An investigation of self-injury in female adolescents: a qualitative study

Abstract

This study aimed to investigate the phenomenon of self-injury among female adolescents. The research was qualitative, and the sampling method was purposive non-random; the sample size was 20 Iranian girl adolescents aged 13–15 years who had experienced non-suicidal self-injury. Data were collected through semi-structured interviews. The data analysis process was performed during three coding steps (open, axial, selective), through which the basic codes and categories were identified. Study results indicate that the main factors in adolescents' self-injury were individual or psychological (thoughts, emotions, and behaviors) and social (peers, family, communication with different gender, communication with others, media/cyberspace, school, and economic). In the former, the role of emotions was remarkable, while in the latter, the family played a key role. Further, results revealed that communication within the family was an important motivating and sustentative factor in adolescents' self-injury. The results can help counselors in working with adolescents who self-injure; results can also be used in the development and implementation of treatment plans.

Introduction

Adolescence is a phase of growth biologically, psychologically, and socially. Functional and structural changes in the brain are paralleled by psychosocial development (Foulkes and Blakemore 2018). In this period, adolescents need balancing, especially between emotions and cognitions, understanding the existential value of self, improving self-awareness, choosing goals in life, being independent of the family emotionally, maintaining psychological and emotional stability against environmental stressors, and establishing healthy relationships with others (Dryfoos 1991). If these characteristics of adolescence are ignored, there is a potential risk of many unhealthy behaviors, one of which is intentional self-injury (Dryfoos 1991).

Due to its increasing prevalence, non-suicidal self-injury (NSSI) was included in the diagnostic classes of the fifth edition of the Diagnostic and Statistical Manual of Mental Disorder (Zetterqvist 2015). NSSI is defined as intentional and deliberate body tissue injury with no suicidal intention, including self-hitting, cutting, or burning. NSSI does not include unintentional and indirect self-injury, suicidal attempts, and socially acceptable behaviors, such as piercing, tattooing, or religious practices (Hooley et al. 2020).

Currently, there are at least 33 terms related to self-injury. Such confusion and uncertainty in the definition of variables affect the identification and exploration of the nature of self-injurious behavior (Han 2019).

There are several models for classifying self-injury. According to one of these models, self-injury is divided into two categories: (1) stereotypic self-injury prevalent in pervasive developmental disorders and psychotic symptoms and (2) impulsive self-injury occurring as impulsive behaviors (Jacobson and Gould 2007). In an alternative classification, self-injury behaviors are divided into direct and indirect classes. The former refers to intentional actions that directly cause harm to the body (e.g., cutting), and the latter refers to behaviors that do not intentionally and directly injure body tissue (e.g., abusive relationships, substance abuse, risky or reckless behavior, or eating disorder; (Germain and Hooley 2012). In their Serbian youth sample, Kostić et al. (2019) found the most common method of self-injury to be cutting (60%), followed by severe biting and scratching (14%).

One of the questions raised about non-suicidal self-injury is how to determine its relationship with suicide attemptsNon-suicidal self-injury and suicide are distinguishable concerning intention, methods, chronicity, lethality, cognitions, responses, outcomes, prevalence, and demography (Muehlenkamp 2005; Walsh 2012). Studies have shown comorbidity among these two behaviors, even though they are regarded as independent and different. Nock et al. (2006) reported a total of 70% of adolescents involved with NSSI have a lifetime suicide attempt while 55% have multiple suicide attempts. It is alarming that those who engage in NSSI are at increased risk for attempting suicide (Klonsky and Glenn 2009) and NSSI is one of the main predictors for the next suicide and self-injury attempts. Baseline NSSI history and hopelessness are significant predictors in the most parsimonious model predicting time to incident suicide attempts (Asarnow et al. 2011). According to these results, a significant overlap is observed between suicide attempts and NSSI.

Studies indicate that the rate of self-injury behaviors in adolescents is prevalent (Prinstein et al. 2010) across cultures. For instance, Muehlenkamp et al. (2012) found an 18% lifetime prevalence for samples across the globe, whereas Monto et al. (2018) found a 17.7–30.8% prevalence in adolescent girls and a 6.4–14.8% prevalence in adolescent boys in the United States. Plener et al. (2016) found a 3.1% lifetime prevalence in the German population, Tang et al. (2018) found a 12.2% prevalence in Chinese adolescents, and Gandhi et al. (2018) found a 21% prevalence in adolescents of Dutch-speaking part of Belgium. These studies highlight the prevalence of NSSI across diverse populations.

