (Spiess 2022) | Insufficient pregnancy data; avoid concentrated supplements; small amounts in beer not the question (alcohol exposure dominates) |
| Lavender (oral) | Silexan oral preparation has some anxiolytic RCT data; no insomnia RCT in pregnancy; linalool BeWo cells inconspicuous | Aromatherapy with topical/diffused lavender is generally considered low-risk; oral supplements should be avoided pending pregnancy data |
| Ashwagandha (Withania somnifera) | Adaptogen marketed for sleep and anxiety; mixed adult data; historical use as an abortifacient in Ayurvedic medicine; no controlled pregnancy data | Avoid in pregnancy — abortifacient history is a contraindication signal even without controlled data |
| Kava (Piper methysticum) | Some adult anxiolytic data; FDA warning: acute fatal hepatotoxicity; VA/DoD "strong against" recommendation | Absolutely contraindicated — hepatotoxicity risk plus no pregnancy data |
| California poppy (Eschscholzia californica) | GABAergic mechanism; minimal RCT data; BeWo cells safe at ≤30 µg/mL (Spiess 2022) | Insufficient pregnancy data; not recommended |
| 5-HTP / tryptophan | Serotonin precursor; modest adult sleep data; theoretical serotonin syndrome risk with SSRIs/SNRIs | No pregnancy data; avoid particularly if on SSRI; the supplement industry's eosinophilia-myalgia syndrome history is a cautionary precedent |
| "Sleep formula" blends (combo products) | Variable ingredients; often contain undisclosed melatonin, valerian, passionflower | The unknowns multiply when ingredients are combined; ingredient lists are not reliable; avoid |
Ginger is the one herbal exception with reasonable pregnancy data — but as an antiemetic for nausea/vomiting, not a sleep aid (Sarecka-Hujar & Szulc-Musioł, Pharmaceutics 2022).
Comparative Evidence Table
| Agent | Adult Insomnia Efficacy | Pregnancy Safety Data | Clinical Position |
|---|---|---|---|
| Valerian | Valente 2024 umbrella (8 SRs): no objective efficacy; AASM/VA/DoD: no benefit | Animal: NOAEL ≥14 g/kg, no malformations at 65× human dose. In vitro BeWo cells safe ≤30 µg/mL. Zero human data. | Avoid — AAFP: "insufficient information" |
| Chamomile | No benefit (VA/DoD) | Polyphenol-mediated PG inhibition; Cuzzolin 2010 cohort: 21.6% threatening miscarriage / 21.6% preterm in regular users | AAFP safety concerns: ductus arteriosus closure, preterm birth, LBW. Occasional tea OK; concentrated/regular avoid (esp. T3) |
| Passionflower | One small PSG-RCT (Lee 2020, n=110): TST +23 min vs placebo, borderline p=0.049 | Ozturk 2018 (5 women): 1 neonatal death, 2 PROM, 2 MAS, 1 PPHN. Harman alkaloids uterotonic | Avoid in pregnancy |
| Magnesium (nutritional) | Mah 2021 (3 RCTs, 151 older adults): SOL −17 min; low-quality evidence | Cochrane: 10 RCTs, 9,090 women; no convincing benefit but safe at nutritional doses (≤350 mg/day supplemental) | OK as nutrition, not as a hypnotic; magnesium glycinate preferred |
| L-Theanine | Bulman 2025 (19 RCTs, 897 adults): modest improvement (SOL SMD 0.15; quality SMD 0.43) | Zero pregnancy data. GRAS for food only | Insufficient data — better-studied options exist |
| CBD | Not FDA-approved for sleep; inverted U-shaped dose-response for anxiety | Crosses placenta; CYP2C19/3A4 inhibitor; ~21% products contain undeclared THC; ~32% pregnant women ever used; animal pup weight ↓ | ACOG 2025: contraindicated. "No medical indications for cannabis use in pregnancy" |
| St. John's Wort | Antidepressant (not hypnotic); modest mild-moderate depression data | Schäfer 2021: 5.8% MCM in 312 babies (vs ~3% background). Potent CYP3A4 inducer — disqualifies regardless of teratogenicity | Avoid — drug interaction profile alone is disqualifying |
| Kava | Some anxiolytic data; VA/DoD: "strong against" | No pregnancy data; FDA hepatotoxicity warning | Absolutely contraindicated |
| Ashwagandha | Mixed adult sleep/anxiety data | Historical abortifacient in Ayurvedic medicine; no controlled data | Avoid in pregnancy |
Guideline Recommendations
AAFP — OTC Medications in Pregnancy (Powers et al., Am Fam Physician 2023)
"Approximately 29% of women worldwide reported taking herbal supplements during pregnancy." The AAFP summary table:
- Valerian: "Insomnia: possibly effective. Safety in pregnancy: insufficient information: avoid use."
