What Happens When You Take an SSRI During Pregnancy?

SSRIs are among the most commonly prescribed medications during pregnancy, and the decision to take, continue, or stop them is rarely straightforward. The evidence consistently shows that these drugs cross the placenta in meaningful amounts and can affect neonatal outcomes, but untreated depression in pregnancy carries its own serious risks, including preterm birth, low birth weight, and worsening maternal health. For most women with moderate-to-severe depression, the research points toward a nuanced calculus rather than a blanket warning: SSRI-related risks to the baby tend to be modest and often shrink once researchers account for the effects of the depression itself, while the consequences of going untreated can be severe for both mother and child.

What Untreated Depression Does to a Pregnancy

Any honest conversation about SSRIs in pregnancy has to start here, because the alternative to medication is not a risk-free baseline. Depression during pregnancy is linked to preterm birth, low birth weight, stillbirth, and increased rates of operative delivery and postpartum depression.¹ A meta-analysis comparing pregnant women with untreated depression to women without depression found that the depressed group had roughly 56% higher odds of preterm birth and nearly double the odds of delivering a low-birth-weight baby.² Depression also drives unhealthy behaviors during pregnancy, including poor nutrition, missed prenatal visits, and substance use, all of which compound the direct biological effects of the illness.³

This is the backdrop that makes the SSRI question so difficult. You are not choosing between a drug and nothing. You are choosing between the risks of the drug and the risks of an active illness. Researchers who study this topic have stressed that many women, unaware of the harms of untreated depression, refuse treatment out of fear of teratogenic effects, a decision that can itself lead to worse outcomes.⁴

Do SSRIs Cause Birth Defects?

This is probably the first fear that comes to mind, and the short answer is that for most SSRIs, the evidence is reassuring once you account for confounding. A large study of nearly 950,000 pregnancies initially found a small increase in cardiac defects among SSRI-exposed infants, but the association disappeared after restricting the analysis to women who had depression and adjusting for other risk factors. The fully adjusted relative risk of any cardiac defect with SSRI use was 1.06, meaning no statistically significant increase.⁵ That same study found no significant link between paroxetine and a specific heart defect that had been flagged in earlier work, or between sertraline and ventricular septal defects.

A meta-analysis pooling four cohort studies with nearly two million participants reached a similar conclusion: SSRIs as a class were not associated with increased risks of heart defects, with a pooled estimate of 1.06 that was not statistically significant.⁶ A separate study in the New England Journal of Medicine also found no significant associations between overall SSRI use in early pregnancy and congenital heart defects or most other categories of birth defects.⁷

Not all the data is perfectly clean, though. A large Scandinavian population study found that first-trimester exposure to SSRIs or venlafaxine was associated with a slightly higher rate of major birth defects: about 3.7% compared with 3.2% in unexposed infants.⁸ And a Danish cohort study found an apparent increase in heart malformations with SSRI exposure, but the same increase appeared in women who had stopped their SSRI before becoming pregnant, suggesting the underlying condition rather than the drug was the driver.⁹ That pattern, where the association fades once you control for the depression itself, shows up repeatedly across this literature.

Preterm Birth and Low Birth Weight

The link between SSRIs and preterm birth is somewhat more persistent than the birth-defect data, though even here the effect sizes are small. A systematic review and meta-analysis found that women taking SSRIs had higher rates of preterm delivery than controls even after adjusting for confounders, with an adjusted odds ratio of about 1.24. That held even when the comparison group was limited to depressed women not taking SSRIs, where the preterm rate was 6.8% in the SSRI group versus 5.8% in the untreated-depression group.¹⁰ In absolute terms, that one-percentage-point difference is real but modest.

A separate meta-analysis looking at antidepressants more broadly found a similar pattern for low birth weight, with a pooled relative risk of about 1.44.¹¹ However, the heterogeneity across studies was high, meaning results varied widely from one study to the next, and nearly a third of the included studies individually found no significant link. A large sibling-comparison study, which controls for all the genetic and environmental factors siblings share, did confirm a modest association between first-trimester SSRI exposure and preterm birth but found no link to babies being born small for their gestational age.¹²

What Happens to Newborns Right After Delivery

One of the most visible effects of SSRI exposure shows up in the first few days of life. Some newborns experience what is called neonatal adaptation syndrome: a cluster of symptoms including jitteriness, irritability, feeding difficulty, and respiratory issues. In one study, about 30% of SSRI-exposed term infants showed some symptoms of this kind, though most cases were mild and none required treatment. The symptoms peaked within the first 48 hours of life and resolved on their own.¹³ Paroxetine was disproportionately represented among infants with more severe symptoms in that study, an observation consistent with its known shorter half-life, which can make the newborn’s adjustment more abrupt.