Across cultures, NSSI tends to emerge during early adolescence. According to a systematic review of longitudinal investigations on NSSI, the peak NSSI prevalence is 15–16 years (Plener et al. 2016). Kostić et al. (2019) examined NSSI behavior among adolescents in southeastern Serbia. This study, which utilized a sample of 50 adolescent girls and boys aged 13–18 years, demonstrated an average age of onset of 14.12 years for NSSI. Gandhi et al. (2018) found the probability of age of onset peaked around the age of 14–15 years in Dutch Belgium adolescents. A study of a sample of Caucasian, African American, Asian, and biracial adolescents identified the average age of NSSI onset was 13.90 years (Ammerman et al. 2018).

Gender disparities in NSSI are of interest to researchers and clinicians alike. A study of 15- to 16-year-old adolescents in Australia, Belgium, England, Hungary, Ireland, the Netherlands, and Norway shows that girls self-injure at a rate of almost double that of boys, and in four out of seven countries, at least one in ten girls has self-injured within the past year (Madge et al. 2011). Based on a meta-analysis conducted by (Bresin and Schoenleber 2015), it is more likely for female youth and adults to engage in NSSI compared to males. Intentional self-injury is a common and often veiled issue in adolescents, particularly in females, indicating differences and similarities globally (Madge et al. 2008).

To understand self-injury behaviors in adolescents, it is important to know why a particular behavior occurs at a particular time, under a particular condition, and by a particular individual (Suyemoto 1998). Adolescent self-injury behaviors are influenced by individual, social, and cultural factors (Han 2019). Forrester et al. (2017) have indicated a significant negative relationship between self-esteem and self-injury; their results revealed that self-esteem is lower in people who experience NSSI than others who do not self-injure. Further, Baldwin et al. (2019), studying 2232 English children from birth to 18 years old, indicated that the risk of self-injury increases in survivors of abuse, neglect, sexual assault, domestic violence, peer or sibling bullying, cyber-bullying, and interpersonal crime. Laye-Gindhu and Schonert-Reichl (2005) found that adolescents who self-injure report significant increases in antisocial behaviors, emotional distress, anger problems, and health-threatening behaviors, as well as decreased self-esteem. The results of this study highlight the importance of emotion in NSSI. Although research on self-injury has increased rapidly in recent years, more issues (individuals, society, culture) are still to be identified and explored in more depth (Han 2019).

Models of NSSI offer a means through which to understand its emergence and how it sustains across time. Intentional self-injury is described by the Experiential Avoidance Model as a way of escaping from or avoiding undesirable emotional experiences through negative reinforcement (Chapman et al. 2006). Although applicable in some cases, this model provides a narrow explanation of NSSI. Providing a more expansive conceptualization, Nock and Prinstein (2004) proposed a four-factor theoretical model, with two dimensions, to explain NSSI. Firstly, according to Nock and Prinstein (2004), NSSI is either reinforced intra-personally and automatically (e.g., to decrease tension or make a more favorable state) or socially (e.g., to change one's environment). Secondly, NSSI is reinforced either positively (i.e., rewarded with a positive stimulus) or negatively (i.e., rewarded by escaping a negative interpersonal demand). This model provides some support for NSSI as a learned behavior, which might presume that the behavior can be altered or unlearned.

Researchers have also explored the functions of NSSI—the purposes it serves for those who self-injure. Klonsky and Glenn (2009) identified 13 functions of NSSI (categorized into interpersonal and intrapersonal functions), which apply to regulate emotion, establish interpersonal boundaries, punish oneself for perceived flaws or mistakes, care oneself, avoid dissociation, prevent suicide, seek sensation, build peer-belonging, influence others interpersonally, increase toughness, show distress, get revenge, and achieve autonomy. Clients may report multiple functions of NSSI (Turner et al. 2012) and those functions can change across time.

Emotion/affect regulation is a commonly reported function of NSSI (Andover and Morris 2014). As such, much attention has been given to understanding emotion, alexithymia, and emotion regulation as they pertain to NSSI. Research indicates that alexithymia is a multi-dimensional deficit in emotion recognition and regulation. Alexithymia includes four main aspects: (1) difficulty identifying and describing subjective feelings, (2) trouble differentiating between feelings and the physical sensations of emotional arousal, (3) limited imaginative processes, and (4) an externally-oriented cognitive style (Timoney and Holder 2013). A meta-analysis showed significant positive relationship between alexithymia, difficulties identifying feelings, difficulties defining feelings, and NSSI (Greene et al. 2020).

