Shen Hanyang, Stafford Ciera, Meijsen Joeri, Zhang Lijin, Reiter Jacob, Lawn Rebecca B, Smith Alicia K, Vermuri Mytilee, Duncan Laramie E
Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford, CA, USA.
Department of Epidemiology and Population Health, Stanford University, Stanford, CA, USA.
Transl Psychiatry. 2025 Aug 6;15(1):268. doi: 10.1038/s41398-025-03482-5.
Testosterone has been theorized to influence the development of post-traumatic stress disorder (PTSD). However, the relationship between testosterone level and PTSD is still not well understood. We evaluated the potential association between testosterone and subsequent development of PTSD symptoms using a large sample size, in a civilian context, inclusive of both males and females. Out of around 500,000 total UK Biobank participants, our sample had 130,471 participants who: had testosterone measures, completed the mental health questionnaire, and passed outlier exclusion. After adjusting for relevant covariates, we used linear regression to assess the relationship between testosterone level and future development of symptoms, in males and females separately (N = 61,758, N = 67,053). In both males and females, small but significant nonlinear (and oftentimes U-shaped) relationships were observed between testosterone levels and PTSD symptoms. When grouping participants into deciles of testosterone for both sexes, the strongest associations between testosterone levels and PTSD symptoms were observed in the central deciles. For example, for total testosterone, compared to decile 1: individuals in decile 7 had the lowest PTSD symptom scores in both males (beta = -0.16, p = 1.58 × 10) and females (beta = -0.23, p = 3.04 × 10). We also found that body mass index (BMI) moderated the relationship between testosterone and PTSD symptoms, such that the relationship was considerably stronger among individuals with higher BMI. Results were similar for depression and anxiety measures. Analyses using calculated free testosterone (cFT) and the free androgen index (FAI) were generally consistent with total testosterone (TT) results. These findings suggest that mid-range testosterone levels are associated with the lowest risk of PTSD symptoms in both sexes, and future work should seek to examine if this relationship is causal.
有理论认为睾酮会影响创伤后应激障碍（PTSD）的发展。然而，睾酮水平与PTSD之间的关系仍未得到充分理解。我们在平民背景下，纳入男性和女性，使用大样本量评估了睾酮与PTSD症状后续发展之间的潜在关联。在英国生物银行约50万名参与者中，我们的样本有130471名参与者：进行了睾酮测量、完成了心理健康问卷并通过了异常值排除。在调整相关协变量后，我们使用线性回归分别评估男性和女性（N = 61758，N = 67053）中睾酮水平与症状未来发展之间的关系。在男性和女性中，均观察到睾酮水平与PTSD症状之间存在小但显著的非线性（且通常为U形）关系。当将参与者按睾酮水平分为十分位数时，在中间十分位数中观察到睾酮水平与PTSD症状之间的最强关联。例如，对于总睾酮，与第1十分位数相比：第7十分位数的个体在男性（β = -0.16，p = 1.58×10）和女性（β = -0.23，p = 3.04×10）中PTSD症状得分最低。我们还发现体重指数（BMI）调节了睾酮与PTSD症状之间的关系，使得在BMI较高的个体中这种关系更强。抑郁和焦虑测量的结果相似。使用计算的游离睾酮（cFT）和游离雄激素指数（FAI）进行的分析总体上与总睾酮（TT）结果一致。这些发现表明，中等范围的睾酮水平与两性中PTSD症状的最低风险相关，未来的研究应设法检验这种关系是否具有因果性。
