New Insight Into Drivers of Self-Harm in Teens

Pauline Anderson

June 13, 2023


Prepandemic cortisol response to stress and amygdala emotion-evoked activation predicted persistent teen engagement in nonsuicidal self-injury (NSSI) among teensduring the COVID-19 pandemic.


  • The analysis included 64 mostly White and middle class or upper middle class female patients in Minneapolis, Minnesota (mean age, 16.2 years) who were part of a larger study of the neurobiology of NSSI.

  • Before the pandemic, researchers assessed the presence of NSSI and measured cortisol levels in saliva while the participant was experiencing stress, such as when giving a speech (less cortisol in response to stress is a sign of HPA axis hyporeactivity); adolescents were assessed for depression and underwent neuroimaging.

  • In the early stages of the pandemic, adolescents were assessed for recent engagement in NSSI.

  • Researchers classified adolescents into three NSSI groups: never (n = 17), desist (a history of NSSI but did not report it during the pandemic; n = 26), or persist (a history of NSSI and reported it during the pandemic; n = 21).


  • Lower prepandemic levels of under the curve ground (AUCg), an index of overall activation of cortisol levels (B = −0.250; standard error [SE] = 0.109; P = .022) and lower prepandemic amygdala activation (B = −0.789; SE = 0.352; P = .025) predicted desistance of NSSI, compared to persistence of NSSI, during the pandemic.

  • This remained significant after controlling for pandemic-related stressors that could exacerbate underlying risk factors

  • When depression was included as a covariate, decreased cortisol AUCg and amygdala activation remained significantly predictive of desistance. Decreased medial prefrontal cortex resting state functional connectivity and decreased depressive symptoms were also predictive of desistance of NSSI.


The results "may give insight into predictors of maladaptive patterns of coping with negative emotions" for those with a history of NSSI, the authors note.


The study was conducted by Katherine A. Carosella, Department of Psychology, University of Minnesota, and colleagues. It was published online February 12 in Psychoneuroendocrinology.


The study was relatively small, and the investigators could not make causal inferences or rule out the possibility that different stages of development affected the data. Measures employed during COVID were not identical to those used in the prepandemic assessment.


The study received support from the National Institute of Mental Health and the University of Minnesota. The authors have disclosed no relevant financial relationships.

For more Medscape Psychiatry news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.