{"paper_id":"c2560723-13e9-49b5-bacc-e6ec762316f7","body_text":"Across large-scale population-based studies, adolescent HC use has been consistently associated with an elevated risk of depression diagnosis or antidepressant use compared with adults and adolescent nonusers, particularly within the first 2 years of use ( 38 ,  39 ,  40 ,  41 , 45 ). Associations appear strongest for progestin-only formulations, whereas COCs generally show weaker or null associations ( 41 ). See  Table 1  for descriptions of HC formulations. Recent registry studies further suggest a dose-dependent pattern among adolescent levonorgestrel-releasing intrauterine device (LNG-IUD) users, with high-dose systems (52 mg like Mirena) associated with a greater depression risk than low-dose systems (19.5 mg like Kyleena), especially when used as the first method of hormonal contraception ( 46 , 47 ). This finding raises important questions regarding mechanisms: One hypothesis is that higher-dose LNG-IUDs, despite their primarily local contraceptive action, result in greater systemic absorption of progestin, which could strongly influence neuroendocrine pathways. However, the precise biological mechanisms linking progestins to depression risk remain underresearched and warrant further investigation ( 48 ). Long-acting reversible contraceptives (LARCs), including the LNG-IUD, are increasingly prescribed to adolescents because of their high reliability, lack of user error, frequent use for relieving menstrual symptoms, and inclusion in some national reimbursement schemes ( 1 , 49 ,  50 ,  51 ,  52 ). Given this context, understanding potential mechanisms linking LNG-IUDs to depression risk is crucial for supporting adolescent health needs.\nAlthough the population-level association between adolescent HC use and depression appear robust, whether adolescent HC use causes depression is far from resolved. Adolescents initiating HC use often differ from nonusers in important ways, including reasons for initiation, timing of pubertal maturation, and underlying gynecological or menstrual conditions, which may confound associations between HC use and depression ( 1 ). While some registry studies statistically adjusted for medical indications such as polycystic ovary syndrome and endometriosis ( 39 ,  40 ,  41 ), these conditions are often not diagnosed formally until late adolescence due to typical early-cycle irregularity ( 53 , 54 ) (see  Consideration 2: Developmental Features of Adolescent Menstrual Cycles  and  Consideration 3: Context of Use ), meaning that residual confounding likely remains. Moreover, other relevant gynecological conditions, such as premenstrual dysphoric mood disorder (PMDD) and dysmenorrhea, were not consistently considered, despite being related to mood problems ( 55 , 56 ). Importantly, psychosocial factors, such as stressful life events, interpersonal stress, and peer and family dynamics, are not typically captured in registry studies although these factors may influence both HC use and mental health risk ( 26 , 57 ).\nFindings from these large-scale studies are challenged by findings from studies that have reported that COCs may have a positive effect on adolescents’ mood, especially in individuals with preexisting psychiatric conditions ( 58 ). Moreover, randomized controlled trials (RCTs) have generally reported little or no effect of HCs on mood, regardless of HC formulation ( 59 ). However, a systematic review and meta-analysis by de Wit  et al.  ( 59 ) found only 14 eligible RCTs on HCs and depressive symptoms published between 1961 and 2020 ( 60 ,  61 ,  62 ,  63 ,  64 ,  65 ,  66 ,  67 ,  68 ,  69 ,  70 ,  71 ,  72 ,  73 ), only one of which included an adolescent sample ( 69 ). Importantly, RCTs are prone to healthy user bias: Participants tend to be healthier than in population studies and less likely to discontinue HC use, limiting generalizability. Furthermore, adolescents who discontinue HCs due to negative side effects of HCs are especially underrepresented, further complicating efforts to disentangle adolescence as a sensitive developmental period—characterized by neurobiological and hormonal changes—from the impact of first-time HC use ( 74 ). Ethical and practical challenges compound these sampling biases further, making RCTs in adolescent populations particularly difficult to conduct.\nA recent Swedish study using an emulated target trial design sought to overcome these challenges and replicated earlier findings of increased depression risk being related to adolescent HC use, particularly during the first 2 years of use ( 75 ). Notably, adolescent HC use also predicted elevated depression risk in adulthood, consistent with other findings suggesting that HC exposure during adolescence may confer lasting vulnerability to depression even after HC discontinuation ( 74 ). While these designs represent important methodological advances, they remain constrained by residual confounding, particularly psychosocial factors that influence both HC use and depression risk. Ongoing efforts to integrate causal-inference frameworks within large-scale longitudinal data will be essential for clarifying temporal and mechanistic links between adolescent HC use and depression ( 76 ).