Emotion regulation skills are divided into six sub-skills, including emotional acceptance, emotional awareness, goal-directed behavior during the emotional experience, impulse control, access to emotion regulation strategies (emotional skills), and emotional clarity (Gratz and Roemer 2004). Deficits in emotion regulation skills have been implicated in NSSI by both theory and research. Research shows that NSSI is performed as an emotion regulation, as it often decreases the experience of negative emotion. Interventions have demonstrated effectiveness in reducing NSSI involve a focus on emotion regulation skills (Andover and Morris 2014). Diverging slightly from affect regulation as a primary function of NSSI, Lloyd-Richardson et al. (2007) found that trying to get the attention of someone, exercise control over a condition, and prevent negative feelings are the main reasons for NSSI. Preventing negative feelings might be akin to a form of affect regulation, thus highlighting the important role of emotion in NSSI. Examining the functions NSSI serves for each person is fundamental to therapeutic intervention; this process is called a “functional analysis” of NSSI (Bentley et al. 2017). However, it is important to note that due to the relationship between NSSI and alexithymia (Norman and Borrill 2015), people who self-injure may find it difficult to identify the functions NSSI serves for them.

The prevalence of self-injury behaviors among adolescents is concerning. There is limited information on the lived experience of NSSI and factors that influence NSSI from the adolescent perspective. Therefore, the present study, using a qualitative research method, aimed to investigate and identify the nature and factors related to self-injury behaviors with no suicidal ideation among Iranian female adolescents.

Literature Review

Research on NSSI has examined the etiology of this behavior from different perspectives, including developmental, intrapersonal, and social factors. Additionally, the literature supports a coping function of NSSI, in which NSSI might actually serve an anti-suicide function (Klonsky et al. 2014). Although various factors may contribute to the development of NSSI, the interaction of factors may best explain the emergence of this behavior.

Nock (2009) integrated theoretical model suggests an interaction of etiological factors, such that both intrapersonal (e.g., affect regulation) and interpersonal factors (e.g., help-seeking) contribute to the development of NSSI, danger of NSSI is increased by factors that contribute to problems with emotion regulation or interpersonal relationship (e.g., childhood abuse) and by special factors that influence the decision to use NSSI rather than some other behavior to serve these functions (e.g., social modeling). Psychological, psychiatric, familial, cultural, and social factors, as well as genetic vulnerability, are major contributors to self-injury (Hawton et al. 2012).

Emotion plays an important role in the experience of self-injury. Edmondson et al. (2016) found that one of most common reasons for NSSI is distress. Further, depression and anxiety are related to self-injury during adolescence (Hankin and Abela 2011; Jacobson and Gould 2007; Moran et al. 2012). For many people who self-injure, NSSI is applied to deal with stressful adverse affective states, particularly depression, anger and mixed emotional conditions, NSSI is related to various internalizing and externalizing situations (Peterson et al. 2008). According to Chapman and Dixon-Gordon (2007) anger is the most commonly reported emotion preceding self-injury and, following self-injury, a positive emotional shift characterized by relief and diminished anger is commonly experienced. A study by Madge et al. (2011) in secondary schools in Belgium, the United Kingdom, Hungary, Ireland, the Netherlands, Norway, and Australia demonstrated that increased self-injury is associated with high levels of depression, anxiety, impulsivity, and low self-esteem. The researchers stated that psychological characteristics and stressful life events reinforce adolescent self-injury, and that these factors are more severe in girls.

Emotional regulation may also affect self-injury (McKenzie and Gross 2014). Based on a study of a sample of female students, limited access to effective emotion regulation strategies and emotional disclarity can predict NSSI (Gratz and Roemer 2008). In further research on hospitalized adolescents with different psychiatric diagnoses, a significant inverse relationship is reported between self-injury and emotion regulation (Perez et al. 2012). alexithymia (problem with recognizing feelings) and rumination are appeared to be the robust factor in predicting self-injury (Borrill et al. 2009).

The role of coping in NSSI has also drawn attention. Chapman and Dixon-Gordon (2007) research on the emotional antecedents and consequences of NSSI exemplifies the coping functions of self-injury. Further, a study by Borrill et al. (2009) indicated that people who self-injure tend to score significantly higher in maladaptive coping styles those who do not report NSSI. Moreover, Wan et al. (2020) suggest that, in order to prevent NSSI in adolescents, reducing negative coping styles in girls/boys and improving positive ones in girls with negative childhood experiences might be helpful.