- Chamomile: "Lack of evidence of effectiveness. Safety: ductus arteriosus closure, increased preterm birth, low birth weight."
- Melatonin: "Insufficient human trials to support its safe use in pregnant patients with sleep disturbance" (see Chapter Six).
- First-generation H1 antihistamines (e.g., diphenhydramine, doxylamine): "Acceptable for as-needed use" for sleep — a notably stronger endorsement than any herbal agent receives (see Chapter Three).
ACOG Committee Opinion No. 762 (2019) — Prepregnancy Counseling
"All prescription and nonprescription medications should be reviewed during prepregnancy counseling. This review also should include nutritional supplements and herbal products that patients may not consider to be medication use but could affect reproduction and pregnancy."
ACOG Clinical Consensus No. 10 (2025) — Cannabis & CBD
"There are no medical indications for cannabis use during pregnancy and the postpartum period." Applies to CBD as well as THC-containing products. Prevalence of cannabis use among pregnant individuals ranges 3.9% to 16.0%.
AASM Clinical Practice Guideline (Sateia et al. 2017)
Reviewed valerian; "the majority of patients would not be likely to use valerian compared to no treatment." Neither valerian, chamomile, nor any herbal agent is recommended for chronic insomnia.
VA/DoD Clinical Practice Guideline (2019)
No benefit of valerian or chamomile over placebo for any critical insomnia outcome. "Strong against" recommendation for kava due to acute fatal hepatotoxicity risk.
Systematic Review of CAM Insomnia Guidelines (Zhao et al., Front Public Health 2023)
17 clinical practice guidelines reviewed; "the only consensus was that four phytotherapeutics including valerian, chamomile, kava, and aromatherapy were not recommended for insomnia management because of risk profile and/or limited benefits."
European Insomnia Network / Marcè Society (2022)
CBT-I should be the preferred treatment for insomnia during the peripartum period. No specific endorsement of any herbal agent for peripartum insomnia.
The Synthesis
No major guideline recommends any herbal sleep agent for insomnia in pregnancy — or even for insomnia in the general adult population. The consensus across AAFP, ACOG, AASM, VA/DoD, the Marcè Society, and the European Insomnia Network: CBT-I first, FDA-regulated pharmacotherapy when needed, and herbal supplements discouraged until robust safety data are available.
CBT-I — The Evidence-Based Path
The single most important alternative to herbal sleep aids in pregnancy is not another medication — it is cognitive behavioral therapy for insomnia (CBT-I). Every major guideline ranks CBT-I as first-line for chronic insomnia, including during pregnancy. Unlike every herbal agent in this chapter, CBT-I has substantial RCT evidence specifically in pregnant populations.
RCT of 208 pregnant women with insomnia symptoms randomized to digital CBT-I (Sleepio) vs standard care. At the 18-week post-randomization follow-up, the digital CBT-I group showed a between-group effect size of d = −1.03 on insomnia severity (Insomnia Severity Index), with benefits maintained at follow-up. Depression and anxiety symptoms also improved. This is the largest pregnancy-specific RCT of any insomnia intervention and demonstrates that scalable, non-pharmacologic treatment is effective.