\nWhile the increased risk of arterial thrombosis associated with HC use is well documented and routinely considered in prescribing ( 77 ), potential effects on mood symptoms remain inadequately addressed and largely absent from clinical guidelines ( 78 , 79 ). This omission is concerning given that mood-related side effects are the primary reason for HC discontinuation ( 79 , 80 ). Integrating robust evidence on HC-related mood effects into research and prescribing practices is therefore essential for aligning contraceptive care with a holistic view of adolescent well-being.\n\nPreclinical work has demonstrated that endogenous E2 and P4 exert widespread effects on the brain, influencing neuroplasticity, neurogenesis, and neurotransmission ( 81 ). E2 and P4 receptors are expressed throughout the brain, including in the amygdala, hippocampus, and prefrontal cortex (PFC) ( 2 , 3 ). The brain is sensitive to E2 and P4 changes over months and years, such as those occurring during puberty ( 11 ), pregnancy, and menopause ( 82 ), as well as shorter-term fluctuations across the menstrual cycle ( 83 , 84 ). Cyclic fluctuations have been linked to small whole-brain functional ( 85 ,  86 ,  87 ) and, to a lesser extent, structural ( 88 ) changes. Once thought to be concentrated in the medial temporal lobe ( 89 ,  90 ,  91 ), these changes are now understood to be distributed across cortical and subcortical regions, reflecting the importance of network perspectives.\nPreclinical findings have also played a central role in elucidating potential mechanisms linking HCs to changes in the brain and subsequent behavior, in particular, through interactions with HPA axis functioning and stress responsivity ( 30 , 92 , 93 ). Converging with emerging evidence from humans, rodent models have shown that HCs appear to blunt cortisol/corticosterone responses to stress, which is consistent with a reduced release of glucocorticoids and altered downstream effects on stress signaling pathways ( 30 ). For example, HCs have been associated with increased circulating FK506 binding proteins (FKBP5). FKBP5 dampens glucocorticoid receptor sensitivity and modifies the negative feedback loop of the HPA axis, resulting in altered stress reactivity ( 94 ). Recent adolescent rodent models have further demonstrated HC-related changes in HPA axis function and alterations in the fold change expression of genes related to neuroimmune, hormone, GABAergic (gamma-aminobutyric acidergic), and monoamine signaling in the hypothalamus and the medial PFC (mPFC) [see ( 93 ) for details]. Although some of these HC-related changes may be specific to adolescence, such as gene expression alterations in the hypothalamus and mPFC, further research in adult rodents is required to replicate these findings. Importantly, the analogous brain regions in humans have a high density of E2 and P4 receptors and undergo substantial maturation during adolescence. Alterations in these stress- and emotion-related brain regions have also been linked to an elevated risk of depression ( 18 ,  19 ,  20 ). Together, these findings outline potential neuroendocrine mechanisms through which HCs may shape resilience and vulnerability to depression, although direct evidence in humans, especially adolescents, remains limited.\nGiven the well-established influence of gonadal hormones on brain structure and function ( 11 , 95 ) and mechanistic evidence from rodent models ( 88 ,  89 ,  90 ), the relative scarcity of neuroimaging research on HCs in humans is striking. A 2020 systematic review ( 96 ) identified only 33 structural and functional imaging studies, with just 1 including participants under 18 years (55 users vs. 55 nonusers, age range: 13.5–15.5 years) ( 97 ). Existing evidence, drawn largely from small adult samples, provides tentative and mixed findings that HC use may relate to differences in brain structure and function, although interpretations remain unclear ( 43 , 96 ). Methodological variability, including menstrual cycle phase, HC formulation, duration of use, small sample sizes, and small effects, likely contribute to inconsistent results ( 37 , 43 ). Findings on cortical thickness have been relatively consistent, with most studies reporting a localized thinner cortex, particularly within frontolimbic regions, in HC users compared with nonusers ( 42 ).