Lead
A large, registry-based study spanning 1996 to 2018 in Finland examined more than 1.27 million singleton births to assess the effects of selective serotonin reuptake inhibitor antidepressant use during pregnancy. Researchers compared pregnancies with two or more SSRI purchases during gestation to women with diagnosed depression who did not use antidepressants and to women who discontinued SSRIs before pregnancy, and included sibling comparisons. The study found that prenatal SSRI exposure was linked to a modestly higher risk of gestational diabetes and to neonatal signs of delayed adaptation, while it was associated with lower risks of preterm birth and low birth weight. The team concludes that SSRI use has effects beyond maternal depression that require individualized clinical weighing of maternal mental health and neonatal monitoring.
Key Takeaways
- The study used Finnish national registers and covered 1,272,587 singleton live births from 1996 to 2018.
- Women with two or more SSRI purchases during pregnancy numbered 19,020 and were compared with 19,625 depressed but unmedicated women and 3,145 women who discontinued SSRIs before pregnancy.
- SSRI use during pregnancy was associated with higher odds of gestational diabetes, adjusted OR 1.20, 95% CI 1.09–1.32, after accounting for depression severity indicators.
- SSRI exposure correlated with increased risk of neonatal adaptation problems: low 5-minute Apgar (OR 2.02, 95% CI 1.78–2.30), breathing problems (OR 1.61, 95% CI 1.48–1.75), and neonatal care unit treatment (OR 1.23, 95% CI 1.16–1.31).
- For birth timing and size outcomes, SSRI use was linked to lower rates of caesarean section, very preterm birth (<32 weeks), late preterm birth (32–36+6 weeks), and low and very low birth weight.
- Third trimester SSRI exposure further raised the risk of a low 5-minute Apgar score, OR 3.44, 95% CI 2.93–4.04.
- Compared with women who stopped SSRIs before pregnancy, continued SSRI use associated with lower odds of late preterm birth (OR 0.83, 95% CI 0.70–0.999) and low birth weight (OR 0.78, 95% CI 0.64–0.96).
- No increased risk of major congenital malformations was detected among SSRI-exposed newborns in this cohort.
Background
Prenatal antidepressant safety has been contested for years because observational studies must untangle the direct effects of medication from effects of underlying maternal psychiatric illness. Depression in pregnancy is itself associated with adverse obstetric and neonatal outcomes, including risk of preterm birth, low birth weight, and impaired maternal self-care, which complicates causal inference about drugs. Randomized trials of SSRI use in pregnancy are ethically and practically limited, so large registry studies that apply multiple comparison groups and within-family designs are used to approximate causal effects.
Previous research has reported mixed signals: some studies linked SSRIs to neonatal adaptation syndromes and rare congenital malformations, while others showed reduced preterm birth compared with untreated depression. Confounding by indication, severity of illness, smoking, socioeconomic factors, and comorbid conditions remains a major challenge. The Finnish registers allow detailed linkage of prescriptions, diagnoses, birth outcomes, and family relationships, which makes sibling comparisons and multiple control groups feasible.
Main Event
The international team, led by investigators at the University of Turku with collaborators at Columbia University, identified women with two or more SSRI purchases during pregnancy as exposed cases. They contrasted outcomes with women diagnosed with depression who did not fill antidepressant prescriptions during pregnancy and with a subgroup who discontinued SSRIs before conception. The authors also performed sibling-pair analyses to control for shared familial factors such as genetics and early environment.
Across 1,272,587 singleton births, adjusted analyses that accounted for indicators of depression severity found an increased odds of gestational diabetes among SSRI users, with ORs clustered around 1.14 to 1.20 depending on the comparison. Simultaneously, SSRI exposure was associated with lower odds of several adverse perinatal outcomes linked to prematurity and small size for gestational age, including very preterm birth and low birth weight.
Newborn clinical adaptation signals were consistently higher after SSRI exposure: infants had greater odds of low 1- and 5-minute Apgar scores, respiratory problems at birth, and need for neonatal care or neonatal intensive care. Notably, exposure in the third trimester amplified the risk of a low 5-minute Apgar score to OR 3.44. Despite these adaptation signs, the study did not find an elevated risk of major congenital anomalies.