Studies also suggest that both behavioral and cognitive/affective indicators of interpersonal problems predict NSSI onset (Victor et al. 2019). Edmondson et al. (2016) highlight the role of interpersonal factors in NSSI. Poor relationships in the form of adverse interactions in close relationships and lack of social support (Hankin and Abela 2011) and history of sexual abuse (Jacobson and Gould 2007) are risk factors for self-injury. People who engage in dysregulated behaviors, such as NSSI, commonly report adverse family environments (Hasking et al. 2020). Many studies show the variable of family functioning significantly predicts NSSI (Fortune et al. 2016; Halstead et al. 2014; Ren et al. 2018; Sitton et al. 2020). Family communication patterns, specifically conversation orientation, has a positive correlation with lifetime NSSI behaviors but not current NSSI behaviors and competence in communicating increases, NSSI behaviors decrease, In addition, alexithymia mediates the impact of communication skills on NSSI behaviors (Wester and King 2018). Parental harsh punishment, low parental monitoring, and poor quality of attachment to parent predicted new onset of NSSI (Victor et al. 2019). Moreover, Oldershaw et al. (2008) indicate that parents have challenges in understanding, resolving, and overcoming adolescent self-injury behavior; they need support and advice to manage self-harming behaviors.

The role of interpersonal factors in NSSI is noteworthy, particularly regarding social contagion. Peer factor (perceptions of peers, and perceptions of one’s own social competence and worth in relation to peers) predicted new onset NSSI (Victor et al. 2019). A study by Prinstein et al. (2010) suggests the role of peer socialization in the development of NSSI in adolescent females after controlling for depressive symptoms as a predictor. The researchers found that adolescents' NSSI was associated with increasing perceptions of their friends' engagement in depressive/self-injurious thoughts and behavior (Prinstein et al. 2010). Further, in a systematic review, social contagion related NSSI was found in 16 studies (Jarvi et al. 2013). According to this review, the initial involvement with NSSI, in particular, might be extremely affected by social contagion (e.g., acquaintances or friends involved in NSSI, or being exposed to NSSI in the media, particularly the Internet), whereas, the NSSI preservation is very likely to be associated with intrapersonal functions developing over time (Jarvi et al. 2013).

Media and contagion affect self-injury dramatically, with the internet playing a major role in contemporary role (Hawton et al. 2012). Jacob et al. (2017) have indicated that images rather than textual interactions are the main reasons for using the internet for self-injury; images evoke a physical reaction and motivate the action. Seeing pictures online plays a major role in the self-injury of many young people (Jacob et al. (2017) Increased attention is being given to social and interpersonal influences related to NSSI.

Social support and connection are protective factors in some African American and Hispanic youth (Rojas-Velasquez et al. 2020). Studies should concentrate on improving the adolescents’ sense of positive identity and connection to parents and prosocial friends (Taliaferro et al. 2020).

Method

The present study utilized a qualitative approach to capture participants’ perception and experience of self-injury from a phenomenological perspective. A qualitative approach is appropriate for this study because qualitative research generally seeks to examine individuals in their natural environment to examine how their behaviors and experiences are formed by the contexts, including the social, cultural, physical, and economic, in which they live (Hennink et al. 2020). Qualitative research allows the researcher to understand the meaning participants assign to their experience. Using Grounded Theory (Chun et al. 2019), the researcher was able to gain understanding of participants’ lived experience of self-injury and generate a theory/model through which to conceptualize their experience.

Research team and researcher reflexivity statement

This manuscript is based on the Ph.D. dissertation of first author. The research team included the first author (Ph.D. student in counseling and school counselor), second author (supervisor, Ph.D. in educational psychology and associate professor in University of Hormozgan (Iran)) and two advisor professors (third author, Ph.D. in counseling and associate professor in University of West Georgia (USA) and the fourth author, Ph.D. in family counseling and associate professor in University of Hormozgan (Iran)). Because the first author is a school counselor, she had some clinical experience working with adolescents who engaged in NSSI. The second author supervised the research work, analyzed the contents of interviewees, and assisted the research team in the coding process. Advisor professors (third and fourth authors participated in coding process, reviewing the extracted codes, and modifying them if needed. All team members had a relative high knowledge about NSSI, which could contribute to bias in the coding process. Therefore, issues of bias were discussed during coding meetings, and multiple measures were taken to increase trustworthiness.

Participants and procedure

The statistical population included adolescent females aged 13–15 years in Shiraz, Iran. The target individuals were identified using purposive sampling. The sampling design was on the basis of the researcher's opinion about who would offer the best information to achieve the goals of the research (Etikan and Bala 2017). The primary researcher is a school counselor and works in schools. She identified the sample through her place of employment. Participation was voluntary, as described below.

