Testosterone changes linked to emotional distress in pubertal girls

News | Articles | October 19, 2026 Key Takeaways Testosterone changes linked to emotional distress: Greater increases in testosterone levels between ages 10 and 12 were associated with more symptoms of depression and anxiety in girls, even after accounting for other pubertal hormones.

Hormonal changes may be more predictive than physical maturation: The study found that hormonal fluctuations were more closely associated with emotional difficulties than visible signs of puberty, suggesting vulnerability may emerge before physical changes become apparent. Early puberty presents opportunities for intervention: Findings highlight the importance of monitoring emotional well-being during early puberty, potentially allowing pediatricians to identify distress and provide support before symptoms escalate.

SHOW MORE Researchers discuss how testosterone changes during early puberty may influence depression and anxiety symptoms in girls and implications for pediatricians. Hormonal changes during early puberty may play a more significant role in emotional development among girls than previously understood, according to new research from the University of Georgia published in Psychoneuroendocrinology.

The study found that greater increases in testosterone levels between ages 10 and 12 were associated with increased symptoms of depression and anxiety, even after accounting for other pubertal hormones. The findings suggest that hormonal changes, rather than visible signs of physical maturation, may be important indicators of emotional vulnerability during early adolescence.

1,2 Researchers analyzed data from more than 5,400 girls participating in the ongoing Adolescent Brain Cognitive Development (ABCD) Study, examining associations between pubertal hormone changes and emotional distress. Testosterone emerged as the strongest hormonal predictor of depression- and anxiety-related symptoms, even after accounting for dehydroepiandrosterone (DHEA) and estradiol.

Although the findings do not establish a causal relationship between testosterone and mental health disorders, they suggest that early puberty may represent a period of heightened sensitivity to social feedback and emotional stress. Notably, these changes may occur before physical signs of puberty become apparent, potentially making emotional difficulties harder for parents and clinicians to recognize.

The findings highlight opportunities for pediatricians to identify emerging emotional concerns and provide developmentally appropriate support before symptoms escalate into more significant mental health problems. Understanding the relationship between hormonal development, social experiences, and emotional well-being may also help inform earlier screening and preventive interventions for preadolescent girls.

In the following Q&A with Contemporary Pediatrics, study authors Avary Evans, a graduate assistant in the University of Georgia's Department of Human Development and Family Science, and Assaf Oshri, PhD, professor and director of the Georgia Center for Developmental Science, discuss the study's findings, the potential role of testosterone in emotional vulnerability, and what pediatricians should consider when supporting girls navigating early puberty.

Contemporary Pediatrics : Previous research has often focused on estrogen as a driver of mood changes during puberty. What prompted you to investigate testosterone as a potential contributor to anxiety and depression symptoms in adolescent girls? Avary Evans: Estrogen has been the default assumption for a long time, but many of those studies have actually found pretty inconsistent links between estradiol and mood in girls.

That's part of what got me interested. The other part is that androgens have mostly been studied in boys, so there was a gap: nobody was really asking what androgens do in girls. Yet girls still go through a substantial amount of androgenic change during puberty.

DHEA is a hormone that rises very early, sometimes around ages 7 and 8, and it drives some of the earliest visible signs of puberty. Testosterone rises more gradually, and it has been shown to act on neural circuits involved in reward, threat detection, and emotion regulation. This study aimed to begin exploring that.

By putting estradiol in as a covariate instead of treating it as the star, we could ask whether the androgens, DHEA and testosterone, add anything beyond what estrogen accounts for, and we found that they did. Assaf Oshri, PhD: On the broader picture, part of what we are pushing back on is the habit of treating puberty as one thing.

It's actually two axes on different clocks: the adrenal axis, which turns on around age 8 or 9 with DHEA, and the gonadal axis, which comes online later. If you collapse them, you lose the signal. Our approach was to model puberty as a coordinated system rather than pick a single hormone and call it the driver.

Contemporary Pediatrics : One of the key findings was that rising testosterone levels predicted internalizing symptoms between the ages of 10 and 12, even after accounting for estradiol and physical signs of puberty. Why do you think this early pubertal window appears to be particularly important?

Evans: Ages 10 to 12 seem to be when gonadarche really ramps up for most girls. In our sample, the average age at menarche, or first period, was a little over 11. So this seems to be when hormonal signals are moving much faster than any visible markers of pubertal maturation.

But to answer your question, what I find most interesting is that testosterone and estradiol essentially changed places in predicting internalizing symptoms. Testosterone change predicted symptoms at the earlier waves and then faded out. Estradiol did the opposite; it only showed up at the later waves.

So I don't want this to be read as 10 to 12 being the single most important window for intervention. Rather, the mechanism in that earlier window is different from the mechanism in the later window. Earlier on, the risk looks more androgen-driven, and later it may have more to do with sensitivity to fluctuating estrogen.

Pinning down what each window means helps us stop treating puberty as one long stretch of combined biological and social change and gets at what's generating risk at 10 versus what's generating risk at 13. More specifically, it helps us know how to intervene for a specific girl at a specific age.

Oshri: The methodological point underneath this is that we modeled within-person change rather than comparing girls to each other as a group. The question isn't whether a girl has high testosterone relative to her classmates; it's how fast her own system is changing over time. This is a longitudinal study.

That's a different question, and it turned out to really matter. Contemporary Pediatrics : Your study also suggested that hormonal changes and visible pubertal development may influence mental health through different pathways. How should pediatricians think about these findings when screening for anxiety and depression during routine well visits?

