Non-suicidal self-injury (NSSI), which involves deliberate, self-inflicted damage of body tissue without suicidal intent, is a serious and prevalent problem (1). Research has shown that NSSI most commonly occurs among 13–15-year olds (2–4), while other studies have highlighted the severe psychological, medical, social, legal and ethical consequences associated with NSSI (5–8). Several international studies have examined NSSI in adolescent populations in different samples. Lifetime prevalence rates of NSSI of 17–18% in adolescents were reported in two systematic reviews (9, 10), with considerable international variability across community samples [from 2.9 to 42% in (9) and from 9.4 to 49.2 in (10)]. In clinical samples the numbers tend to be higher with the prevalence rates varying between 50 and 61% (11, 12). Another review (13) suggested that the aggregate lifetime prevalence of NSSI tended to be higher in non-Western (32.6%) than Western countries (19.4%). Despite this research, there are still many settings where there has been little focus on NSSI. This is the case for the Russian Federation where few studies have been conducted so far and national statistics on the prevalence of NSSI are limited (14, 15). Nevertheless, there is some indication that NSSI might be prevalent in Russia as a recent study found that 13% of adolescents reported at least one previous episode of self-injury (16).

Associations between self-injury and mental health problems are common and the vast majority of adolescents who engage in NSSI also have some degree of comorbidity, including post-traumatic stress disorder (PTSD) (17, 18), with generally higher levels of traumatic exposure (19, 20), depressive disorders (21), obsessive-compulsive disorder (18), anxiety (22), symptoms of borderline personality disorder (23, 24), specific phobia (21), autism (25), intellectual disability (26), schizophrenia (27), and eating disorders (28). A strong association has also been noted between NSSI and externalizing problems where conduct disorder (CD) and attention deficit hyperactivity disorder (ADHD) are most common (29, 30). Adolescents engaging in NSSI may also have substantial difficulties in social interaction, such as problems with peers and poor academic achievement at school (31, 32). These conditions may be related to worse NSSI in terms of frequency/severity and all represent detrimental outcomes that can complicate efforts to treat NSSI or exacerbate the effects of NSSI.

An additional problem is posed by the possible association between self-injurious behavior and suicide. While some research has suggested that a large proportion of self-injury cases tend to be non-suicidal in nature and that self-injurious behavior is intended to effect a change in the environment, rather than result in an individual's death (33), other studies have consistently reported that NSSI is a robust predictor of suicidality (34–37) with suicidal ideation and attempts being more common among individuals with NSSI as compared to individuals without NSSI (2, 38). It has also been found that NSSI predicts suicidal behavior over and above mental health problems, such as depression (35), hopelessness and borderline personality traits (34), PTSD and a history of child abuse (39).

Among the most well-known and widely researched factors linked to self-harming behavior are emotion regulation (40) and impulsivity (41), where NSSI is thought to occur as a result of emotional and cognitive dysregulation (42), as well as impaired impulse control (43). From this perspective, personality has been thought to influence the use of emotion regulation strategies (44) and the ability to delay actions in order to plan and make choices by identifying a solution (43). Indeed, personality research among the general population using Cloninger's biopsychosocial personality model (45) has demonstrated a relationship between personality traits and emotion regulation as well as impulsive behavior (46–48). Regarding NSSI, it has been reported that specific personality traits are associated with NSSI in adolescents (49, 50). Tschan et al. (51) found that adolescents with a history of NSSI had a personality profile that included higher scores on the personality traits novelty seeking (NS) and harm avoidance (HA) and lower scores on the character traits self-directedness (SD), persistence (P) and cooperativeness (C) compared to adolescents with mental disorders but without NSSI. In addition, another study (52) found that while adolescents with occasional and repetitive NSSI differed significantly from those without NSSI only on low SD, individuals with repetitive NSSI differed on several personality traits, such as higher NS, HA and self-transcendence (ST) and lower SD. These findings suggest that researchers and clinicians might benefit from identifying personality traits that are linked to NSSI in order to gain a better understanding of the mechanisms that are involved in NSSI and to develop treatment approaches that consider personality (53, 54). Self-injurious behavior is a serious problem in offender populations (55). Nonetheless, while reviews on NSSI among incarcerated adults are available (56, 57), the comparable literature for adolescents is still somewhat limited. Several studies have suggested that youth from the juvenile justice system may represent an especially high risk population for self-injurious and suicidal behavior (17, 58, 59). Two studies (60, 61), included in a review by Casiano et al. (62), reported a lifetime prevalence of “deliberate self-harm (acts of both suicidal and non-suicidal self-injury)” in incarcerated juveniles that ranged from 6 to 40%. A high prevalence of self-harm among young offenders is not surprising given that many of the risk factors for NSSI, such as poverty, violence and poor mental health, exist at higher rates among this population when compared to the general youth population (63). Moore et al. (64) suggested that cumulative exposure to multiple risk factors, which may take place among youth in custody, might further increase the risk of NSSI.