\nHowever, findings for other brain metrics have been variable. Some studies have reported lower global cortical ( 98 , 99 ) and intracranial volumes ( 100 ) or lower regional gray matter volumes in prefrontal [e.g., middle and superior frontal gyri ( 100 )], temporal [e.g., anterior cingulate and fusiform gyrus ( 101 )], and subcortical [e.g., the hippocampus ( 98 ), putamen ( 102 ), and amygdala ( 99 )] regions. However, other studies have reported volumetric increases in some of these regions ( 100 , 101 , 103 ). Functional imaging findings have also been mixed: Some studies have reported lower prefrontal activity in HC users during emotion-processing tasks ( 104 ), while others—including the only task-based adolescent study reported to date—found higher temporal lobe activity ( 97 ). Resting-state functional connectivity findings remain inconclusive, with studies reporting both higher and lower connectivity or no differences at all between HC users and nonusers ( 96 ).\nNeuroimaging research on adolescent HC use remains extremely limited. In adolescents, Marečková  et al.  ( 97 ) reported higher temporal lobe activity in COC users compared with nonusers, diverging from adult findings of reduced prefrontal activation during a similar emotion-processing task ( 104 ). More recently, Heller  et al.  conducted the first whole-brain analysis of cortical morphology in the ABCD (Adolescent Brain Cognitive Development) Study and found lower cortical thickness in the paracentral gyrus in adolescent HC users compared with nonusers after controlling for age, pubertal stage, and intracranial volume ( 105 ). Although this was the only finding surviving multiple comparison correction, the small number of HC users versus nonusers ( n  = 65 vs. 1169; mean age 14 years) reduced statistical power and generalizability. Moreover, limited data in the ABCD Study on HC formulation, duration of use, and menstrual cycle or intake phase (active or placebo pill phase) at time of scanning also restrict interpretations. Importantly, the functional significance of HC-related brain differences remains unclear, including in adults. One study in adults found that prefrontal cortical thickness differences associated with COC use were not linked to depressive symptoms ( 106 ). Considering that adolescence is a period of heightened hormonal change, brain development, and increased depression risk, this represents a critical knowledge gap.\nTo advance the field, a range of thoughtfully designed neuroimaging approaches is needed, collecting new data or leveraging existing datasets ( 18 ). Longitudinal and multimodal studies, whether large-scale or dense sampling, offer opportunities to examine developmental trajectories and within-person fluctuations. For example, the ABCD Study will be well positioned to examine how brain development and depression trajectories differ between HC users and a matched group of nonusers. However, it currently lacks detailed HC-related information from individuals, limiting making inferences beyond the group level. Moreover, a dense sampling study with an adult COC user demonstrated that brain structural dynamics across the menstrual cycle differed significantly from those observed in a naturally cycling individual. These differences were likely driven by the E2-dominant hormonal milieu associated with COC use ( 88 ). Extending such designs to adolescents would allow researchers to examine how endogenous and exogenous hormonal fluctuations relate to brain features and mood across the adolescent menstrual cycle and during HC use. It would also allow investigation of how these associations vary across HC types, formulations, and individuals. At the same time, well-powered cross-sectional studies can provide valuable mechanistic insights, such as by comparing adolescents using different HC formulations, delivery methods, or mechanisms of action. All new data collection should include careful characterization of HC formulation, duration and indication of use, endogenous and exogenous hormone levels, menstrual cycling, pubertal status, and psychosocial and clinical measures ( 37 , 107 ). Crucially, the associations between HCs, adolescent brain development, and depression must be examined within the broader developmental and social context of adolescence (see  Consideration 3: Context of Use ) and, where possible, using designs that strengthen causal inference, including within-person or prospective approaches and detailed consideration of potential confounders.\nExisting evidence underscores that little is known about how HC use interacts with depression risk and brain development during adolescence. A lack of adolescent-specific data, methodological differences, and limited attention to the developmental and social context, all constrain interpretation of existing findings. Addressing these challenges requires a shift toward more developmentally and mechanistically grounded research frameworks. In the following section, we outline 3 key considerations to guide future research on HCs, the brain, and mental health in adolescence: 1) heterogeneity in HC formulations and mechanisms of action; 2) developmental features of adolescent menstrual cycles; and 3) the broad individual, interpersonal, and societal contexts in which adolescent HC use occurs. Together, these considerations are aimed at supporting a more nuanced understanding of how exogenous ovarian hormones may modulate adolescent brain development and depression risk and inform future study design while guiding interpretations.