Additional Pregnancy RCT Evidence
- Manber et al. (Obstet Gynecol 2019; n=194): face-to-face CBT-I in pregnancy. ISI improved from 15.4 → 8.0 (CBT-I) vs 15.9 → 11.2 (control). 64% remission rate in the CBT-I group.
- Zheng et al. systematic review (J Sleep Res 2023; 8 RCTs, 743 women): ISI mean difference −4.25; significant improvements in sleep quality, SOL, anxiety, and depression.
- Shang et al. (Sleep Med 2023; 9 RCTs, 978 women): CBT-I improved insomnia severity and sleep quality immediately and at short-term follow-up.
- Bacaro et al. (Sleep Med Rev 2020): CBT-I first-line for peripartum insomnia.
Practical Implementation
- Digital CBT-I programs (Sleepio, Somryst, SHUTi) are scalable and have specifically been studied in pregnancy. Recommend the Sleepio app or equivalent for patients who prefer self-paced.
- In-person CBT-I with a behavioral sleep medicine specialist is the gold standard but limited by specialist availability.
- Brief behavioral treatment for insomnia (BBTI) — a 4-session abbreviated version — is feasible in primary care or OB settings with brief provider training.
- Core components: sleep restriction therapy, stimulus control, cognitive restructuring of dysfunctional sleep beliefs, sleep hygiene education, relaxation training.
- Pregnancy-specific adaptations: address nocturia, pregnancy-related discomfort, RLS screening (prevalence ~22% in T3), heightened anxiety about delivery and infant care.
When Pharmacotherapy Is Still Needed
For patients who have failed CBT-I or for whom the severity of insomnia requires immediate pharmacologic relief, the evidence-based hierarchy is:
- Z-drugs (zaleplon, zolpidem IR for onset; zolpidem ER, eszopiclone for maintenance) — 4.3M-pregnancy safety database (Fung JAMA Psychiatry 2025; aRR 1.01 for MCMs). See Chapter One.
- First-generation antihistamines (doxylamine, diphenhydramine) — AAFP "acceptable for as-needed use." See Chapter Three.
- Sedating antidepressants (low-dose mirtazapine, trazodone) when comorbid depression. See Chapter Four.
- Benzodiazepines — only when none of the above is appropriate. See Chapter Two.
The hierarchy above is anchored in actual evidence. The herbal aisle is not.
Clinic-Ready Tools
The First Prenatal Visit Screening Question
"Are you taking any vitamins, supplements, herbal teas, CBD products, or anything from the health food store? I ask everyone — patients often don't think of these as medications, but I need to know about all of them so we can keep your pregnancy as healthy as possible."
The Patient Script — Shared Decision-Making
"I understand why you'd reach for something natural — it feels safer than a prescription. But 'natural' and 'studied in pregnancy' aren't the same thing. Valerian, chamomile supplements, passionflower, and CBD have essentially no pregnancy safety data — we don't know what they do to a developing baby because nobody has done the studies. And unlike prescription medications, supplements aren't regulated by the FDA — the amount in the pill can be very different from what the label says, and some products have been found to contain ingredients that aren't listed at all. We have much better-studied options: a structured sleep program called CBT-I that has strong evidence in pregnancy — there's an app called Sleepio that was tested in over 200 pregnant women — and if you need a medication, there are prescription sleep aids with safety data from millions of pregnancies. Magnesium as a regular supplement is fine — it's a mineral your body needs — but it's not a sleeping pill. And CBD is something we specifically recommend against in pregnancy — the medical societies are clear on that one."
EMR Documentation Template
- Valerian: No objective evidence of efficacy for insomnia (Valente, Eur Neuropsychopharmacol 2024, umbrella review of 8 SRs; AASM 2017; VA/DoD 2019). No human pregnancy safety data. AAFP 2023: "insufficient information: avoid use." → Counseled to discontinue.
