Hormonal Contraception and the Brain: Bridging a Gap in Steroid Neuropharmacology

In 1944, a pharmaceutical company was founded in Mexico City with a specific purpose: to use precursors found in Mexican wild yams to synthesize steroid-based medicines. The firm, Syntex Laboratories, was remarkably successful. It developed a highly economical process for manufacturing cortisone, greatly increasing access to life-saving corticosteroids. Syntex was also responsible for oxymetholone, one of the most potent anabolic-androgenic steroids ever marketed. Oxymetholone was successfully used for the treatment of anemia and HIV-associated wasting before modern advances in hematopoietic growth factors and antiretroviral therapy made it obsolete. Syntex’s crown jewel, however, was norethindrone—the first orally available hormonal contraceptive (1).

More than 80 years after Syntex’s founding, it is clear that steroid hormones exert profound effects on the brain. Yet there remains a curious imbalance in neuropsychiatric research on Syntex’s products. The central effects of corticosteroids have been extensively studied and are routinely used to model stress and depression in animals. Likewise, the effects of anabolic-androgenic steroids—particularly at supraphysiologic doses—have been the subject of longstanding investigation in both animal models and humans. In contrast, comparatively little is known about the neuropsychiatric effects of the progesterone and estrogen derivatives used in hormonal contraception.

This gap may, in part, reflect the broader tendency of medical research to neglect issues specific to women’s health. At the same time, studying the neuropsychiatric effects of contraception carries societal risk. Despite the well-established safety and efficacy of these medications, their use remains controversial in some circles outside medicine, and findings of adverse effects may be selectively leveraged to discourage their use. Against this backdrop, it is understandable that the medical establishment has emphasized the safety of hormonal contraceptives. Until February 2026, an advocacy website of the American College of Obstetricians and Gynecologists described changes in libido and mood with hormonal contraception as a myth with “no basis in clinical data” (2). The language of the website has since been revised, better reflecting evidence that a subset of women experience changes in mood and sexual function with hormonal contraception (3,4). Determining the prevalence, mechanisms, and risk factors underlying these effects will require further study. Nonetheless, it should be possible to acknowledge the essential role these medications play in reproductive freedom while still demanding a more complete understanding of their neuropsychiatric effects.

Despite the relative paucity of research, efforts to more fully characterize the central effects of hormonal contraception are emerging. A prime example is the study by Davignon et al. (5) in the current issue of Biological Psychiatry: Cognitive Neuroscience and Neuroimaging. This study builds on a body of cross-species research suggesting that the synthetic steroids found in hormonal contraception have different effects on brain function compared with their endogenous counterparts. Specifically, multiple lines of evidence have established that 17β-estradiol, the primary endogenous estrogen in adult mammals, helps facilitate extinction of conditioned fear. However, the administration of combined oral contraceptives (OCs) containing a synthetic estrogen (ethinyl estradiol) along with a progestin appears to impair fear extinction (6). Anything that facilitates or impairs extinction has important clinical implications, especially when it comes to various types of psychotherapy that often directly or indirectly engage extinction processes.

To evaluate the impact of OCs on fear regulation, Davignon et al. (5) leveraged psychophysiological and neural measures through a combination of skin conductance response (SCR) and functional magnetic resonance imaging (fMRI). Participants were grouped as current OC users, naturally cycling women in the periovulatory phase (high endogenous 17β-estradiol), naturally cycling women in the early follicular phase (low endogenous 17β-estradiol), and men. Differences emerged during extinction recall: current OC users exhibited greater recovery of fear (as measured by SCR) compared with women in the high-estradiol state, suggesting impaired fear extinction. These findings replicate the previously mentioned work showing that 17β-estradiol facilitates fear extinction, whereas OC use is associated with impaired extinction (6).

Despite structural similarities and comparable peripheral actions, ethinyl estradiol—when administered daily at doses used in most OCs—does not appear to recapitulate all central effects of endogenous 17β-estradiol at periovulatory concentrations. Several non–mutually exclusive hypotheses have been put forward as to why this is the case (7). But one clue exists that could simultaneously help identify mechanisms and provide a solution. Data from both humans and rodents suggest that acute administration of exogenous but bioidentical 17β-estradiol can improve extinction recall (6). 17β-estradiol is commonly used in menopausal hormone treatment, whereas the synthetic derivative is more common in OCs. However, formulations of hormonal contraception containing 17β-estradiol do exist. Whether replacing the synthetic ethinyl estradiol in most OCs with bioidentical 17β-estradiol would prevent deficits in fear extinction remains an open question deserving further study.

Another important question is whether OC use can have lasting neuropsychiatric effects after discontinuation. Women frequently use OCs for extended periods during adolescence and young adulthood, when the brain is still developing. Some studies have reported associations between prior long-term OC use during adolescence and changes in brain morphology (8), as well as risk for depression (9). However, establishing causality remains challenging, and at least some observed effects on brain morphology appear to be reversible (8,10). In the present study, Davignon et al. (5) explored associations between past OC use and fear extinction. Current OC users exhibited impaired extinction relative to never users, but not to past users. Further exploratory analyses suggested that past OC users in a low-estradiol state demonstrated deficits comparable to those of current users. This raises the possibility that OC exposure may have enduring effects that become most apparent during periods of low endogenous 17β-estradiol.

Alternative explanations must be considered. Without prospective, randomized assignment of OC initiation and discontinuation, causal inference is limited. Moreover, stratification by both OC history and 17β-estradiol state resulted in small subgroup sizes, and these analyses were exploratory and uncorrected for multiple comparisons. Findings related to past OC use should therefore be interpreted with caution.

Even so, Davignon et al. (5) contribute essential replication to work showing an association of current OC use with impaired fear extinction. Whether this translates into clinically meaningful differences is less clear. Notably, two studies of exposure-based therapy for spider phobia have reported reduced treatment response among OC users relative to naturally cycling women (6), suggesting that these effects may extend beyond laboratory paradigms into clinical contexts.

These findings do not justify broad changes in contraceptive practice. It would be premature to instruct patients undergoing exposure-based psychotherapy to discontinue an otherwise effective and well-tolerated OC. Decisions regarding the initiation or discontinuation of hormonal contraception should remain individualized and based on a patient’s history, reproductive goals, and preferences. However, the study’s findings do underscore important next steps. To determine clinical significance, future work should examine clinical populations, including large-scale observational studies of exposure-based therapy outcomes, as well as randomized trials assessing the impact of OC administration on treatment response in anxiety and trauma-related disorders. Mechanistic studies are also needed. In the present study, no group differences emerged in fMRI blood oxygen level–dependent responses within canonical fear-related regions, highlighting the need for more refined neurobiological approaches, potentially incorporating prospective designs and network-level analyses.

The chemists at Syntex demonstrated that structural modification of steroids could yield compounds that preserved key biological effects while improving clinical utility through enhanced potency and bioavailability. In doing so, they transformed modern medicine. Yet in prioritizing peripheral efficacy, we may have implicitly assumed that synthetic steroids are interchangeable with their endogenous counterparts. Davignon et al. (5) suggest that this assumption may not always hold in the brain.

This calls not for retreat from hormonal contraception, but for refinement. The relevant question is no longer simply whether these medications are broadly safe and effective, but which formulations, in which individuals, under which conditions, optimize both reproductive and mental health. Davignon et al. (5) do not provide definitive answers. They do, however, make clear that these are questions the field can no longer afford to ignore.