\n\nA central challenge in interpreting findings on HCs, the brain, and depression is the substantial heterogeneity in HC active substances and formulations. As outlined in  Table 1 , HCs differ in hormonal composition, mechanism of action, and systemic hormone absorption, factors that may differentially influence neuroendocrine function, brain development, and mood, and accounting for this variation is essential. However, many studies, including global estimates ( 49 , 108 ), have not distinguished between specific HC formulations. For example, “the pill” is often reported as a single category without differentiating between COCs, which typically contain ethinyl estradiol (synthetic estrogen) and one of at least 12 progestins, whereas progestin-only pills (POPs) do not contain a synthetic estrogen. These differences are not trivial: Hormonal composition determines both the physiological mechanisms and the potential neurobiological effects of each HC type. Below, we summarize these mechanisms and refer readers to existing work for more detailed discussions ( 107 , 109 ,  110 ,  111 ,  112 ,  113 ).\nHCs can cross the blood-brain barrier, as most synthetic forms of estrogen and progestins are sufficiently lipophilic to enter the central nervous system. Once in the brain, they bind primarily to estrogen and P4 receptors, although some progestins also interact with androgen, mineralocorticoid, and glucocorticoid receptors and can modulate GABA A  receptor–related neurosteroid pathways. Most HCs act through negative feedback on the HPG axis, suppressing gonadotropin-releasing hormone (GnRH) secretion from the hypothalamus. In turn, reduced GnRH lowers luteinizing hormone and follicle-stimulating hormone release from the pituitary, inhibiting follicular maturation and, for most HC types, suppressing ovulation. (Note: For progestin-only HCs, ovulation suppression can be inconsistent across formulations, and ovulation suppression is not a typical characteristic of IUDs.)\nAs a result, endogenous E2, P4, and testosterone levels are markedly reduced, often resembling the early follicular phase in naturally cycling individuals, although the exogenous hormones introduced by most HCs create a pharmacologically stabilized hormonal milieu distinct from natural menstrual cycling. HCs also work by thinning the endometrial lining (via mechanisms that differ somewhat between COCs and POPs), often referred to as endometrial stabilization. Moreover, HCs can thicken cervical mucus to impede sperm movement. Notably, the progestins used in HCs vary in progestational potency and androgenicity (e.g., levonorgestrel is androgenic whereas drospirenone is antiandrogenic), which can have different therapeutic purposes or psychological consequences ( 110 , 111 ). We also highlight that while ethinyl estradiol is the synthetic form of estrogen used in most COCs, newer formulations can contain estradiol hemihydrate (bioidentical to endogenous E2) or the estradiol ester prodrug estradiol valerate. More recently, estetrol, a naturally occurring estrogen produced by the human fetal liver during pregnancy, and proposed to have fewer cardiovascular risks, is used ( 114 ).\n\nUnderstanding menstrual patterns typical of adolescence is essential for interpreting the effects of HCs on adolescent brain development and depression risk. Exogenous hormones introduced via HCs, and the accompanying suppression of endogenous hormones depending on formulation (see  Table 1 ), may interact with or obscure the typical hormonal fluctuations that occur as neuroendocrine systems change. Such interactions could complicate interpretation of neurodevelopmental and mood effects related to HC use.\nAdolescence is characterized by the ongoing maturation of the HPA and HPG axes, which regulate the increasing production and activity of adrenal and gonadal hormones, respectively ( 12 , 115 ). Within these broader neuroendocrine changes, menarche (first menstrual period) is a salient milestone in female development. It typically occurs between ages 12 and 13 ( 116 , 117 ), although age at menarche has been declining across recent decades ( 117 , 118 ). The years following menarche are marked by substantial hormonal variability, including irregular cycles, as feedback mechanisms within the HPG axis continue to mature ( 119 ). Early cycles are often anovulatory—up to 85% within the first 2 years postmenarche—resulting in low P4 levels and wide variability in cycle length ( 111 , 120 ,  121 ,  122 ). A systematic review of adolescent menstrual patterns ( 120 ) reported mean cycle lengths of around 34 days, with most cycles ranging from 21 to 45 days. Menstrual cycles tend to shorten and become more regular across adolescence, but it can take up to 6 years postmenarche to stabilize, with adult norms of 21 to 35 days ( 123 , 124 ).