- Chamomile: No evidence of efficacy (VA/DoD 2019). AAFP 2023 safety concerns: ductus arteriosus closure, increased preterm birth, low birth weight. Polyphenol-mediated prostaglandin inhibition. Cuzzolin 2010: 21.6% threatening miscarriage / preterm in regular users. → Counseled to avoid concentrated/regular use; occasional tea acceptable.
- Passionflower: Only human pregnancy report (Ozturk, Complement Ther Med 2018, 5 women): 1 neonatal death, 2 PROM, 2 MAS, 1 PPHN. Harman alkaloids are uterotonic. → Counseled to discontinue.
- Magnesium: Safe as nutritional supplement at ≤350 mg/day elemental supplemental (IOM UTL). Not evidence-based as a hypnotic. → May continue at nutritional doses (magnesium glycinate preferred).
- L-Theanine: Zero pregnancy data. GRAS for food use only. Cannot recommend or strongly warn against. → Counseled about evidence gap; better-studied options available.
- CBD: ACOG Clinical Consensus No. 10 (2025): contraindicated. Crosses placenta. CYP2C19/3A4 inhibitor. ~21% products contain undeclared THC. → Counseled to discontinue immediately.
- St. John's wort: Potent CYP3A4 inducer — will reduce levels of Z-drugs, BZDs, antipsychotics, OCs. Schäfer 2021 German claims: 5.8% major malformations in 312 exposed babies. Serotonin syndrome risk with SSRIs. → Counseled to discontinue; transition to SSRI with pregnancy data if antidepressant needed.
- "Sleep formula" / blend products: Ingredient lists unreliable; combined exposures unknown. → Counseled to discontinue.
- Discontinue [herbal agent(s)] as documented above.
- Initiate CBT-I (digital — Sleepio or equivalent — or in-person). RCT evidence in pregnancy: Felder JAMA Psychiatry 2020 (n=208, digital CBT-I d=−1.03); Manber Obstet Gynecol 2019 (n=194, 64% remission).
- If pharmacotherapy needed: Z-drug preferred (zaleplon or zolpidem IR for onset; zolpidem ER or eszopiclone for maintenance) — 4.3M-pregnancy safety database (Fung JAMA Psychiatry 2025; aRR 1.01).
- Magnesium glycinate 200–400 mg elemental daily may continue if desired (nutritional, not hypnotic).
- Screen for RLS (IRLSSG four questions) before prescribing any sleep intervention — RLS prevalence ~22% in T3.
- Document herbal supplement counseling. Reassess each visit.
Clinical Pearls
Herbal supplement use is frequently not reported to healthcare professionals, and providers often do not ask (Muñoz Balbontín 2019). At the first prenatal visit, explicitly ask: "Are you taking any vitamins, supplements, herbal teas, CBD products, or anything from the health food store?" Patients often do not consider these "medications" and will not volunteer the information unless directly asked. Document the response. This single question prevents the most common herbal-pregnancy exposure scenario: months of undisclosed use before anyone thinks to ask. Update the screen at each prenatal visit — patients start new supplements throughout pregnancy in response to evolving symptoms.
The most effective counseling framework for herbal supplements in pregnancy is not "these are dangerous" (which patients dismiss as paternalistic) but "these haven't been studied" (which respects their autonomy while conveying the evidence gap). The three phrases that land:
- "Natural doesn't mean studied, and studied doesn't mean safe in pregnancy."
- "We have prescription sleep medications with safety data from over 4 million pregnancies. We have zero pregnancy data for valerian, passionflower, or CBD."
- "The supplement industry isn't regulated the way prescription drugs are — the amount in the pill can be very different from what the label says."
