Caspian Journal of

1. Introduction
The frequency of self-harming behaviors in adolescence is increasing. One study demonstrated that the level of non-suicidal self-harming behaviors raised from 4% in 2002 to nearly 16% in 2017 [1]. Self-harming rate in Iranian adolescents was estimated between 13% to 22% [2]. It may result in several psychological and social outcomes. In most cases, it is difficult to recognize the main reason of this behavior [3]. Self-harming may contrive committing suicide or self-murder [4]. Therefore, studying the risk factors of self-harming is critically important. Although self-harming behavior is in contrast with life motive, but various motives are attributed to it. Yet there is no consistent and comprehensive information about the process of forming, continuity, and repetition of this behavior [5]. The contributing factors in self-harming could be generally considered as internal and external. 
Nock et al. categorize self-harming motives in four-function model (FFM) as positive attention attracting from others, creating positive emotional moods, release of negative emotions, and avoiding irritating people [6]. Although the impulsive behavior contribute to predicting harming behavior, Lockwood et al. [7] in a review article concluded that several cognitive and emotional variables can play role in explaining and retaining this behavior. Among cognitive variables, emotion dysregulation is the most accepted etiologies of self-harming behaviors in teenagers [8]. 
On the other hand, the role of peers in adolescence behaviors is considerable. The peers’ factors in adolescence especially modeling of them, copying them, the tendency to be accepted by them and the experience of being victim or bullying among them correspond with self-harming and committing suicide. Previous studies indicated that self-harming behaviors correspond with the motivations of group acceptance, agitation tool, anti-suicide, self-proving, self-distinguishing from others, anti-resolution and revenge ties with the motives of non-suicidal self-harming [9]. Emotion dysregulation contribute to a wide range of strategies which people use in different stressful situations. Individuals who experience emotion dysregulation have difficulty accepting their emotional response which can lead to behavioral dysregulation including excessive substance use, impulsivity or self-harming [10]. It should be noted that self-harming includes two groups of direct behaviors such as non-suicidal self-harming and indirect behavior like drug abuse and risk-taking behaviors. In direct behaviors, a person hurt his peripheral tissues consciously and deliberately but does not intend to suicide and the behavior is not socially accepted by culture and in indirect self-harming a person behaves voluntarily which in long term would harm, but there is no intention of self-harming at the time of doing behavior [11].
The role of emotion regulation defects in self-harming has been reported [12]. There is dispersed information about the factors and processes which may build links between self-harming and defects of emotion regulation. The present research could be novel and innovative with the aim of studying emotion regulation and modeling the self-harming of peers in predicting self-harming behaviors in teenage girls.
2. Materials and Method
Study type and population
The current descriptive-cross sectional study was conducted on delinquent teenage girls in Correction and Rehabilitation Center of Mashhad from January to April 2021. The participants were selected voluntarily and the eligibility criteria were: 16-18 age range; the ability to read and write and the consent for participating in research. Moreover, we obtained the informed consent of their legal guardians. The sample size was estimated 150 out of the total of 242 teenage girls in Correction and Rehabilitation Center of Mashhad, using Cochran formula. The estimation was based on maximum prevalence of 0.5, precision and significance level of 0.05 and considering finite population correction for calculation.
Data collection and measures
The required data was collected using four questionnaires that were given to the psychologist and the consultant of the center in order to distribute between participants. They also received the instructions on how to fill them. 
Self-harming behaviors was assessed using Self-harm Inventory of Sanson et al. (1994) containing 22 items scored as yes (1) or no (0). The range of score was 0-22. High score indicates the high rates of self-harming behavior. Several studies have supported the validity and reliability of the tool. In the first study of Soanson et al. [13] the reliability of questionnaires according to Alfa Cronbach was 0.83. Tahbaz et al. [14] translated the scale into Persian. They reported adequate content validity and reliability coefficient was 0.76. 
