Testosterone levels linked to PTSD symptoms in both men and women, study finds

A new study published in Translational Psychiatry suggests that testosterone may play a more complicated role in mental health than previously understood. Researchers found that both low and high testosterone levels were associated with higher levels of post-traumatic stress symptoms, while mid-range levels were linked to the fewest symptoms. The findings, which were consistent for both men and women, raise questions about whether hormonal balance might shape the risk for stress-related mental health problems—and whether this relationship might be influenced by body weight.

Testosterone is a hormone often associated with physical traits such as muscle mass and male sexual development, but it also has important roles in brain function and emotional regulation. In both men and women, testosterone is believed to affect how the body and brain respond to stress. Past studies have suggested that testosterone may suppress the activity of the stress-response system known as the hypothalamic-pituitary-adrenal axis, or HPA axis, which controls the release of stress hormones like cortisol.

Because stress and trauma are known to affect the risk of developing post-traumatic stress disorder, or PTSD, some researchers have speculated that testosterone might influence how likely someone is to develop PTSD symptoms. However, existing evidence has been inconsistent, often limited by small sample sizes, military-only populations, or by including only male participants.

The new study aimed to address these gaps using a large, civilian sample that included both men and women, and by exploring more complex patterns in the data. The researchers also looked at whether body mass index, or BMI, might alter the relationship between testosterone and PTSD symptoms.

“We know from previous research that naturally occurring hormones – including testosterone and estrogen — can influence the risk of developing psychiatric disorders such as depression, PTSD, and anxiety disorders,” said study author Laramie Duncan, an assistant professor at Stanford University. “But most of these studies were done decades ago on small numbers of people so the results weren’t reliable. We took advantage of a large, modern study and evaluated data from over 100,000 people to more definitively answer the question of whether PTSD and depression symptoms are correlated with natural testosterone levels.”

The researchers used data from the UK Biobank, a large research database that contains genetic, health, and lifestyle information from about half a million adults in the United Kingdom. From this dataset, they focused on 130,471 participants who had both testosterone measurements and completed a mental health questionnaire in 2016 that included items related to PTSD symptoms. To ensure the reliability of their analysis, they excluded people whose testosterone levels were unusually high or low compared to the rest of the population.

PTSD symptoms were assessed using a short survey that asked about emotional responses such as irritability, avoidance, feelings of detachment, upsetting memories, and difficulty concentrating. Although this measure does not capture the full diagnostic criteria for PTSD, it provides a general indication of symptom severity across the population.

The researchers examined three different types of testosterone metrics: total testosterone, calculated free testosterone (which estimates the biologically active portion of the hormone), and the free androgen index, which is another way of estimating testosterone availability. They looked at how these hormone levels related to PTSD symptom scores using both standard statistical models and models that allowed for nonlinear relationships, such as U-shaped curves.

They also adjusted their analysis for several factors known to affect testosterone, including age, time of day the blood sample was taken, socioeconomic status, ethnicity, and BMI. Finally, they tested whether the relationship between testosterone and PTSD symptoms differed depending on a person’s BMI category (underweight, healthy weight, overweight, or obese).

The researchers found that the relationship between testosterone and PTSD symptoms followed a U-shaped curve. In both men and women, people with mid-range testosterone levels reported the lowest levels of PTSD symptoms, while those with either very low or very high levels had higher symptom scores. This pattern held true across all three testosterone measurements.

“A talented PhD student in our lab, Hanyang Shen, thought to ask whether there might be a non-linear relationship between testosterone and mental health symptoms, and there was,” Duncan told PsyPost. “She found that ever-higher testosterone was not correlated with better mental health symptom scores. Rather, our results suggest that only especially low testosterone may be a concern for mental health symptoms, for most people at least.”

For example, when participants were divided into ten equal groups based on testosterone levels, those in the seventh decile (just above the middle of the distribution) had the lowest average PTSD scores. Those in the lowest decile had the highest scores. These patterns remained significant even after accounting for other variables such as age and BMI.

“The nature of the relationship we observed between testosterone and mental health symptoms was very similar in males and females,” Duncan noted. “This is interesting because testosterone is often thought of as a male hormone, but it’s also important for women.”

When the researchers looked more closely at the interaction between testosterone and BMI, they found that the association between testosterone and PTSD symptoms was stronger among people with higher BMI. In other words, for individuals who were overweight or obese, the link between out-of-range testosterone levels and PTSD symptoms was more pronounced. This finding was consistent across both men and women.

The researchers also performed several additional analyses to ensure their findings were robust. These included looking at whether similar patterns emerged when they used anxiety and depression scores instead of PTSD symptoms. The results were largely similar, suggesting that the relationship between testosterone and mental health symptoms may not be limited to PTSD alone.

While the study benefited from an unusually large and well-characterized dataset, the authors caution against drawing firm conclusions about cause and effect. One major limitation is that testosterone and PTSD symptoms were measured at different time points. Testosterone was assessed during the initial recruitment phase between 2006 and 2010, while PTSD symptoms were measured in 2016. Because the study did not track participants over time with repeated measurements, it is unclear whether hormone levels predicted future symptom development or reflected existing mental health issues.

“We found a small correlation between lower testosterone levels and higher symptom scores for both PTSD and depression,” Duncan told PsyPost. “However, we want to be 100% clear that the results do not say whether low testosterone causes PTSD or depression (or vice versa) but do suggest there is a relationship between them. The results are consistent with other studies showing that treating clinically low testosterone can have mental health benefits. It is important to know that testosterone supplementation is not effective or safe for everyone with mental health symptoms, which is why it is critical to talk to a doctor before taking them.”

Another limitation is that the study could not include cortisol or other stress hormone data. Some researchers suggest that the balance between testosterone and cortisol may be more informative than either hormone alone. Other hormones, such as estrogen, could also play a role.

“We would really like to measure the effect of estrogen on women’s mental health symptoms,” Duncan said. “This is currently almost impossible, however, even with a sample size of over 100,000 people (as we had here) because estrogen levels fluctuate so much across shorter (monthly) and longer (lifetime) timescales in women.”

“Also, the relationship between estrogen and mood seems to be more complex than with testosterone because the fluctuations in estrogen may be as important (or more) than the levels themselves. We need a device like a continuous glucose monitor to really understand how differences estrogen levels impact mental health symptoms, for different women.”

The study, “Associations between testosterone and future PTSD symptoms among middle age and older UK residents,” was authored by Hanyang Shen, Ciera Stafford, Joeri Meijsen, Lijin Zhang, Jacob Reiter, Rebecca B. Lawn, Alicia K. Smith, Mytilee Vermuri, and Laramie E. Duncan.