Magnesium is the only agent in this chapter that can be endorsed without reservation — but only as a nutritional supplement, not as a hypnotic. Many pregnant women have inadequate magnesium intake, and supplementation at 200–400 mg/day elemental is safe and reasonable (RDA 350–400 mg/day in pregnancy). Magnesium glycinate is preferred over magnesium oxide for GI tolerability. The IOM upper tolerable supplemental intake is 350 mg/day — above this, expect diarrhea, nausea, abdominal cramping. The sleep-promoting effect, if real, is modest (SOL −17 min in 3 small RCTs of older adults; Mah 2021) and not established in pregnant populations. The honest framing: "Magnesium is a mineral your body needs, and most pregnant women don't get enough. It might help a little with sleep, but it's not a sleeping pill. Think of it as good nutrition, not a treatment."
CBD requires more direct counseling than the herbal agents because patients perceive it as a wellness product, not a drug. The decisive facts:
- ACOG (2025) says there are no medical indications for cannabis use during pregnancy — this includes CBD.
- CBD crosses the placenta and may affect placental transporters.
- ~21% of CBD products contain undeclared THC.
- CBD inhibits CYP2C19 and CYP3A4 — it can raise levels of SSRIs, anticonvulsants, and other medications.
- ~32% of pregnant women have ever used CBD; ~4% report use in the past 30 days during pregnancy.
The framing: "I know CBD is marketed as natural and safe, but the medical societies are clear on this one — it's not recommended in pregnancy. It crosses the placenta, it can contain THC even when the label says it doesn't, and it can interfere with other medications you're taking."
Even if St. John's wort were proven safe in pregnancy (it is not — Schäfer 2021: 5.8% major malformations in 312 babies vs ~3% background), its CYP3A4 induction profile would disqualify it. St. John's wort will reduce blood levels of zolpidem, eszopiclone, quetiapine, aripiprazole, midazolam, alprazolam, suvorexant, lemborexant, daridorexant, oral contraceptives, cyclosporine, antiretrovirals, and warfarin. A patient taking St. John's wort alongside any CYP3A4-substrate medication is at risk of therapeutic failure of that medication. Add the serotonin syndrome risk with SSRIs and the contraceptive failure risk and the case is overwhelming. Discontinue at the first prenatal visit. Transition to an SSRI with established pregnancy safety data if antidepressant treatment is needed.
The Ozturk & Kalayci case series (5 women, 2018) is tiny and uncontrolled — but it is the only human pregnancy data for passionflower, and the outcomes are alarming: one neonatal death, two cases of premature rupture of membranes, two infants with meconium aspiration syndrome, and one infant with persistent pulmonary hypertension. Passionflower contains harman alkaloids with documented uterotonic properties, providing a plausible mechanism. The efficacy evidence is marginal (one small RCT with a borderline p-value). There is no clinical scenario in which passionflower is the best available option for a pregnant patient.
Patients will ask: "Can I still drink chamomile tea?" The answer is nuanced. An occasional cup of chamomile tea contains far less apigenin and polyphenols than a concentrated chamomile extract capsule or multiple daily cups of strong tea. The AAFP's safety concerns (ductus arteriosus closure, increased preterm birth, low birth weight) are most relevant to regular, concentrated use — particularly in the third trimester, when the ductus arteriosus is most vulnerable to prostaglandin-synthesis inhibition. The practical answer: "An occasional cup of chamomile tea is unlikely to cause harm. Daily concentrated chamomile supplements or multiple cups of strong tea should be avoided, especially in the third trimester."
Many patients assume that anything sold at a pharmacy chain is FDA-regulated, vetted, and safe. This is not true for dietary supplements. Under DSHEA (1994), supplements do not require pre-market FDA approval for safety or efficacy. Manufacturers do not have to prove the labeled dose is accurate. The FDA can only act after evidence of harm emerges. The Erland & Saxena melatonin findings from Chapter Six — content variability of −83% to +478%, serotonin contamination in 26% of products — apply to the entire supplement industry, not just melatonin. The phrasing that works: "It's sold next to the Tylenol, but it's regulated more like a vitamin or a protein powder than like a prescription medication. The label tells you what they intended to put in it — not necessarily what they actually put in it."