Emotion Dysregulation Scale questionnaire by Garenfsky et al. consists of 36 items rated on a 5-point Likert scale (ranging from 0=almost never to 5=almost always). The range of score was 0-180. The questions measure 6 dimensions; Non-acceptance of emotional responses, difficulty engaging in goal-directed behavior, impulse control difficulties, lack of emotional awareness, limited access to emotion regulation strategies, and lack of emotional clarity. The six-factor structure of scale confirmed through exploratory factor analysis, and the validity coefficient was 0.93 [15]. The factor structure of the questionnaires has been confirmed in Iran. The internal consistency was 0.92 and test-re-test correlation during two weeks for sub-scales was estimated between 0.86 to 0.88 [16].
The questionnaire of modeling peer’s self-harming by Clitor and Nixon is a part of Ottawa self-harming inventory. This sub-scale contains 6 items with three types of questions; yes/no (0/1) questions and multiple-choice questions (little (1), sometimes (2), a lot (3) and one question scored 1-5). Higher scores indicate more modeling of peers and scores ranged between 4-14. In Nixon et al. [17] reliability coefficient was reported 0.82. In Iran, Khanipour [18] internal consistency of the questionnaire according to Cronbach’ Alfa was 0.73.
Suicide Probability Scale by Cull and Gill comprise of 36 questions and 4 subscales of disappointment, suicide thoughts, hostility-aggression and negative self-evaluation. The questions are scored according to a four-rated scale. The scores are categorized to high risk (75-100), average risk (50-74), and low risk (25-49) of suicide. The scores ranged from 25 to 100. In Cull and Gill study the Cronbach’s’ Alfa of the total score for scale was reposted 0.93 and for subscales ranged from 0.62 to 0.89 [19]. The scale was applied to adolescents and the four-factor structure was confirmed, also the validity was reported 0.91 for all questions using Cronbach’s Alfa [20]. The convergent validity of the scale according to the correlation between test score and symptoms was verified in a study on teenagers in Iran [21].
Statistical analysis
Research data were described through the Mean±SD. According to the research aims, a model was developed and the model parameters were estimated by the method of partial least squares. Suicide probability was exogenous variable, self-harming behavior was endogenous, and emotion dysregulation and peer modeling were the mediating variable of the model, which were defined as a latent variable. To measure the fit of the research conceptual model, structural equation modeling (SEM) was used, which is a kind of latent variable analysis, which allows sets of multiple regression equations to be tested simultaneously to test the relationships between variables [22]. After specifying the model, an estimation of overall goodness of fit of the model with the observed data was performed. For this purpose, Chi-square/df index of fit, the comparative fit index (CFI), and Standardized root mean square residual (SRMR) were used [23]. The R2 evaluates the portion of the variance of the endogenous variables, which is explained by the structural model. It indicates the quality of the adjusted model. For the area of the social and behavioral sciences, Cohen suggests that R2=2% as classified as having a small effect, R2=13% as a medium effect, and R2=26% as having a large effect. The Cohen’s Indicator (f2) is obtained by the inclusion and exclusion of model constructs (one by one). The values of 0.02, 0.15, and 0.35 considered as small, medium and large effect [24]. The significance level was considered 0.05. Correlation matrix was reported using Pearson correlation coefficient. Data analysis was conducted in SPSS software, version 22 and SEM was performed using SmartPLS-3 software.
The reliability of the variables was evaluated using composite reliability (CR), which is a measure of the internal fit of the model, in the SmartPLS-3 software. Convergent validity of latent variables was tested with the average variance extracted (AVE). Convergent validity was defined as AVE greater than or equal to 0.50. The AVE values should be above the 0.50. The traditional indicator Cronbach’s Alpha (CA), is based on the variables inter-correlations. CA values above 0.60 and 0.70 are considered fitting in exploratory studies and composite reliability (CR) values of 0.70 and 0.90 are considered satisfactory [22].
3. Results
The mean age of participants was 16.21±2.42. All of participants were female. The majority of them were student of high school (94.9%). The education level of the majority of parents of participants (69.1%) were elementary, 18.4 % were illiterate, and the rest educated up to high school. According to the classification of suicide probability, 91 individuals (60.7%) was at low-risk category, 57 individuals (38%) at average risk, and 2 (1.3%) at high risk. 
Table 1 represents the results of descriptive statistics bivariate correlation analysis between self-harming behavior with psychological and peers’ related factors.