\nAlthough individuals vary in the length and pattern of their menstrual cycles, a mature cycle typically involves distinct phases: 1) follicular, marked by steadily rising E2 as ovarian follicles mature, peaking just before ovulation; 2) ovulation, triggered by LH, the dominant follicle releases an egg; and 3) luteal, when the corpus luteum (remnant of the dominant follicle) secretes E2 and P4. In the absence of fertilization, the corpus luteum regresses causing E2 and P4 to fall rapidly, resulting in menses, marking the start of a new cycle. These recurring fluctuations create a dynamic environment that remains in flux as neuroendocrine feedback mechanisms undergo refinement during adolescence ( 12 ).\nThis developmental and individual variability has direct implications for research on HCs, the adolescent brain, and depression ( 125 ). For example, the degree of endogenous hormone suppression, varying across HC formulations, may influence how exogenous hormones interact with the refinement of the HPG axis during adolescence, which could in turn influence structural and functional brain development. Moreover, adolescence is a period of heightened hormone sensitivity, and thus neurodevelopmental and mood changes may be similarly related to hormone variability rather than to absolute concentrations ( 126 ). Furthermore, individuals may differ considerably in their degree of hormonal sensitivity due to a combination of biological and environmental factors. As discussed in  Consideration 3: Context of Use , it is therefore important to contextualize how HC use relates to brain development and depression risk.\nAlthough variability in adolescent menstrual cycles can pose challenges for research ( 121 ), menstrual cyclicity offers a valuable, noninvasive window into both reproductive and general health. The menstrual cycle is increasingly recognized as a vital sign—an indicator of underlying physiological and psychosocial well-being ( 123 ). Features such as cycle length, pain, bleeding patterns, hypermenorrhea, or mood changes can reflect individual differences in hormonal regulation and could point to early signs of gynecological problems as well as to hormonally driven mood disorders such as PMDD. Intensive longitudinal designs integrating daily hormone assays (see  Box 1 ), menstrual cycle characteristics, and mood tracking are needed to ascertain how menstrual cycle regularity stabilizes during adolescence and how variability may signal risk for gynecological or mental health conditions. This is particularly important in the study of HCs, as they are often prescribed to treat these conditions ( 1 ). This work in turn could inform more personalized contraceptive counseling that integrates individual, developmental, and mental health risk profiles, and importantly, maintains follow-up to monitor the emergence and trajectory of mood symptoms after initiating or switching HCs. Box 1 Measuring Hormones and Menstrual Cycle Dynamics \n Measuring endogenous hormonal fluctuations (i.e., E2, P4, LH, FSH) across the menstrual cycle and exogenous hormones introduced by HCs is crucial for linking endocrine dynamics to adolescent neurodevelopment. However, heterogeneity in biospecimens, assay types, and sampling frequency can limit interpretability ( 82 ). Below, we outline considerations to improve reliability in developmental neuroendocrine research. Biospecimens • Blood-based measures remain the gold standard but are invasive and often impractical for repeated sampling ( 149 ). • Saliva and urine are less invasive alternatives, although hormone concentrations are typically lower. Urine measures are also affected by dilution and therefore require normalization to creatinine ( 149 , 150 ). • DUS methods allow feasible daily sampling with precision comparable to blood or liquid urine ( 121 , 150 , 151 ). Cycle staging and sampling • Peripubertal cycles can often show irregular ovulation, variable lengths, and large fluctuations in hormone concentrations. • When vaginal ultrasound is not feasible, ovulation can be inferred by an LH peak followed closely by a rise in progesterone (<7 days) ( 121 ). • Daily collection at consistent times (up to approximately 45 days) can help detect individual-specific hormonal surges and account for cycle-length variability ( 82 , 111 , 121 ). • Less intensive designs, such as detecting LH peaks within each cycle, can be more practical but provide less cycle staging ( 121 ). Assay considerations • Immunoassays are cost efficient and user friendly but may show cross-reactivity and reduced specificity ( 152 , 153 ), particularly for exogenous hormones as these molecules can differ structurally from