References
- Muñoz Balbontín Y, Stewart D, Shetty A, Fitton CA, McLay JS. Herbal medicinal product use during pregnancy and the postnatal period: a systematic review. Obstet Gynecol. 2019;133(5):920–932. doi:10.1097/AOG.0000000000003217. 74 articles, 1,067,071 women. Almond oil PTB OR 2.09; raspberry leaf C-section AOR 3.47; heavy licorice early PTB 3.07-fold. "Should be discouraged until robust evidence of safety is available."
- Powers EA, Tewell R, Bayard M. Over-the-counter medications in pregnancy: a review. Am Fam Physician. 2023;108(4):360–369. AAFP: valerian "insufficient information: avoid use"; chamomile "ductus arteriosus closure, increased preterm birth, LBW"; melatonin "insufficient human trials"; first-gen H1 antihistamines "acceptable for as-needed use."
- ACOG Committee Opinion No. 762. Prepregnancy counseling. Obstet Gynecol. 2019;133(1):e78–e89. All supplements and herbal products should be reviewed at prepregnancy counseling.
- ACOG Clinical Consensus No. 10. Cannabis use during pregnancy and lactation. Obstet Gynecol. 2025;146(4):600–611. doi:10.1097/AOG.0000000000006053. "No medical indications for cannabis use during pregnancy and the postpartum period." Applies to CBD.
- Sarecka-Hujar B, Szulc-Musioł B. Herbal medicines — are they effective and safe during pregnancy? Pharmaceutics. 2022;14(1):171. Ginger is the most extensively analyzed; other herbs give conflicting results; all should be used with special caution.
- Chen C, Li Y, Wang W, et al. Benefits and risks of herbal medicine use during pregnancy on offspring outcomes: a systematic review and meta-analysis. Pharmacol Res. 2026;224:108092. 111 studies; adjusted observational studies do not indicate a clear increase in adverse offspring outcomes overall; a few specific herbs linked to potential safety concerns.
- Valente V, Machado D, Jorge S, Drake CL, Marques DR. Does valerian work for insomnia? An umbrella review of the evidence. Eur Neuropsychopharmacol. 2024;82:6–28. doi:10.1016/j.euroneuro.2024.01.008. 8 systematic reviews; "no evidence of efficacy for the treatment of insomnia" by objective measures.
- Ell J, Schmid SR, Benz F, Spille L. Complementary and alternative treatments for insomnia disorder: a systematic umbrella review. J Sleep Res. 2023;32(6):e13979. 15 eligible studies; "light exposure and valerian did not significantly improve sleep outcomes."
- Sateia MJ, Buysse DJ, Krystal AD, Neubauer DN, Heald JL. AASM clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults. J Clin Sleep Med. 2017;13(2):307–349. Low-quality evidence for valerian; "the majority of patients would not be likely to use valerian compared to no treatment."
- Spiess D, Winker M, Chauveau A, et al. Medicinal plants for the treatment of mental diseases in pregnancy: an in vitro safety assessment. Planta Med. 2022;88(12):1036–1046. BeWo b30 placental cells; valerian extract safe at ≤30 µg/mL; valtrate cytotoxic at ≥10 µM.
- Spiess D, Winker M, Dolder Behna A, Gründemann C, Simões-Wüst AP. Advanced safety assessment of herbal medicines for the treatment of non-psychotic mental disorders in pregnancy. Front Pharmacol. 2022;13:882997. Hyperforin inhibited trophoblast differentiation at ≥1 µM; valtrate cytotoxic at ≥10 µM; valerenic acid and linalool inconspicuous.
- Spiess D, Abegg VF, Chauveau A, et al. Transplacental passage of hyperforin, hypericin, and valerenic acid. Front Pharmacol. 2023;14:1123194. Valerenic acid equilibrates between maternal and fetal compartments; hyperforin minimal passage; hypericin does not cross.