There was significant strong and direct correlation between self-harming with emotion dysregulation (r=0.85), peers modeling (r=0.77), and suicide probability (r=0.87) (P<0.001). 
The reliability of the variables was assessed by composite reliability. The results showed that the minimum and maximum composite reliability values were related to emotion dysregulation (0.75) and self-harm (0.90), respectively, indicating the acceptable reliability of latent variables. The convergent validity of latent variables was tested with AVE. The lowest AVE (0.52) was related to emotion dysregulation and the highest AVE (0.67) belonged to “self-harming”. Since the mean AVE was greater than 0.5, the convergent validity of latent variables was confirmed. Also, due to the larger CR values than AVE, there was a convergent validity condition. 
Figure 1 shows the structural equation model of research variables.

The results showed that all direct and indirect relationships are significant at the 0.05 level. The strongest relationship was observed between self-harming behaviors(β=0.86, P=0.001) and emotion dysregulation (β=0.34, P=0.001) and weakest relationship was between peers modeling and suicide probability (β=0.25, P=0.04). The model account for 87 % of total variance of suicide probability.
The results represented in Table 2 shows significant indirect effect of self-harming behaviors to suicide probability through emotion dysregulation (β=037) and peers modeling (β=0.17). 


4. Discussion 
High prevalence of self-harming behavior in adolescence lead to permanent outcomes in their future life. On the other hand, these behaviors are increasing in girls. Hence, the study was conducted in order to analyze the effect of behavioral (self-harming), cognitive (emotion dysregulation), and environmental (peers) factors in suicide probability. The results indicated the indirect and direct relation between self-harming and suicide probability with emotion dysregulation. These findings were in consistent with findings of Hasanvand et al., Peh et al., and Breton and Macglinchi [25, 26, 27]. Kedmati [16] concluded that difficulty in emotion regulation could predict the suicide probability and self-harming. 
To explain this finding, inability in accepting emotional strategies may cause to negative emotions. People with low emotional regulation ability show more irritabilities in response to negative emotions. Since adolescent with suicide thoughts cannot avoid negative emotions, they cannot recognize appropriate strategy to face stressful situations. They do not have enough access to emotional strategies and this leads to choose maladaptive strategies. Because of higher impulsive behaviors related to emotional dysregulation the probability of risky behaviors such as drug abuse and self-harming is prevalent among them [28]. According to emotion-regulation model the adolescents with more emotional dysregulation who have problems in accepting, evaluating and controlling emotional responses have more potential for self-harming suicide behaviors [27].
The results also indicated the indirect and direct relation between self-harming and suicide probability with peers’ roles. In line with the findings in the study by Khanipour [9]; Prinstein et al. [29] highlighted the role of peers in self-harming. In a model presented by Nouk, one of the main reasons for teenagers with self-harming is delivering social messages. This leads to social responses as it is a high risk behavior and impose cost on family and social systems. The second motive for non-suicidal inter-personal injury is self-affirmation. It is more in delinquent adolescents to demonstrate their abilities to others. This was more evident in our study because the group were delinquent. They believed they are identified as a popular and powerful individual among peers. Self-other distinction is another interpersonal non-suicidal self-harming motive. The teenagers who have difficulty in understanding, accepting and extending compassion to themselves might display self-injurious behavior to be distinguished from others and achieve a feeling of self-consistency.
 In general, the findings regarding the indirect and direct relationship between emotional regulation and peer modeling with self-harm and the possibility of suicide can be explained by several models. According to the four-function model [6], self-harm is a maladaptive emotional strategy that is used by people with emotional regulation deficits in stressful situations. Also, based on the experiential avoidance model, self-harm behavior is considered as an avoidance behavior in distress conditions to avoid distress conditions. This theory highlights the role of emotional skills. Also, Bandura’s social cognitive model can be helpful in explaining this finding because it justifies the peer modeling of teenagers. It is possible that one of the peers achieves their desires with self-harming behaviors and pretending to commit suicide, and this makes others eager to copy their behavior for self-harming or tendency to suicide.
5. Conclusion
The results of the study showed was self-harming was strongest predictor of emotion dysregulation, thus it is suggested to pay more attention on self-harming behaviors in rehabilitation centers. As the emotion dysregulation could predict the suicide probability, it is recommended to give necessary information about relation between variables and consider them in creating mediators and consultation to correct and improve the emotion dysregulations of cases with self-harming. 
The study had limitations as it was confined to delinquent teenagers of rehabilitation center of Mashhad and it is not definite whether the research variables have the same share in predicting other populations. The research was done on a limited number of people; thus generalizing the results to other studies should be done carefully. It studied the relation between self-harming behaviors with limited variables, while other variables including emotional intelligence, personality and gender can contribute.