A number of studies have explored factors associated with suicidal ideation and attempts among delinquent youths (65, 66), but as yet, research on NSSI has been limited, and no previous study has investigated the prevalence of NSSI among incarcerated youth in Russia. Expanded knowledge regarding different aspects of NSSI among incarcerated youth, such as its prevalence, comorbidity and association with personality, can potentially facilitate the development of more effective interventions for self-harming behavior.

The current study aimed to: (1) estimate the prevalence of NSSI in incarcerated adolescents in Russia; (2) evaluate what types of psychiatric diagnoses and self-reported mental health problems are associated with a history of NSSI; and (3) examine whether specific personality traits are related to NSSI in this population. We expected that adolescents from the NSSI group would have higher rates of comorbid psychopathology and suicidal ideation. We also expected that NSSI would be associated with specific personality traits.

These scales were translated into Russian following established standard procedures (77). The Russian translation of the K-SADS-PL was done at the Department of Psychology, Moscow State University.

The data were analyzed using the Statistical Package for the Social Sciences (SPSS-25.0). Chi-square tests were used to compare the prevalence of psychiatric disorders between the NSSI and No-NSSI groups. Independent samples t-tests were used in order to compare the levels of problem scores between the groups. The level of statistical significance was p < 0.05 (two-tailed).

This study aimed to determine the prevalence of NSSI among incarcerated adolescents, as well as to explore the potential associations between NSSI and psychopathology and personality. Just over 18% of our subjects reported lifetime engagement in NSSI. The prevalence of NSSI in the current study was lower than the mean prevalence reported in community adolescent samples in international studies where the aggregated lifetime prevalence of NSSI was reported to be as high as 22.1% (13). However, the lifetime prevalence rates in previous studies have varied considerably between countries and in some were as low as 13% (78). In addition, the lifetime prevalence of NSSI in our population was higher than that (13%) reported in the only previous Russian study on the prevalence of NSSI in late adolescents (16). It is uncertain what underlies these differences across studies although Muehlenkamp et al. (10) proposed that dissimilarities in assessment methodologies (including settings and instruments) and in the classification of self-injury might explain the variation in rates between studies.

The number of adolescents with NSSI who reported recurrent thoughts of death and suicidal ideation was over two times higher than those in the No-NSSI group, while almost one-third of the NSSI adolescents also described their previous suicide attempts as serious (i.e., described a definite suicidal intent at the time of the attempt), as compared to only one-tenth of adolescents in the No-NSSI group. These findings demonstrate that while some researchers have suggested that these two types of behavior differ from each other in some fundamental ways (79), they nevertheless often co-occur. This indicates a possible relationship between the engagement of delinquent youths in self-harm and a risk for suicidal behavior. Indeed, these findings are in line with previous reports that suicidal thoughts are more common among self-injurious adolescents, they are more often involved in making suicide attempts (11, 24) and that NSSI may be a predictor of suicidal behavior (80).

Previous research has shown that an abundance of different internalizing and externalizing symptoms are common among individuals with NSSI (36, 81, 82). Findings from our study suggest that incarcerated adolescents with NSSI represent a high-risk group, as they tend to have higher rates of anxiety and PTSD and they also scored higher on almost all types of problem behavior, as assessed by the YSR. In addition, the present study also suggests that community violence exposure is more prevalent among adolescents with NSSI. This further emphasizes the importance of screening for NSSI in juvenile offender populations, especially among those exposed to violence, and that screening protocols should be included in the standardized assessment procedures available in the juvenile justice system.

Although non-significant, the difference in the prevalence of MDD was near significance level, while self-reported depressive symptoms (both the BDI and Anxious/Depressed scale of the YSR) were significantly higher in the NSSI group. This finding is underpinned by earlier studies where depressive symptoms were highly comorbid with NSSI (13, 83). In incarcerated adolescents, MDD symptoms were found among those who reported NSSI twice as often compared to those without a history of self-injury (84). It is not clear why we obtained different results (significant and non-significant) when we assessed the association between NSSI and depression using different measures. However, previous studies that have assessed depressive symptoms by self-reports and clinician-administered interviews have similarly reported varying results. Martin et al. (85) after comparing expert and self-assessments, concluded that self-reports were not reliable in identifying participants with a lifetime diagnosis of MDD. In contrast, although Stuart et al. (86) referred to the clinical interview as the “gold standard” for identifying depression, they also acknowledged that simple self-report methods did have a role when it came to identifying symptoms of depression. Differences in the results between self-reports and clinical interviews might also reflect the sensitive nature of the topic and poor memory due to current mood (87, 88).