- Zielinsky P, Piccoli AL Jr, Manica JL, Nicoloso LH. New insights on fetal ductal constriction: role of maternal ingestion of polyphenol-rich foods. Expert Rev Cardiovasc Ther. 2010;8(2):291–298. Polyphenol-rich diets (>1,089 mg/day) in T3 associated with ductal constriction.
- Cuzzolin L, Francini-Pesenti F, Verlato G, Joppi M, Baldelli P, Benoni G. Use of herbal products among 392 Italian pregnant women: focus on pregnancy outcome. Pharmacoepidemiol Drug Saf. 2010;19(11):1151–1158. 37 regular chamomile users: threatening miscarriage 21.6%, preterm labor 21.6%.
- Ozturk Z, Kalayci CC. Pregnancy outcomes in psychiatric patients treated with Passiflora incarnata. Complement Ther Med. 2018;36:30–32. doi:10.1016/j.ctim.2017.11.008. 5 women; 1 neonatal death, 2 PROM, 2 MAS, 1 PPHN. Only human pregnancy report for passionflower.
- Lee J, Jung HY, Lee SI, Choi JH, Kim SG. Effects of Passiflora incarnata Linnaeus on polysomnographic sleep parameters in subjects with insomnia disorder: a double-blind randomized placebo-controlled study. Int Clin Psychopharmacol. 2020;35(1):29–35. n=110; PSG-based TST +23.05 min vs placebo (P=0.049).
- Miroddi M, Calapai G, Navarra M, Minciullo PL, Gangemi S. Passiflora incarnata L.: ethnopharmacology, clinical application, safety and evaluation of clinical trials. J Ethnopharmacol. 2013;150(3):791–804. Excessive use during pregnancy and lactation should be avoided.
- Mah J, Pitre T. Oral magnesium supplementation for insomnia in older adults: a systematic review & meta-analysis. BMC Complement Med Ther. 2021;21(1):125. 3 RCTs, 151 participants; SOL −17.36 min (P=0.0006); TST +16 min (NS); low to very low quality evidence.
- Arab A, Rafie N, Amani R, Shirani F. The role of magnesium in sleep health: a systematic review. Biol Trace Elem Res. 2023;201(1):121–128. Observational studies suggest association; RCTs show "uncertain association."
- Makrides M, Crosby DD, Bain E, Crowther CA. Magnesium supplementation in pregnancy. Cochrane Database Syst Rev. 2014;(4):CD000937. 10 RCTs, 9,090 women; no convincing evidence of benefit for pregnancy outcomes.
- Bulman A, D'Cunha NM, Marx W, et al. The effects of L-theanine consumption on sleep outcomes: a systematic review and meta-analysis. Sleep Med Rev. 2025;81:102076. 19 RCTs, 897 participants; SOL SMD 0.15 (P=0.04); sleep quality SMD 0.43.
- Cotter J, Caddick CE, Harper JL, Ebajemito JK. Examining the effect of L-theanine on sleep: a systematic review of dietary supplementation trials. Nutr Neurosci. 2025;:1–15. 13 trials, 550 participants; "200–450 mg/day appears to be a safe and effective way to support healthy sleep in adults."
- Azubuike C, Deng A, Goodin A. Nationwide prevalence of cannabidiol use in pregnancy and in women of reproductive age. Obstet Gynecol. 2026;147(5):754–756. doi:10.1097/AOG.0000000000006140. ~32% of pregnant women have ever used CBD; ~4% in past 30 days during pregnancy.
- American Psychiatric Association. Resource document on cannabidiols. 2023. CBD crosses placenta; CYP substrate and inhibitor; THC detected in 21% of CBD-only products; FDA recommends avoiding in pregnancy.
- Lo JO, Hedges JC, Girardi G. Impact of cannabinoids on pregnancy, reproductive health, and offspring outcomes. Am J Obstet Gynecol. 2022;227(4):571–581. Cannabis use associated with SGA, PTB, fetal neurodevelopmental consequences.