their endogenous counterparts and may not be detected by assay antibodies ( 111 , 154 ). • MS offers greater specificity and sensitivity across biospecimens (including DUS) ( 155 ), enabling simultaneous detection of multiple hormones from lower sample volumes and circulating concentrations ( 150 , 151 ). Such precision is particularly well suited for studies evaluating menstrual cycle dynamics and comparing hormonal profiles between naturally cycling and HC users ( 154 ). • Methodological advancements continue to improve MS sensitivity and accessibility, facilitating more precise hormone assessments across developmental and contraceptive contexts. DUS, dried urine spot; E2, 17β-estradiol; FSH, follicle-stimulating hormone; HC, hormonal contraceptive; LH, luteinizing hormone; MS, mass spectrometry; P4, progesterone.\nMeasuring endogenous hormonal fluctuations (i.e., E2, P4, LH, FSH) across the menstrual cycle and exogenous hormones introduced by HCs is crucial for linking endocrine dynamics to adolescent neurodevelopment. However, heterogeneity in biospecimens, assay types, and sampling frequency can limit interpretability ( 82 ). Below, we outline considerations to improve reliability in developmental neuroendocrine research. Biospecimens • Blood-based measures remain the gold standard but are invasive and often impractical for repeated sampling ( 149 ). • Saliva and urine are less invasive alternatives, although hormone concentrations are typically lower. Urine measures are also affected by dilution and therefore require normalization to creatinine ( 149 , 150 ). • DUS methods allow feasible daily sampling with precision comparable to blood or liquid urine ( 121 , 150 , 151 ). Cycle staging and sampling • Peripubertal cycles can often show irregular ovulation, variable lengths, and large fluctuations in hormone concentrations. • When vaginal ultrasound is not feasible, ovulation can be inferred by an LH peak followed closely by a rise in progesterone (<7 days) ( 121 ). • Daily collection at consistent times (up to approximately 45 days) can help detect individual-specific hormonal surges and account for cycle-length variability ( 82 , 111 , 121 ). • Less intensive designs, such as detecting LH peaks within each cycle, can be more practical but provide less cycle staging ( 121 ). Assay considerations • Immunoassays are cost efficient and user friendly but may show cross-reactivity and reduced specificity ( 152 , 153 ), particularly for exogenous hormones as these molecules can differ structurally from their endogenous counterparts and may not be detected by assay antibodies ( 111 , 154 ). • MS offers greater specificity and sensitivity across biospecimens (including DUS) ( 155 ), enabling simultaneous detection of multiple hormones from lower sample volumes and circulating concentrations ( 150 , 151 ). Such precision is particularly well suited for studies evaluating menstrual cycle dynamics and comparing hormonal profiles between naturally cycling and HC users ( 154 ). • Methodological advancements continue to improve MS sensitivity and accessibility, facilitating more precise hormone assessments across developmental and contraceptive contexts. DUS, dried urine spot; E2, 17β-estradiol; FSH, follicle-stimulating hormone; HC, hormonal contraceptive; LH, luteinizing hormone; MS, mass spectrometry; P4, progesterone.\nBlood-based measures remain the gold standard but are invasive and often impractical for repeated sampling ( 149 ).\nSaliva and urine are less invasive alternatives, although hormone concentrations are typically lower. Urine measures are also affected by dilution and therefore require normalization to creatinine ( 149 , 150 ).\nDUS methods allow feasible daily sampling with precision comparable to blood or liquid urine ( 121 , 150 , 151 ).\nPeripubertal cycles can often show irregular ovulation, variable lengths, and large fluctuations in hormone concentrations.\nWhen vaginal ultrasound is not feasible, ovulation can be inferred by an LH peak followed closely by a rise in progesterone (<7 days) ( 121 ).\nDaily collection at consistent times (up to approximately 45 days) can help detect individual-specific hormonal surges and account for cycle-length variability ( 82 , 111 , 121 ).\nLess intensive designs, such as detecting LH peaks within each cycle, can be more practical but provide less cycle staging ( 121 ).\nAssay considerations\nImmunoassays are cost efficient and user friendly but may show cross-reactivity and reduced specificity ( 152 , 153 ), particularly for exogenous hormones as these molecules can differ structurally from their endogenous counterparts and may not be detected by assay antibodies ( 111 , 154 ).\nMS offers greater specificity and sensitivity across biospecimens (including DUS) ( 155 ), enabling simultaneous detection of multiple hormones from lower sample volumes and circulating concentrations ( 150 , 151 ). Such precision is particularly well suited for studies evaluating menstrual cycle dynamics and comparing hormonal profiles between naturally cycling and HC users ( 154 ).