The research, drawn from the MONTRA survey conducted in France between 2010 and 2011, analyzed data from 796 women of childbearing age, defined as 18 to 50 years old, all diagnosed with bipolar disorder according to DSM-IV criteria. A randomly selected national sample of 439 psychiatrists, drawn from a registry of 12,000 practitioners, recorded detailed sociodemographic, clinical, and pharmacological information over four consecutive days of routine care. The scale and representativeness of this design give the findings unusual weight for a question that has long been studied only in small, single-center samples.

The headline number is 61.2 percent: the proportion of women reporting current use of effective contraception, as defined by World Health Organization criteria. Effective methods included combined oral contraceptives, progestogen-only pills, patches, vaginal rings, implants, intrauterine devices, female sterilization, condoms, and diaphragms. Withdrawal, notably, did not count. That figure sits slightly below the roughly 70 percent rate reported in the general French population, and within the range of 52 to 73 percent documented in earlier studies of women with bipolar disorder. In a population where unplanned pregnancy carries disproportionate risks, the gap is clinically meaningful.

Why does this matter so much? Bipolar disorder is a chronic, severe condition marked by alternating manic, hypomanic, and depressive episodes, and the perinatal period is a well-established window of vulnerability. Discontinuing pharmacological treatment during pregnancy or after delivery sharply increases the risk of illness recurrence. Meanwhile, some of the most effective mood stabilizers — anticonvulsants such as valproate and carbamazepine, and lithium itself — carry teratogenic risks or interfere with contraceptive efficacy. An unplanned pregnancy in this population can therefore force an agonizing choice between maternal mental health and fetal safety, a choice that effective contraception is designed to prevent.

The statistical core of the new study is a logistic regression model that identified which variables independently predicted contraceptive use. Women aged 26 to 35 were more than twice as likely to use effective contraception as the reference group (odds ratio 2.40), and those aged 36 to 45 were nearly twice as likely (odds ratio 1.92), both highly significant. Education produced the single largest effect: women with higher educational attainment had more than five times the odds of using effective contraception compared with those with only primary education (odds ratio 5.07). Being professionally active (odds ratio 1.63) and not living alone (odds ratio 1.41) also raised the odds, while a greater number of manic or mixed episodes halved them (odds ratio 0.50).

What is striking is what did not predict contraceptive use. The type of bipolar disorder, history of hospitalization, suicide attempts, rapid cycling, and comorbid anxiety or substance use disorders showed no significant association. Nor did the nature of pharmacological treatment. Around 28 percent of the women were prescribed medications with high or intermediate teratogenic risk, yet there was no difference in contraceptive use between those taking such drugs and those who were not. Only 3 percent received valproate, suggesting clinicians generally avoid the most dangerous option, but the absence of any link between teratogenic prescribing and contraception hints that reproductive risk is not systematically factored into treatment planning.

The authors argue this points to a blind spot in psychiatric practice. Previous research has found that a substantial share of women with mental illness — over 32 percent in one cited study — had never received information on sexual education or family planning, and that psychiatrists often avoid the topic, whether because they lack knowledge of contraceptive options, view it as outside their remit, or feel uncomfortable raising it. The new findings suggest that this counseling gap mirrors broader social inequalities: the same demographic factors that shape contraceptive behavior in the general population — age, education, employment, and social support — appear to operate with equal or greater force among women with bipolar disorder.

One novel observation deserves attention. Women who were judged autonomous in managing their illness and their medication were significantly more likely to use effective contraception, even though autonomy did not emerge as an independent predictor in the final regression model. More than a quarter of participants lacked autonomy in illness management, and more than 10 percent lacked autonomy in treatment management, reflecting the well-documented functional impairment that bipolar disorder imposes. If contraceptive behavior tracks with the capacity to self-manage a chronic illness, then contraceptive counseling may need to be tailored to functional status, not just to reproductive history.

The study’s limitations are candid and considerable. The data were collected 15 years ago, before long-acting reversible contraceptives such as hormonal and copper IUDs and subdermal implants rose from under 4 percent to around 10 percent of contraceptive uptake in comparable settings, and before European and global regulators tightened restrictions on valproate prescribing in women of reproductive potential. The cross-sectional design rules out causal inference, the contraceptive data did not specify exact methods, and including women up to age 50 may have captured some in perimenopause. Clinical and functional constructs were not measured with standardized instruments. These caveats temper the precision of the estimates but not the direction of the central finding.

Even so, the public health implications are hard to ignore. Women with bipolar disorder experience higher rates of unplanned pregnancy, adverse obstetric outcomes including gestational hypertension and hemorrhage, and elevated postnatal relapse risk. The authors call for systematic integration of contraceptive counseling and family planning into psychiatric care, closer collaboration between psychiatry and gynecology, and targeted training so that mental health professionals routinely assess reproductive health alongside mood symptoms. In a condition where the stakes of an unplanned pregnancy extend to both mother and child, contraception is not a peripheral concern — it is relapse prevention. The message of this study is that the women least likely to be protected are those already most disadvantaged by their social circumstances, and that closing that gap may be one of the most consequential interventions psychiatry can offer.

Subject of Research: Contraceptive use among women of childbearing age with bipolar disorder

Article Title: Contraceptive use in women of childbearing age with bipolar disorder

Article References: Rodríguez-Toscano, E., de Chazeron, I., Belzeaux, R., Llorca, P. M., & Samalin, L. (2026). Contraceptive use in women of childbearing age with bipolar disorder. BMC Psychiatry, 26(1), Article 773. https://doi.org/10.1186/s12888-026-08348-8

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08348-8

Keywords: bipolar disorder, contraception, unplanned pregnancy, women's health, psychiatry, reproductive health, teratogenic risk, mood stabilizers, social determinants, perinatal mental health, MONTRA survey, public health