Ethical Considerations
Compliance with ethical guidelines
The study protocol was approved by Islamic Azad University, Torbat Jam Branch (Code: IR.TRJUMS.REC.1398.011). 

Funding
This research did not receive any grant from funding agencies in the public, commercial, or non-profit sectors.

Authors' contributions
All authors equally contributed to preparing this article

Conflict of interest
There is no conflict of interest in this study

Acknowledgements
I would like to express my deep gratitude to dear participants, manager and staff of the Prisons Organizations of Mashhad who have provided me with great assistance and this work would have not been possible without their support. 

1. McManus S, Gunnell D, Cooper C, Bebbington PE, Howard LM, Brugha T, et al. Prevalence of non-suicidal self-harm and service contact in England, 2000-14: Repeated cross-sectional surveys of the general population. Lancet Psychiatry. 2019; 6(7):573-81. [DOI:10.1016/S2215-0366(19)30188-9] [PMID]
2. Ezakian S, Mirzaian B, Hosseini SH. [A review on non-suicidal self-injury in Iranian young adults and adolescents (Persian)]. Clin Excell. 2018; 8(2):14-25. [Link]
3. Schneider LA, King DL, Delfabbro PH. Maladaptive coping styles in adolescents with internet gaming disorder symptoms. Int J Ment Health Addict. 2018; 16(4):905-16. [DOI:10.1007/s11469-017-9756-9]
4. Mars B, Heron J, Klonsky ED, Moran P, O’Connor RC, Tilling K, et al. Predictors of future suicide attempt among adolescents with suicidal thoughts or non-suicidal self-harm: A population-based birth cohort study. Lancet Psychiatry. 2019; 6(4):327-37. [DOI:10.1016/S2215-0366(19)30030-6] [PMID]
5. Khosravi M, Kasaeiyan R. The relationship between neuroticism and suicidal thoughts among medical students: Moderating role of attachment styles. J Family Med Prim Care. 2020; 9(6):2680-7. [DOI:10.4103/jfmpc.jfmpc_1200_19] [PMID] [PMCID]
6. Nock MK. Why do people hurt themselves? New insights into the nature and functions of self-injury. Curr Dir Psychol Sci. 2009; 18(2):78-83. [DOI:10.1111/j.1467-8721.2009.01613] [PMID] [PMCID]
7. Lockwood J, Daley D, Townsend E, Sayal K. Impulsivity and self-harm in adolescence: A systematic review. Eur Child Adolesc Psychiatry. 2017; 26(4):387-402. [DOI:10.1007/s00787-016-0915-5] [PMID] [PMCID]
8. Bjureberg J, Sahlin H, Hellner C, Hedman-Lagerlöf E, Gratz KL, Bjärehed J, et al. Emotion regulation individual therapy for adolescents with nonsuicidal self-injury disorder: A feasibility study. BMC Psychiatry. 2017; 17(1):411. [DOI:10.1186/s12888-017-1527-4] [PMID] [PMCID]
9. Yalvaç EBK, Gaynor K. Emotional dysregulation in adults: The influence of rumination and negative secondary appraisals of emotion. J Affect Disord. 2021; 282:656-61. [DOI:10.1016/j.jad.2020.12.194] [PMID]
10. Ghaderi M, Ahi Q, Vaziri S, Mansoori A, Shahabizadeh F. [The mediating role of self-criticism and distress tolerance in relationship between attachment styles and non-suicidal self-injury behaviors in adolescents (Persian)]. J Res Behav Sci. 2020; 17(4):537-52. [DOI:10.52547/rbs.17.4.552]
11. Asarnow JR, Berk MS, Bedics J, Adrian M, Gallop R, Cohen J, et al. Dialectical behavior therapy for suicidal self-harming youth: Emotion regulation, mechanisms, and mediators. J Am Acad Child Adolesc Psychiatry. 2021; 60(9):1105-15.e4. [DOI:10.1016/j.jaac.2021.01.016] [PMID]
12. Imani M, Taghva A, Shiralinia K, Kazemi M. The role of personal and occupational factors on suicide and self-harm behavior of military personnel. Ibn-e-Sina. 2015; 17(2):43-51. [Link]