\nMethodological advancements continue to improve MS sensitivity and accessibility, facilitating more precise hormone assessments across developmental and contraceptive contexts.\nDUS, dried urine spot; E2, 17β-estradiol; FSH, follicle-stimulating hormone; HC, hormonal contraceptive; LH, luteinizing hormone; MS, mass spectrometry; P4, progesterone.\n\nAdolescent HC use does not occur in isolation. Inspired by Bronfenbrenners’s Ecological Systems Theory ( 127 ) and as illustrated in  Figure 1 , we conceptualize HC use as embedded within interacting individual (micro), interpersonal (meso), and sociocultural (macro) systems that jointly shape who initiates HCs, which methods are chosen, and how use and side effects are experienced. We selectively highlight contextual factors most relevant to adolescent depression and brain development, noting that this is not a comprehensive review. Figure 1 The confluence of hormonal contraceptive use, brain development, and depression risk in adolescent females occurs within a developmental ecosystem of individual, interpersonal, and sociocultural factors. (Figure created in  BioRender .)\nThe confluence of hormonal contraceptive use, brain development, and depression risk in adolescent females occurs within a developmental ecosystem of individual, interpersonal, and sociocultural factors. (Figure created in  BioRender .)\nGynecological conditions are common in adolescence but are often underdiagnosed and underresearched, despite their substantial impact on emotional, social, and academic functioning ( 55 , 128 ,  129 ,  130 ). For example, primary dysmenorrhea, the leading cause of school absenteeism among adolescent females ( 55 , 128 , 131 , 132 ) and heavy menstrual bleeding, more frequent in adolescence than adulthood ( 133 ), are associated with reduced participation in social activities and sports and higher depressive symptoms ( 128 , 134 ). However, these symptoms are often normalized as part of growing up and managed without medical consultation ( 130 ). Given the significant physical and emotional changes associated with puberty, menstrual symptoms can amplify vulnerability to distress and negatively impact well-being ( 135 ). The scarcity of adolescent-specific reproductive health research limits our ability to characterize the full spectrum of menstrual cycle patterns in adolescence, including symptoms that affect functioning and well-being regardless of whether they signal underlying pathology ( 136 ). In turn, this constrains research on brain development and mental health, as unrecognized hormonal variability may obscure key associations ( 82 , 107 ) and may limit how well clinical guidance reflects adolescent-specific developmental physiology ( 123 ). At the same time, HCs are often prescribed to adolescents for noncontraceptive indications, such as managing heavy menstrual bleeding, dysmenorrhea, and endometriosis ( 1 , 137 ). These therapeutic uses play a central role in adolescent health care, improving quality of life and daily functioning. As such, it is important to recognize both the clinical advantages of HC use and the need to balance them with careful evaluation of potential neuroendocrine or mood-related effects. Understanding how endogenous hormonal changes, gynecological conditions, and the exogenous hormones introduced by HCs interact during this sensitive developmental period is therefore critical for interpreting the broader effects of HC use on adolescent brain and mental health.\nBeyond gynecological factors, individual differences in genetic propensity for mental health conditions, pubertal timing, menstrual cycle regularity, hormonal sensitivity, and neurodevelopmental conditions may influence, and be influenced by, initiation and experience of HC use. For example, individuals who initiated HCs before the age of 19 had a higher polygenic score for depression and attention-deficit/hyperactivity disorder (ADHD) on average ( 138 ). These findings suggest potential genetic confounding. However, the interplay of genetic factors with HC use, depression risk, and brain development remains largely unexplored and represents an important direction for future research. Moreover, oral contraceptive use among female adolescents with ADHD may compound an already heightened risk for depression in this group ( 139 ). Knowledge and beliefs about HCs can also affect uptake and adherence ( 57 ); concerns about mood-related side effects, potentially amplified by online misinformation ( 27 ,  28 ,  29 ), may shape expectations, bias self-reported outcomes in research studies, and confound findings. Future studies should also assess antidepressant and other psychotropic medication use, as antidepressant prescriptions are often used as a proxy for depression diagnosis in epidemiological studies. However, the pharmacological and behavioral interactions between HCs and antidepressants remain poorly understood and should be investigated as they could potentially aid the development of tailored treatments. Together, these individual-level influences are important to consider in studies linking HCs to neurobiological or mental health outcomes.