Aim: Non-suicidal self-injury (NSSI) constitutes a risk factor for subsequent behavioral and emotional issues such as depression, aggression, and heightened emotional reactivity. Traumatic experiences among individuals with NSSI also exhibit associations with these psychological problems.
Specific features of NSSI and various types of traumatic experiences may be linked to specific aspects of emotional reactivity, aggression, or the severity of depression, but remain unclear, especially among university students, who typically represent a highly functional cohort. The aim of this study is to thoroughly examine these complex relationships and identify risk factors.
Methods: Using an online survey, 150 university students aged 18 to 25 years, who had self-injured at least once, were queried about traumatic experiences, depression, emotional reactivity, and aggression. Regression analyses were conducted to determine risk factors associated with increased outcome measures.
Results: It was found that childhood emotional abuse contributes to emotional reactivity, aggression, and depression. Risk factors for persistent NSSI beyond adolescence included lower levels of childhood physical neglect (Odds = 0.82), increased automatic positive reinforcement, and reduced social positive reinforcement.
Conclusions: The results indicate that childhood emotional abuse significantly contributes to emotional and behavioral problems and should therefore be considered in NSSI therapy. Furthermore, it was observed that NSSI persists into adulthood when used as a strategy for emotion regulation rather than for communicating needs in an environment of physical neglect.

Aim: Adolescents with non-suicidal self-injury (NSSI) and suicidal behavior do often not seek professional help or do so with delay. The study aimed to examine the association between help-seeking duration after first NSSI, suicidal thoughts or suicide attempt and future intervention outcomes in an outpatient sample of adolescents.
Methods: The sample consisted of 285 adolescents recruited from AtR!Sk; a specialized outpatient clinic for adolescents with risk-taking and self-harming behavior. Frequency of suicide attempts in the previous year and frequency of NSSI in the previous six months were assessed at baseline and 1-year follow-up alongside with the duration from symptom onset to first contact with professional care (i.e. help-seeking duration). Generalized linear mixed models were used to assess associations between help-seeking duration for first NSSI, suicidal thoughts or suicide attempt and NSSI or suicide attempts at 1-year follow-up, controlling for age and gender.
Results: The results show a significant decrease in the frequency of suicide attempts between baseline and 1-year follow-up for seeking help due to suicide attempt (p < 0.001) or NSSI (p < 0.05), and in the frequency of NSSI for seeking help due to NSSI (p < 0.001). However, help-seeking duration for NSSI, suicidal thoughts or suicide attempt was not significantly associated with NSSI or suicide attempts at 1-year follow-up.
Conclusions: Duration between onset of suicidal thoughts and self-harm and first contact with mental health services was not associated with future self-harming behavior. Clinically, this may indicate that symptoms can be effectively reduced by intervention, even if professional help is sought late.

Aim: Suicidal behaviour (SB) and non-suicidal self-injury (NSSI) are highly related phenomena, which are both common within adolescent clinical samples. In the present study, we investigated whether a set of neurobiological markers contributed above depressive symptoms to the distinction of adolescents with NSSI and SB from those with NSSI only using machine-based learning approaches.
Methods: Female adolescents with engagement in NSSI (n = 161) were recruited from our outpatient clinic for risk-taking and self-harming behaviour (AtR!Sk). Different machine-based learning models (linear and logistic regression, elastic net regression, random forests, gradient boosted trees) with repeated cross-validation were applied. Depressive symptoms were included as a covariate in every analysis to account for their potential confounding effects. We tested whether a) a broad set of neurobiological markers versus b) only selected markers (CRP, interleukin-6, salivary cortisol, DHEA-S, TSH, dopamine, norepinephrine, and ACTH), which are known to be associated with suicidal behavior, could distinguish between the two groups (NSSI + SB vs. NSSI only).
Results: After controlling for depressive symptoms, models showed low to moderate predictive performance (AUC between 0.62 and 0.71). Predictors with the highest predictive performance were high DHEA (OR = 4.37; p = 0.03) and high TSH (OR = 1.95; p = 0.03). Nonlinear models slightly outperformed linear ones and feature selection modestly increased predictive performance.
Conclusion: The study underscores the potential of specific neurobiological markers for the risk-stratification of self-harm. Further research is needed to replicate these findings longitudinally.

Background: Homeostasis models posit that nonsuicidal self-injury (NSSI) serves, in part, to upregulate the endogenous opioid system in order to compensate for an opioid deficiency. A few studies have demonstrated lower basal levels of beta-endorphin (BE), an endogenous opioid, in individuals with NSSI. However, longitudinal studies are missing. Hence, the present study aimed to investigate the longitudinal associations between NSSI, comorbid psychopathology (i.e., borderline personality disorder and depressive symptoms), pain sensitivity and basal BE levels in adolescents with NSSI.
Methods: N = 53 adolescents with NSSI disorder undergoing specialized treatment participated in baseline and one-year follow-up assessments. BE was measured in plasma; pain sensitivity was assessed with a heat pain stimulation paradigm. Associations between BE and change in NSSI, borderline personality disorder and depressive symptoms as well as pain sensitivity were examined using negative binomial and linear regression analyses.
Results: We found that an increase in basal BE was associated with a decrease in depressive symptoms (b = -0.06, p = .005). No associations between BE and NSSI, borderline personality disorder symptoms or pain sensitivity were observed.
Conclusion: Our findings may confirm a role of plasma BE in the etiology of depressive symptoms but challenge current endogenous opioid homeostasis models of NSSI.

Background: Non-suicidal self-injury (NSSI) is a highly prevalent phenomenon during adolescence. Nonetheless, research on predictors of the clinical course of NSSI over time is still scarce. The present study aimed at investigating the impact of adverse childhood experiences (ACE) and hypothalamus-pituitary-adrenal (HPA) axis functioning on the longitudinal course of NSSI.
Methods: In a sample of n = 51 help-seeking adolescents engaging in NSSI, diurnal cortisol secretion (CAR, cortisol awakening response; DSL, diurnal slope), hair cortisol concentrations and ACE were assessed at baseline. Clinical outcome was defined by change in the frequency of NSSI in the past 6 months measured 12 and 24 months after the baseline assessments. Mixed-effects linear regression models were used to test for effects of ACE and HPA axis functioning on the course of NSSI.
Results: ACE and HPA axis functioning did not show main but interaction effects in the prediction of NSSI frequency over time: Adolescents with a low severity of ACE and either an increased CAR or a flattened DSL showed a steep decline of NSSI frequency in the first year followed by a subsequent increase of NSSI frequency in the second year.
Conclusions: Our findings could be interpreted in the sense of high diurnal cortisol concentrations in the absence of ACE being favorable for clinical improvement on the short-term but bearing a risk of allostatic load and subsequent increase of NSSI frequency. In contrast, adolescents with severe ACE may benefit from elevated cortisol concentrations leading to slower but lasting decreases of NSSI frequency.