13. Sansone RA, Wiederman MW, Sansone LA. The self‐harm inventory (SHI): Development of a scale for identifying self‐destructive behaviors and borderline personality disorder. J Clin Psychol. 1998; 54(7):973-83. [DOI:10.1002/(sici)1097-4679(199811)54:7aid-jclp11>3.0.co;2-h] [PMID]
14. Tahbaz Hoseinzadeh S, Ghorbani N, Nabavi S. [Comparison of self-destructive tendencies and integrative self-knowledge among multiple sclerosis and healthy people (Persian)]. Contemp Psychol. 2011; 6(2):35-44. [Link]
15. Garnefski N, Kraaij V. The cognitive emotion regulation questionnaire: Psychometric features and prospective relationships with depression and anxiety in adults. Eur Psychol Assess. 2007; 23(3):141-10. [DOI:10.1027/1015-5759.23.3.141]
16. Khedmati N. [The Relationship between rumination and difficulty of emotion regulation with self-harm behaviors in students (Persian)]. Rooyesh-e-Ravanshenasi J. 2020; 8(11):19-26. [Link]
17. Nixon MK, Levesque C, Preyde M, Vanderkooy J, Cloutier PF. The Ottawa self-injury inventory: Evaluation of an assessment measure of nonsuicidal self-injury in an inpatient sample of adolescents. Child Adolesc Psychiatry Ment Health. 2015; 9:26. [DOI:10.1186/s13034-015-0056-5] [PMID] [PMCID]
18. Khanipour H. [Self-injurious behaviors in adolescents: nature, possibility of suicide and the role of psychological factors and peers [Persian)] (PhD thesis). Tehran: Allameh Tabatabaie University; 2014. [Link]
19. Cull J, Gill W. Suicide probability scale (SPS) manual. Torrance: Western Psychological Service; 1982. [Link]
20. Eltz M, Evans AS, Celio M, Dyl J, Hunt J, Armstrong L, et al. Suicide probability scale and its utility with adolescent psychiatric patients. Child Psychiatry Hum Dev. 2007; 38(1):17-29. [DOI:10.1007/s10578-006-0040-7] [PMID]
21. Sharifian MS, Lavasani MG, Ejei J, Taremian F, Amrai K. The relationship among classroom community, attitude toward parents, anxiety disorders and depression with adolescent suicide probability. Procedia Soc Behav Sci. 2011; 15:520-5. [DOI:10.1016/j.sbspro.2011.03.134]
22. Hair JF. Multivariate data analysis: An overview. In: Lovric, M. International encyclopedia of statistical science. Heidelberg: Springer; 2011. [DOI:10.1007/978-3-642-04898-2_395]
23. Ringle CM, Da Silva D, Bido D. Structural equation modeling with the SmartPLS. Braz J Mark. 2015; 13(2). [DOI:10.5585/remark.v13i2.2717]
24. Cohen J. Statistical power analysis. Curr Dir Psychol Sci. 1992; 1(3):98-101. [DOI:10.1111/1467-8721.ep10768783]
25. Hasanvand M, Dabbaghi P, Rabiei M. [Assessment of relationship between negative emotion and non-suicidal self-injury behavior: With emphasis on the mediator role of body image and emotion regulation among soldiers (Persian)]. Ebnesina. 2016; 18(3):56-60. [Link]
26. Peh CX, Shahwan S, Fauziana R, Mahesh MV, Sambasivam R, Zhang Y, et al. Emotion dysregulation as a mechanism linking child maltreatment exposure and self-harm behaviors in adolescents. Child Abuse Negl. 2017; 67:383-90. [DOI:10.1016/j.chiabu.2017.03.013] [PMID]
27. Brereton A, McGlinchey E. Self-harm, emotion regulation, and experiential avoidance: A systematic review. Arch Suicide Res. 2020; 24(sup1):1-24. [DOI:10.1080/13811118.2018.1563575] [PMID]
28. Laghi F, Bianchi D, Lonigro A, Pompili S, Baiocco R. Emotion regulation and alcohol abuse in second-generation immigrant adolescents: The protective role of cognitive reappraisal. J Health Psychol. 2021; 26(4):513-24. [DOI:10.1177/1359105318820715] [PMID]
29. Prinstein MJ, Heilbron N, Guerry JD, Franklin JC, Rancourt D, Simon V, et al. Peer influence and nonsuicidal self injury: Longitudinal results in community and clinically-referred adolescent samples. J Abnorm Child Psychol. 2010; 38(5):669-82. [DOI:10.1007/s10802-010-9423-0] [PMID] [PMCID]