A rarer but more worrying neonatal outcome is persistent pulmonary hypertension of the newborn, or PPHN, a condition where the baby’s circulation does not switch over normally to breathing air after birth. An early, widely cited study found a six-fold increase in PPHN among infants exposed to SSRIs after the 20th week of pregnancy.¹⁴ A later study adjusted for maternal depression and other factors and still found a four-fold increase with late-pregnancy SSRI use, though not with use before 20 weeks.¹⁵ Even with these elevated odds ratios, the condition is extremely rare in absolute terms. PPHN affects roughly one to two per thousand live births overall, so a several-fold increase of a very small baseline number still translates to a low absolute risk. Still, it is the single most serious neonatal complication linked to late-pregnancy SSRI use, and it is one reason clinicians pay attention to the timing of exposure.

Postpartum Bleeding

A less commonly discussed risk involves the mother rather than the baby. Serotonin plays a role in platelet function and blood clotting, and SSRIs reduce serotonin uptake by platelets. Several studies have found modestly higher rates of postpartum hemorrhage among women taking SSRIs. A large Swedish registry study found that postpartum hemorrhage prevalence was about 9.1% among SSRI-treated women, compared with 7.0% among healthy women and 7.6% among women with psychiatric illness but no SSRI use.¹⁶

A hospital-based study found more dramatic numbers: roughly a two-fold increase in postpartum hemorrhage and postpartum anemia among SSRI users who delivered vaginally, with average blood loss about 86 milliliters higher in the SSRI group.¹⁷ And preliminary results from a registry at Massachusetts General Hospital suggested the bleeding risk may be especially elevated after cesarean section, with an adjusted odds ratio above 2.¹⁸ The clinical takeaway is straightforward: obstetric teams should be aware if a patient is on an SSRI so they can plan for potentially higher blood loss at delivery, especially during surgical births. This is a manageable risk when anticipated, not a reason to stop medication.

The Autism and ADHD Question

Few topics in prenatal SSRI research have generated more anxiety than the possible link to autism spectrum disorder. Early studies did find statistical associations. One case-control study reported that SSRI exposure was nearly three times as common among boys later diagnosed with autism compared with typically developing boys.¹⁹ A Canadian cohort study found about a two-fold increase in autism risk associated with SSRI use in the second and third trimesters.²⁰

But these were observational studies that struggled to separate the drug’s effect from the underlying illness. Mothers who take SSRIs have depression or anxiety; those conditions are themselves linked to neurodevelopmental outcomes in children, possibly through shared genetics or prenatal stress biology. When researchers used designs that control more tightly for these confounders, the associations weakened or vanished. A large Scandinavian study comparing women who used SSRIs during pregnancy with those who used them before but stopped found no significant increase in autism risk after full adjustment.²¹ Sibling studies, which compare children born to the same mother with and without SSRI exposure, found no significant difference in autism rates between exposed and unexposed siblings.²² Another large sibling comparison similarly showed no association between first-trimester SSRI use and autism or ADHD.²³

A major review synthesizing the evidence on both autism and ADHD concluded that prenatal antidepressant use is unlikely to substantially increase the risk of either condition, and that earlier findings were probably driven by the maternal illness rather than the drug.²⁴ The ADHD picture follows a similar trajectory: a systematic review found that the apparent association between prenatal antidepressant exposure and ADHD disappeared in sibling analyses and in studies restricted to women with a history of mood disorders.²⁵ One Norwegian cohort study did find a somewhat elevated ADHD risk in SSRI-exposed children around ages seven to nine, but also found that maternal depression before pregnancy independently predicted ADHD regardless of medication use, complicating the interpretation.²⁶

Why the Confounding Problem Is So Stubborn

Almost every risk attributed to SSRIs in pregnancy faces the same methodological problem: women who need these drugs are different from women who do not. They have depression or anxiety, they may have poorer sleep, higher cortisol, different eating habits, and different genetic backgrounds. All of these factors can independently affect pregnancy and child development. You cannot randomly assign pregnant women to take SSRIs or a placebo, so every study in this area is observational, and observational studies cannot perfectly separate the drug from the disease it treats.

Sibling studies help because they compare children born to the same mother under different exposure conditions, which automatically controls for genetics and stable environmental factors. When sibling designs are applied to the autism, ADHD, and birth-defect questions, the associations with SSRIs mostly disappear.²⁷ The one exception in sibling analyses is preterm birth, which retains a modest association even after controlling for shared family factors, making it perhaps the most credible standalone risk of SSRI exposure itself.