\nDecisions around HC use are embedded within interpersonal environments, and shaped by partners, family, peers, and increasingly, online influencers. Partner influence is strong, reflecting entrenched gender norms that position contraception as women’s responsibility, subject to male approval ( 57 , 140 ). Family attitudes, particularly parental disapproval of adolescent sexual activity, can deter HC use or encourage secretive HC use ( 57 ). Many adolescents may also feel embarrassed or reluctant to discuss contraception with their parents and therefore rely on peers or online sources for information.\nWhile peers, family, and online spaces can promote autonomy and shared learning among adolescents, anecdotal experiences from these sources often carry more weight than medical advice ( 27 , 57 , 141 ). Social media, in particular, can empower youth with accessible reproductive health information, but it can also amplify misconceptions surrounding HCs, including overgeneralized and exaggerated claims about physical side effects, infertility, and cancer risks ( 27 ,  28 ,  29 , 142 , 143 ). Furthermore, social media has become a prominent space for self-diagnosis and discussion of mental health conditions ( 144 ), highlighting broader shifts in how young people think about and engage with their psychological well-being. These evolving narratives extend to how young people interpret and report their mental health experiences in relation to HC use, which should be considered by researchers ( 18 ). Together, interpersonal and online influences can both facilitate and constrain HC use, shaping the socioemotional environment in which HC initiation and (dis)continuation occur.\nPatterns of HC use vary widely across geographic regions, reflecting differences in reproductive health care access and cost, education, knowledge about reproductive rights, and sociocultural norms ( 57 ). HC use is highest among adolescents in high-income countries (HICs) and lowest in low- and middle-income countries (LMICs) ( 49 ). Furthermore, use of HC types also varies regionally: OCs remain the most used HC by adolescents in HICs, although the use of LARCs has increased recently ( 50 , 51 , 145 ). In LMICs, OCs, injectables, and implants are the most prevalent HCs used by adolescents ( 146 ). Importantly, adolescent-specific epidemiological data remain limited globally, constraining granular comparisons by formulation or sociodemographic factors. For example, reports from the United Nations Department of Economic and Social Affairs typically aggregate contraceptive use across broad age bands (e.g., 15–19 and 20–24 years) and method categories (e.g., modern vs. traditional), without age-specific breakdowns by contraceptive type ( 49 ). Nonetheless, this regional variability underscores the importance of considering the sociocultural context when studying HC use. For example, judgmental attitudes and lack of specialized knowledge on HCs from health care providers can deter use ( 147 ). Broader societal structures and norms regarding, for example, religion, premarital sex, and gender roles also shape whether and how adolescents initiate, continue, or discontinue HC use.\nImportantly, societal stigma and systemic barriers not only limit HC access but also contribute to unmet health needs for adolescents with gynecological conditions or menstrual problems for which HCs are commonly prescribed ( 1 , 148 ). These macrolevel influences intersect with individual and interpersonal contexts to shape who uses HCs, for what reasons, and how their effects—both physiological and psychological—are experienced and reported. Recognizing these broader determinants is critical for situating research on adolescent HC use, brain development, and mental health within a global and developmental framework.\n\nThis is a pivotal moment in women’s reproductive health research, especially for adolescents and young adults. Advances in developmental neuroscience, including large-scale longitudinal studies and investigator-led deep phenotyping studies, provide unprecedented opportunities to conduct rigorous, transparent, and developmentally sensitive research that can transform the understanding of how HCs interact with the adolescent brain and mental health. Such progress is urgently needed, not only to improve scientific understanding but also to stem the rising tide of misinformation and empower young people to make their own informed decisions about their reproductive and mental health.","source_license":"CC-BY-4.0","license_restricted":false}