Evans: I think the main takeaway from our study is that hormonal change and visible development didn't move together as risk factors. Rising testosterone predicted more internalizing symptoms, but our measure of visible pubertal development was not consistently associated with those symptoms.

For us, those are two different signals. This can matter at a well visit because the physical exam is what's in front of you. A girl can still look prepubertal, and you might assume the puberty conversation can wait another year, but our data suggest that hormonal shifts are already underway and starting to carry some risk before you see physical changes take root.

So my overall suggestion would be to rely on age over appearance. At ages 10 to 12, begin screening for mood and anxiety symptoms by talking with her directly. That's where we started to see this association, and that may be earlier than when pediatricians or parents currently consider having these kinds of conversations and screenings.

Oshri: Clinically, the thing I would emphasize is that these symptoms were reported by the girls themselves, not by their parents. So a girl may already be experiencing these symptoms before anyone around her has noticed, and before the physical exam gives you a reason to ask. That's the gap.

You have to ask her directly, and you have to ask earlier than feels natural. Contemporary Pediatrics: Your data did not support routine hormone testing in clinical practice, but they do point to a period of increased vulnerability. What practical message do you hope pediatricians take away from this research when caring for girls entering puberty?

Evans: I think it's important to emphasize that I really don't want the study to be misconstrued as an argument for ordering hormone panels. There's no specific hormone threshold at which a girl is automatically at risk for mental health symptoms. What we found, as Dr.

Oshri said, is that within-person change in hormones across puberty is what tracks with symptoms. A single value from a single blood draw is not going to give you the information you need to guide mental health care. I hope this study helps change what we pay attention to, not what we order.

Where it lands is in the actual care conversation with the patient and their caregiver. When a parent brings in a 10- or 11-year-old who has started to pull back or has become a little more irritable or anxious, our first instinct may be to look for a cause, such as something at school or something with friends.

We may start thinking about the normal changes happening at home during adolescence, and sometimes those are contributing. But often, nothing specific has gone wrong. This is simply a period of major biological change that's under way even before we see it or recognize it, and mood shifts before, during, and after that change are common and expected.

Being a pediatrician who can say that with confidence does a couple of things. It can take the blame and shame off parents when talking about mental health, and it creates a natural opening to discuss mental health options and coping strategies early, rather than waiting for something to develop into a deeper-rooted, long-term problem.

Oshri: Building on Avary's message, which is clear and accurate, it's important to understand that we are developmental scientists who study child and youth development. We're interested in prevention and in producing basic science that informs intervention. But when you do this, you must separate two things that people tend to fuse together: one is distress, and the other is disorder.

What we found is not clinical diagnoses; we're not clinicians. These are precursors: the loneliness, the worry about what others think. Catching them at that stage is exactly the point, because that's when you still have room to intervene for prevention, before anything consolidates.\ Contemporary Pediatrics: What are the next research questions that need to be answered before these findings can influence clinical care, and what studies are you planning to better understand the biological mechanisms linking puberty to mental health?

Evans: There are so many exciting directions that could push this research toward more direct relevance for clinical care. The first, and I think the biggest, is measurement. These samples are collected just once a year, and we're starting to realize that hormones can fluctuate a lot across a single day and across the menstrual cycle.

There's evidence that sensitivity to small hormonal fluctuations, rather than hormone levels or even large-scale hormone change, can matter for mood and affective symptoms. So we would need denser sampling and better accounting for cycle phase, both before menarche and, once menarche has occurred, across the cycle.

Another important step is to solidify the mechanism. In our article, we relied on existing research showing that testosterone affects neural circuits involved in emotion and threat processing. Future studies will need to actually test that as a mediator, a mechanistic step between hormonal change and affective symptoms or psychopathology.

Another necessary route would be follow-up studies that examine whether this association holds in different contexts. Does this pattern look the same in rural and urban settings? Does it look the same across different races and ethnicities?

This kind of specificity matters clinically because it shapes how you tailor care to your specific patient or to your patient population as a whole. But the piece I'm most excited to work on myself is how early environments can speed up or slow down the pace of puberty. The 10-to-12 window we identified is an aggregate across more than 5,000 adolescent girls.

It's not going to be exactly the right window for every child. But if we can work out which environmental factors shift that window earlier or later, we won't have to rely on age alone to start these conversations. We can start timing prevention to when a particular girl, or a particular population of girls, is actually most likely to be moving through these windows.

Oshri: Speaking as the director of the center and as a scholar in the field, the program I'm leading is focused on the other half of the equation: what in a child's environment accelerates or delays that window. Early adversity, chronic stress, and family and neighborhood context all appear to shape the pace of maturation.

If we can model that, we can stop timing prevention by age, as Avary pointed out, and start timing it to the individual child. References Hormonal changes during puberty linked to emotional distress in young girls. News release.

University of Georgia; July 13, 2026. Accessed October 19, 2026. https://www.newswise.com/articles/hormonal-changes-during-puberty-linked-to-emotional-distress-in-young-girls Evans AI, Kogan SM, Koss KJ, House EM, Geier CF, Oshri A. Links between hormonal and pubertal development, and adolescent females’ risk for affective symptoms.

Psychoneuroendocrinology . 2026;188:107849. doi:10.1016/j.psyneuen.2026.107849 Related to this article