What Happens When You Stop an SSRI During Pregnancy

Some women decide to discontinue their SSRI once they learn they are pregnant, and the research on what happens next is sobering. A Danish study of matched pairs found that women who stopped antidepressants during pregnancy had a 25% higher risk of a psychiatric emergency during the perinatal period compared with women who continued their medication. The risk was particularly elevated during the pregnancy itself.²⁸ A multinational Scandinavian study found that women who had been on stable long-term antidepressant therapy and then stopped late in pregnancy had a moderately increased likelihood of needing psychiatric medications postpartum, with the risk especially pronounced among women with a history of mood disorders.²⁹

A relapse of severe depression during pregnancy is not just a psychological hardship. It raises cortisol, disrupts sleep, impairs nutrition, and in severe cases can lead to self-harm. These indirect effects can be more damaging to the pregnancy than the SSRI itself. The decision to stop medication should be made with a clinician who can weigh the severity of the illness against the treatment risks, not made reflexively out of alarm.

Dose Changes and Drug Levels in Late Pregnancy

Your body handles SSRIs differently as pregnancy progresses. Blood volume expands, kidney filtration increases, and liver enzymes that break down many drugs become more active. The net result is that SSRI blood levels can fall in the second and especially third trimester, even if you are taking the same dose. Some women experience a return of depressive symptoms late in pregnancy for this reason alone.³⁰ Your prescriber may need to increase your dose in the third trimester and then scale it back after delivery, when your metabolism returns to its pre-pregnancy state. Therapeutic drug monitoring, where blood levels are periodically checked, is something some specialists recommend during pregnancy to keep medication effective without overshooting.

How Patients Perceive the Risks

The way women weigh SSRI risk tends to be driven more by fear than by evidence. A study of women calling a teratogen information service found that even after receiving evidence-based counseling, about 15% of antidepressant users chose to stop their medication, compared with just 4% of women calling about stomach medications and 1% calling about antibiotics.³¹ The main factors driving their decisions were not the counseling itself but information received beforehand from family, friends, and the internet. This suggests that the cultural narrative around antidepressants in pregnancy, which tends to emphasize danger, is hard to override even with professional reassurance.

There is also a natural asymmetry in how people think about drug risk versus disease risk. Taking a pill feels like doing something active to your baby; letting a disease run its course feels passive and therefore safer, even when the evidence says otherwise. Clinicians who work in perinatal psychiatry often have to spend as much time helping patients understand the risks of untreated depression as they do discussing the drug’s profile.

Non-Drug Options and When They Apply

For women with mild-to-moderate depression, non-pharmacologic approaches may be worth trying, either alone or alongside a reduced medication dose. A review of randomized trials found some evidence supporting acupuncture, massage, and bright light therapy for reducing depression symptoms in pregnancy, though the evidence base remains small.³² Cognitive behavioral therapy and interpersonal therapy have the strongest track records among talk therapies for perinatal depression.³³

The limitation is practical: moderate-to-severe depression, the kind that interferes with daily functioning or involves suicidal ideation, rarely responds adequately to therapy or lifestyle changes alone. For women in that category, SSRIs remain a first-line treatment during pregnancy. The goal is matched intensity: mild illness gets low-intensity intervention, severe illness gets medication, and moderate illness involves an honest conversation with a provider about what the patient is willing to try and what the stakes are if treatment fails.

Breastfeeding After Taking SSRIs in Pregnancy

Women who took an SSRI during pregnancy often wonder whether it is safe to continue while breastfeeding. The evidence here is more straightforward than the pregnancy data. Most SSRIs produce very low or undetectable levels in nursing infants’ blood. Sertraline and paroxetine are generally considered the best first-line options for breastfeeding mothers because of their especially low transfer into breast milk.³⁴ Fluoxetine and citalopram tend to produce somewhat higher infant levels and have been associated with occasional reports of irritability or sedation in nursing infants, though adverse effects remain uncommon overall.

Switching SSRIs solely because of breastfeeding plans introduces its own risk, since changing medications can trigger a period of instability. If a woman is doing well on fluoxetine, for instance, the decision to switch to sertraline should account for the chance that the new drug may not work as well. Many specialists advise continuing whichever SSRI has been effective during pregnancy and monitoring the infant for any unusual symptoms rather than making a prophylactic switch.

Epigenetic Research and Where It Stands

An emerging area of investigation looks at whether prenatal SSRI exposure leaves molecular marks on the baby’s DNA, specifically changes in DNA methylation that do not alter the genetic code but can influence how genes are expressed. One study comparing cord blood from SSRI-exposed, depression-exposed but unmedicated, and unexposed newborns found distinct methylation differences across all three groups, with 18 differentially methylated regions unique to the SSRI-exposed group and 53 unique to the depression-only group.³⁵ What is interesting here is that depression without medication produced more methylation changes than the drug itself. Whether any of these molecular differences translate into meaningful health outcomes for the child is entirely unknown at this point. This research is in its earliest stages, and no clinical decisions should rest on it. But it does reinforce the broader theme of this literature: depression is not a neutral backdrop against which SSRIs act. It is biologically active, and its fingerprints show up even at the molecular level.