- Andrade C. Maternal cannabis use during pregnancy and maternal and neonatal adverse outcomes. J Clin Psychiatry. 2024;85(4):24f15611. Small to moderately increased risks of PTB, SGA, LBW, NICU admission, fetal death.
- Schäfer W, Wentzell N, Schink T, Haug U. Characterization of pregnancies exposed to St. John's wort and their outcomes: a claims data analysis. Reprod Toxicol. 2021;102:90–97. doi:10.1016/j.reprotox.2021.04.005. 496 exposed pregnancies, 420 (85%) T1; 312 babies linked; 18 (5.8%) major malformations; 17/18 exposed in T1; 11% non-live births.
- Mannel M. Drug interactions with St John's wort: mechanisms and clinical implications. Drug Saf. 2004;27(11):773–797. CYP3A4 and P-glycoprotein induction via PXR; contraindicated with cyclosporine, antiretrovirals, OCs, warfarin.
- Kolding L, Pedersen LH, Henriksen TB, Olsen J, Grzeskowiak LE. Hypericum perforatum use during pregnancy and pregnancy outcome. Reprod Toxicol. 2015;58:234–237. 38 exposed pregnancies (DNBC); malformation prevalence 8.1% vs 3.3% (P=0.13); based on only 3 cases.
- Moretti ME, Maxson A, Hanna F, Koren G. Evaluating the safety of St. John's wort in human pregnancy. Reprod Toxicol. 2009;28(1):96–99. 54 exposed pregnancies; malformation rates 5% (SJW) vs 4% (disease comparator) vs 0% (healthy); NS.
- Bacaro V, Benz F, Pappaccogli A, et al. Interventions for sleep problems during pregnancy: a systematic review. Sleep Med Rev. 2020;50:101234. CBT-I first-line; no specific endorsement of herbal agents.
- Zhao FY, Xu P, Kennedy GA, et al. Identifying complementary and alternative medicine recommendations for insomnia treatment and care: a systematic review and critical assessment of comprehensive clinical practice guidelines. Front Public Health. 2023;11:1157419. 17 CPGs; consensus: valerian, chamomile, kava, and aromatherapy not recommended for insomnia.
- Salame A, Mathew S, Bhanu C, et al. Over-the-counter products for insomnia in adults: a scoping review of randomised controlled trials. Sleep Med. 2025;129:219–237. 51 RCTs; most studies showed positive effects but findings inconclusive; no serious adverse events.
- Felder JN, Epel ES, Neuhaus J, Krystal AD, Prather AA. Efficacy of digital cognitive behavioral therapy for the treatment of insomnia symptoms among pregnant women: a randomized clinical trial. JAMA Psychiatry. 2020;77(5):484–492. 208 pregnant women; digital CBT-I (Sleepio) vs standard care; between-group effect d = −1.03 at 18-week follow-up; benefits maintained.
- Manber R, Bei B, Simpson N, et al. Cognitive behavioral therapy for prenatal insomnia: a randomized controlled trial. Obstet Gynecol. 2019;133(5):911–919. 194 pregnant women; ISI 15.4→8.0 (CBT-I) vs 15.9→11.2 (control); 64% remission.
- Zheng X, Zhu Z, Chen J, et al. Efficacy of cognitive behavioural therapy for insomnia or sleep disturbance in pregnant women: a systematic review and meta-analysis. J Sleep Res. 2023;32(2):e13808. 8 RCTs, 743 women; ISI MD −4.25; significant improvements in sleep quality, SOL, anxiety, depression.
- Shang X, Ye L, Wang MP, Lam TH, Lai AYK. A comprehensive insight on cognitive behavioral therapy for insomnia in pregnant women: a systematic review and meta-analysis. Sleep Med. 2023;112:322–332. 9 RCTs, 978 women; CBT-I improved insomnia severity and sleep quality immediately and at short-term follow-up.
Responses