The Development of Non-Suicidal Self-Injury in Adolescence: The Role of Interpersonal and Intrapersonal Risk Factors

The present dissertation aims to improve our knowledge on the longitudinal development of Non-Suicidal Self-Injury (NSSI) and the role of interpersonal and intrapersonal risk factors associated with it. Non-Suicidal Self-injury (NSSI), defined as the direct and deliberate self-inflicted damage of body tissue without suicidal intent, is a serious public health concern worldwide (Kiekens et al., 2018). Adolescents are the most at-risk group, given that the transition into adolescence may represent a critical vulnerability period for the onset of NSSI behaviors (Lloyd-Richardson, 2008). This phenomenon requires attention not only because of its heavy impact in terms of public health and the high incidence within the population, but also for the consequences that engagement in NSSI entails. The long-term effects of self-injurious behavior can be destructive, with consequences for emotional and cognitive development (Baetens et al., 2011). NSSI is used as a maladaptive means of coping with intense emotions. Both interpersonal (e.g., social interaction with peers and family; Brausch & Gutierrez, 2010) and intrapersonal factors (e.g., emotion regulation, self-efficacy, and self-esteem; Baetens et al., 2011) can serve to initiate and maintain NSSI (Nock, 2009; Nock & Prinstein, 2004; Zetterqvist et al., 2013). In the last ten years, the attention given to this issue has become increasingly important. Most of the existing literature has examined this behavior (e.g., prevalence, risks factors) at the cross-sectional level, while few studies have explored the longitudinal development of NSSI, and the role played by interpersonal and intrapersonal factors at the longitudinal level. For these reasons, the general aim of the present dissertation is to analyze the longitudinal development of NSSI and the association with interpersonal and intrapersonal risk factors. Three empirical studies are presented. They cover three main issues: 1) a meta-analysis on the longitudinal development of NSSI; 2) the reciprocal associations between peer problems and NSSI; 3) the mediational role of Covid-19 related stress in the association between pre-existing vulnerabilities and NSSI. In the first study (Chapter 1), we presented a meta-analysis on the development of NSSI from childhood to young adulthood, using a Bayesian approach. The aim was to examine both the occurrence and the frequency of NSSI over time, considering all studies published up until November 2020. Subsequently, we examined the role of possible moderators, such as gender, mean age during the first wave of data collection, and number of months covered by the assessment. The results show the important role of gender (i.e., females) and age in the explanation of the expected proportion and mean changes of NSSI over time. Specifically, what emerges from the findings is how being female represents an important risk factor for the occurrence of this behavior. As for the frequency of this behavior, a higher percentage of females are associated with higher severity of NSSI, but it tends to decrease over time. The results show that mid-adolescence (i.e., 14/15 years) appears to be the period of highest risk for the occurrence of NSSI over time. Instead, over time, findings suggest that the frequency of this behavior is higher in adolescence, at a mean age of 15-16 years of age, and it decreases in late adolescence (e.g., Plener et al., 2015). In the second study (Chapter 2), we investigated the reciprocal associations between peer problems (e.g., peer victimization, friendship stress, and loneliness) and NSSI throughout adolescence, distinguishing between- and within-person effects. Participants were 866 adolescents (54.5% females; Mage = 13.12 years, SD = 0.78), who took part in six waves of data collection. Random Intercept Cross-Lagged Panel Models (RI-CLPM) were used to estimate within-person cross-lagged effects between each peer problem and NSSI from Grade 7 to 12. After accounting for between-person associations between peer problems and NSSI, results indicated that higher-than-usual levels of NSSI predicted higher-than-usual levels of adolescents’ own friendship stress, loneliness, and peer victimization at the subsequent time point. Yet, sensitivity analyses revealed that most of these effects were strongly attenuated and explained by within-person fluctuations in depressive symptoms. No within-person cross-lagged effects from peer problems to NSSI were found. In the third study (Chapter 3), we examined the role of Covid-19 related stress in the association between pre-existing vulnerabilities and the engagement in NSSI during the pandemic. Specifically, the study aimed to examine if adolescents with pre-existing vulnerabilities, including a prior history of NSSI, higher levels of internalizing symptoms, and poorer regulatory emotional self-efficacy, were more likely to show increases in NSSI across the pandemic period through higher levels of Covid-19 related stress. The analysis was conducted on 1061 adolescents (52.4% females; Mage = 15.49 years, SD = 0.76), enrolled in the 9th and 10th grade in Tuscany, Italy, who took part in two waves of data collection. Results showed that adolescents with pre-existing vulnerabilities were at higher risk of engaging in NSSI through the role of Covid-19 related stress. Specifically, adolescents with a prior history of NSSI, higher levels of anxious and depressive symptoms, and poorer regulatory emotional self-efficacy showed a higher level of Covid-19 related stress, which in turn it was associated with an increased risk of occurrence of NSSI. In the final chapter (Chapter 4), the results of the previous three studies have been discussed highlighting their contribution to the literature on the longitudinal development of NSSI, strengths and limitations, and the implications for future studies.