Preview

PDF, English - main document Download (2MB) | Terms of use

Abstract

Non-suicidal self-injury (NSSI) is intended to be harmful to oneself through tissue damage without the intention to die. Individuals suffering from NSSI are affected by many short and long-term negative consequences, for instance, severe wounds, scars, and a high risk for aggravated psychopathology. Theories on NSSI postulated that NSSI is mainly used to regulate aversive inner states such as negative affect (NA), but there seem to be also other relevant triggers for NSSI, for instance, interpersonal problems. These assumptions are supported by self-report and cross-sectional studies, whereas findings in daily life using ambulatory assessment (AA) were more ambiguous. Looking at biological mechanisms of NSSI, evidence is still sparse but current findings suggest that individuals engaging in NSSI differ from those without NSSI in various ways. Mainly, individuals engaging in NSSI show altered activation patterns with regard to the perception and regulation of affect and pain in the brain and in peripheral systems (e.g., pain system) compared to those without NSSI. As NSSI is defined to be a harmful behavior, pain processing seems to be a central biological component involved in the development and maintenance of NSSI. The endogenous opioid system (EOS) is especially involved in the perception and regulation of emotional and physical pain. In the context of NSSI, β-endorphin is one important marker of the EOS, as it is released during tissue damage in central and peripheral regions of the body alike. Therefore, β-endorphin seems to be especially suited for the assessment of NSSI in daily life. Within this thesis, I presented three publications, all based on an AA study. The purpose of the study was to test the affect regulation and interpersonal function of NSSI as well as changes in β-endorphin surrounding NSSI in daily life. I assessed 51 cis-gender women, aged between 18 - 45 years, with DSM-V diagnosis of NSSI disorder and frequent NSSI engagement(min. one NSSI episode/ week > 3 month). Participants attended in the study over a 15-day study period, reporting on five semi-randomized prompts per day as well as on event-based prompts in case of NSSI. Furthermore, I sampled during high-urge for NSSI. To assess changes after NSSI or high-urge moments in the dependent variables more closely, I included a high-frequency sampling in the first thirty minutes (three follow-up prompts, each ten minutes apart) after NSSI or high-urge moments. Participants completed 4,619 prompts, resulting in a high compliance rate of 92.04% and engaged in 155 NSSI acts. The first publication comprised the assessment of salivary β-endorphin in daily life. First, I assessed β-endorphin on eight time-points on a non-NSSI day (first day of study period), to represent the trajectory of β-endorphin across the day. Afterwards, participants completed two additional weeks of the AA study. Saliva samples were collected during high-urges for NSSI, before and after NSSI events, and on three follow-ups in the thirty minutes following NSSI or high-urge moments. In line with my hypothesis, results show that salivary β-endorphin significantly increases from pre to post NSSI, which supports theoretically assumptions on the involvement of the EOS in NSSI. Furthermore, I found a significant association between the severity of the injury and levels of β-endorphin, suggesting that more severe wounds were associated with a higher release of β-endorphin. Contrary to my hypothesis, I did not observe a significant difference in β-endorphin levels in moments of high-urge vs. NSSI moments, which is not in line with theoretical assumptions, which postulate that moments of high-urge should be accompanied by especially low levels of β-endorphin. Interestingly, I also did not find a significant association between pain-ratings and levels of β-endorphin. I speculate that the non-significant finding could be due to cognitive processes involved in pain management. The second publication focused on the assessment of the affect regulation function of NSSI as postulated by almost all theories on NSSI. Those theories posit that NA and tension are increased prior to NSSI and decrease after the engagement in NSSI. I captured NA and tension as two facets of the affect regulation function at random prompts, during high-urge, NSSI moments and follow-up prompts. To tease apart whether effects of NSSI on NA and tension are specific for NSSI engagement or rather due to time effects, I compared moments of NSSI to moments of high-urge without subsequent NSSI (using high-urge moments as a consequent within-group control condition). For NA, I found that the decline in the NSSI condition was not steeper than in the high-urge condition, indicating that NSSI engagement did not outperform the resistance of an urge for NSSI. Results on tension showed that NSSI was associated with a significant linear decrease in tension, whereas resisting an urge was not. Nevertheless, trajectories in the high-urge condition for both, NA and tension, were better described by an inverted U-shaped pattern, leading to a significant decrease in NA and tension. This indicates that resisting an urge is also followed by significant reduction in NA and tension in the same time-window as engaging in NSSI. The described findings show only limited evidence for the affect regulation function of NSSI in my sample, as I only found a significant effect for tension regulation. The findings on NA and tension were descriptively supported by my exploratory analyses. Here, I used k-means clustering to visualize NA and tension trajectories surrounding NSSI and relate these to participants’ self-rated effectiveness of the NSSI events. Taken together, my results indicate that resisting an urge to self-harm may also be effective in managing NA and tension, highlighting the importance of well-established therapeutic interventions such as urge-surfing. The third publication of this thesis concentrated on the evaluation of the interpersonal function of NSSI, which postulated that negative interpersonal events (IPEs) should increase prior to and decrease following NSSI, whereas the number of positive IPEs should increase after NSSI. To test these assumptions in daily life, I presented a set of five negative and positive IPEs on each random prompt. Additionally, I asked participants how distressing the IPE was for them and if the behavior of the other person was a reaction to their last self-harm episode. In case of NSSI, I asked whether participants experienced a positive or negative IPE before they engaged in NSSI and if yes, how distressing the event was for them.I found a positive concurrent association between the number of negative IPEs and the engagement in NSSI. Furthermore, highly distressing negative IPEs were positively associated with concurrent NSSI events and urges and predicted later events. I observed no reduction in negative or increase in positive IPEs following NSSI, which suggests that there is no evidence for positive or negative interpersonal reinforcement through NSSI in this sample. Additionally, participants endorsed interpersonal motives for NSSI only rarely in a trait-level interview, but indicated that others, especially family members, often trigger NSSI. In line with this, participants infrequently endorsed the motive ‘get help/attention’ in daily life. These results suggest that negative IPEs (especially highly distressing ones) trigger NSSI, but participants rarely used NSSI for interpersonal motives. Taken together, I found evidence that IPEs could be seen as triggers for NSSI rather than a mechanism, which is involved in maintaining NSSI through positive or negative reinforcement. For the clinical practice, one implication could be that clinicians and they patients should focus on identifying interpersonal triggers of NSSI and how the amount of distress from IPEs can be regulated to prevent NSSI engagement, rather than focusing on the role of other people’s behavior in response to an NSSI event.