Analysis & Implications
The contrasting pattern of findings indicates that SSRIs may exert multiple biologically mediated effects in pregnancy. Reduction in preterm birth and low birth weight suggests a possible protective pathway, potentially via stabilization of maternal stress biology, improved prenatal care engagement, or direct placental effects that favor term delivery. Conversely, the increased neonatal adaptation problems point to perinatal pharmacologic influences on infant serotonergic signaling, cardiorespiratory regulation, or withdrawal phenomena after in utero exposure, especially when exposure extends into the third trimester.
The observed association with gestational diabetes, which persisted after adjustment for depression severity, raises new clinical and mechanistic questions. Antidepressant-related metabolic effects have been reported in nonpregnant populations, but causality, timing, and agent-specific risks in pregnancy remain unclear. If confirmed, this would argue for closer metabolic screening for pregnant women taking SSRIs and for research into whether risk varies by agent or dose.
Clinically, these results reinforce that treatment decisions in pregnancy must be highly individualized. For some patients, continuing an effective SSRI may reduce the elevated preterm birth risk linked to untreated depression. For others, particularly where metabolic risk or late-pregnancy fetal adaptation is a concern, alternative strategies, dosage adjustments, closer glucose and fetal monitoring, or timing of exposure may be considered. Shared decision making that documents risks and benefits and arranges enhanced perinatal surveillance is essential.
Comparison & Data
| Outcome | Comparison | Reported OR (95% CI) |
|---|---|---|
| Gestational diabetes | SSRI vs depressed unmedicated | 1.14 (1.07–1.22); adjusted 1.20 (1.09–1.32) |
| Low 5-minute Apgar | SSRI vs depressed unmedicated | 2.02 (1.78–2.30); third trimester 3.44 (2.93–4.04) |
| Late preterm birth | SSRI vs discontinued before pregnancy | 0.83 (0.70–0.999) |
| Low birth weight | SSRI vs discontinued before pregnancy | 0.78 (0.64–0.96) |
The table highlights estimated effect sizes from primary comparisons reported in the paper. The authors adjusted for multiple confounders and performed sibling analyses to reduce bias from shared familial risk. Estimates near 1 indicate small to moderate associations, while confidence intervals that exclude 1 suggest statistical precision. Readers should note that odds ratios summarize relative odds, which are not direct measures of absolute risk; absolute risk differences for individual patients depend on baseline probabilities.
Reactions & Quotes
Our results emphasise the significance of individualised treatment decisions during pregnancy. The treatment of depression is important, and the use of SSRIs seems to protect against the risk of preterm birth associated with depression. At the same time, however, it is necessary to closely monitor both the progress of the pregnancy and the health of the newborn.
Docent Heli Malm, lead author, University of Turku
The association we have observed with gestational diabetes requires further research in order to better understand the possible cause and effect relationship and underlying biological mechanisms.
Docent Heli Malm, University of Turku
The authors framed findings as relevant for clinical counseling rather than as a directive for wholesale medication changes. Independent clinicians who reviewed the paper noted that the study strengthens the case for monitoring and for weighing the risks of untreated maternal depression, which itself increases obstetric risks.
Unconfirmed
- Whether SSRI exposure directly causes gestational diabetes through a specific metabolic mechanism remains unproven and requires dedicated mechanistic and agent-specific studies.
- Longer term neurodevelopmental consequences of neonatal adaptation signs after SSRI exposure are not addressed by this study and remain uncertain.
- Applicability of these results to non-SSRI antidepressants or to variations in dose and specific drug agents is not established in this dataset.
Bottom Line
This large, carefully controlled registry study indicates that prenatal SSRI use is associated with a dual pattern: modestly increased maternal metabolic risk and neonatal adaptation signs, alongside lower odds of preterm birth and low birth weight. The persistence of several associations after adjustment and in sibling analyses implies that at least some effects are attributable to the medication itself and not solely to maternal depression.
For clinicians and pregnant patients, the study supports individualized decision making that balances maternal mental health benefits with perinatal monitoring strategies, including glucose surveillance and planning for potential neonatal support when exposure continues into late pregnancy. Further research is needed to clarify mechanisms, agent-specific risks, and long-term outcomes so that counseling can be progressively refined.
Sources
- Neuroscience News (media summary of the University of Turku release)
- Malm H et al., American Journal of Obstetrics and Gynecology MFM (peer reviewed article, open access)
- University of Turku (institutional research centre and press source)