The study also examined personality traits in adolescents with NSSI and found, as expected, that they differed from those in No-NSSI adolescents. More specifically, they had lower scores on self-directedness, but higher scores on self-transcendence. Low self-directedness, reflecting unreliability, impulsivity and indecisiveness (76) was recently reported to predict NSSI among adolescents (40), with self-directedness gradually decreasing as the frequency of NSSI increased. As suggested by Glenn and Klonsky (2), most individuals with NSSI have poor impulse control and the engagement of impulsive individuals in NSSI may be related to seeking immediate relief from tension without thinking about the possible consequences of such behavior. There is also some evidence that difficulties in emotion regulation may be associated with low self-directedness (89) and play a role in a number of mental health problems, including NSSI (42). According to Cloninger et al., a person with low self-directedness is characterized by a decreased ability to influence difficult situations and often has problems resolving them positively (45). As a result, when experiencing negative emotions or being in difficult situations, a combination of low self-directedness and emotion regulation difficulties (as suggested by higher anxiety and depression scores), might lead to self-injurious behavior. In connection with this, as Bernheim et al. (90) have previously reported that psychotherapy might play a role in strengthening self-directedness, this might also be important for NSSI treatment, considering that an association has been documented between NSSI and low treatment adherence and low self-directedness (91, 92).

Findings on the association between NSSI and self-transcendence are also in line with those from previous studies where high self-transcendence has been linked to NSSI in adolescents (42, 52, 93). As suggested by Cloninger (76), high self-transcendence is generally associated with extravagant thinking, a lack of logic and unconventional behavior (94, 95) and NSSI might be a result of such traits, where self-transcendence may reflect a sense of detachment between the individual's self and his or her surroundings, potentially leading to self-injurious behavior (93). Previously, it was shown that specific personality traits are not only associated with a predisposition to different mental disorders, but are also able to impact treatment results (96). As low self-directedness and high self-transcendence are common in different psychopathologies, such as personality disorder, depression, bipolar disorder and NSSI (52, 97, 98), then treatment approaches that take these personality traits into account may be relevant.

In spite of important and positive changes in the Russian juvenile justice system in recent years and a dramatic decrease in the number of incarcerated youths (from 18,600 in 2002 to only 1,592 in 2017) (99), it is still unclear whether incarcerated youth receive help which is sufficient to meet their mental health needs. With regard to interventions, evidence-based and developmentally sensitive strategies need to be elaborated to intervene among youth in the juvenile justice system. While evidence from several international studies on the effectiveness of therapeutic interventions is available (100, 101), it is unclear whether these results can be generalized to the Russian context. There is a need therefore for more research in order to address issues related to NSSI among incarcerated youth and gain a better understanding of their current needs and gaps in the provision of services.

Summing up, our findings point to the importance of targeting adolescents in correctional facilities who have engaged in NSSI, as they may represent a high risk group for suicidal behavior. Intervening early to support young people who self-injure, to help prevent an exacerbation of NSSI should be a top priority. In relation to this, our findings may be useful for workers within the juvenile justice system in helping them to identify adolescents with NSSI earlier in order to more quickly initiate intervention measures. Adequate screening and recognition of mental health problems and needs is not only clinically important but may also help to identify factors associated with persistent NSSI and result in more individualized and suitable intervention and prevention efforts, as well as better treatment outcomes.

This study has several limitations which should be mentioned. One of the main limitations and methodological-ethical concerns relates to the data collection setting, where the participants were placed involuntarily. In such an environment it is not clear whether all of the subjects really did think that they had the freedom to refuse to participate and that the information they provided would be guaranteed to be kept confidential. This might have affected their responses. Similarly, in this confined setting the use of self-reports may have resulted in socially desirable responding. Another limitation is related to the lack of information on the diagnostic history of NSSI and that we did not know when the behavior first appeared, before or during the process of incarceration. In addition, we could not determine the directionality of the observed associations given the cross-sectional design of the study. Future longitudinal studies may be helpful both in further elucidating these relationships and also in helping to determine effective prevention and treatment approaches.

The data used in this study cannot be shared publicly due to the initial decision of the local ethical committee, as well as the restrictions included in the informed consent statement (where it was stated that the data would only be used by the research group and would not be transferred elsewhere). Requests to access the dataset should be directed to Roman Koposov.

The studies involving human participants were reviewed and approved by Northern State Medical University in Arkhangelsk (Russia). Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

VR, RK, and AS were involved in the conceptualization and design of the study. RK and VR conducted the data collection and drafted the manuscript. VR conducted the data analysis. All authors reviewed and edited the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.