The inclusion criteria for the study consisted of the following: (a) adolescent age of 13–15 years, (b) female gender, and (c) a history of at least one episode of NSSI within the past year. Data saturation determined the sample size. In analyzing the data, the category of complete concepts and the relationship between the categories were well defined, and no new codes were extracted from the data analysis; thus, data saturation occurred (Saunders et al. 2018). This goal was achieved by conducting 20 interviews.

Data sources

Semi-structured exploratory interviews were performed to collect data. One of the major data-collection methods for qualitative research is interviews, which allow for collecting rich data from individuals in different roles and circumstances. Semi-structured interviews fall somewhere in between those of structured and unstructured. They involve using certain pre-formulated questions; however, the interview protocol is not adhered to strictly. Rather, during the interview, new questions can arise, which are welcomed. In the semi-structured interviews, the interviewee is provided with an opportunity to add central insights that emerge during the conversation, while formerly organized questions offer some focus (Myers 2019). Twelve points, stated by McGrath et al. (2019), were followed in the present research regarding the development and implementation of interviews: (a) identify when qualitative research interviews are appropriate; (b) prepare yourself as an interviewer; (c) construct an interview guide and test your questions; (d) consider cultural and power dimensions of the interview situation; (e) develop rapport with your respondents; (f) remember the researcher is the data co-creator; (g) talk less and listen more; (h) allow yourself to adjust the interview guide; (i) be prepared to handle unanticipated emotions; (j) transcribe the interviews in good time; (k) check the data; and (l) initiate analysis early.

Based on the indicators identified in the research literature, the interview protocol was compiled and provided to the supervisors and consultants for review. The interview protocol (Appendix A) consisted of ten questions regarding participants’ experience of NSSI, including information related to frequency, functions, interpersonal effects, and coping. After modifying and confirming the interview protocol, the interviews were conducted. This research project was approved in the Ethics Committee of Hormozgan University of Medical Sciences.

At the beginning of the interviews, the research purposes, procedures, benefits, voluntary nature of participating, and confidentiality were stated to the participants. The researcher confirmed participants’ understanding of their right to withdraw from the investigation at any time and in any manner. Further, a signed informed consent form was obtained from the participants. The duration of each interview was 45 to 60 min, depending on the circumstances and participants’ tolerance for discussing self-injury, and interest. Interviews were recorded and then transcribed on paper.

Data analysis

Data analysis was performed according to the Grounded Theory approach, as described by Corbin and Strauss (1990). They offered a thorough procedure-oriented method for coding, including open, axial, and selective coding phases.

Open coding involves breaking down, analyzing, comparing, conceptualizing, and classifying data (Bryant and Charmaz 2019). Open coding is the impressionistic method by which data is analytically decomposed. Its aim is to give new perspectives to the researcher by breaking through conventional ways of thinking about phenomena represented in the data or interpreting them. This puts together conceptually similar events/actions/interactions to create categories and subcategories (Corbin and Strauss 1990). A qualitative research code is most often a word or short phrase that symbolically attributes to a summative, prominent, essence-capturing, and/or evocative attribute to a segment of language-based or visual data (Saldaña, 2021). In the present study, this phase was initiated by the researcher coding paragraph by paragraph, line by line, phrase by phrase, or even word by word, to identify data that stood out as meaningful (micro-analysis). Open coding, therefore, was the extraction of central concepts and initial codes (Bryant and Charmaz 2019).

In axial coding, categories are related to their subcategories, and the relationships are tested against data. Also, further development of categories takes place, and the researcher continues to look for indications of the categories. Through the "coding paradigm" of conditions, context, strategies (action/interaction), and consequences, subcategories are related to a category. During the analytic process, the analyst can draw upon previous experience to think through the conditions. All hypothetical relationships proposed deductively during axial coding must be considered provisional until verified repeatedly against incoming data. Deductively arrived at hypotheses that do not hold up when compared with actual data must be revised or discarded. A single incident is not a sufficient basis to discard or verify a hypothesis. To be verified (that is, regarded as increasingly plausible) a hypothesis must be indicated by the data over and over again. An unsupported hypothesis must be critically evaluated to determine if it is false or if the observed events indicate a variation of the hypothesis. In this study, the codes gained from the prior phase were classified or divided into sub-classes and main classes (Corbin and Strauss 1990).

The third phase, selective coding, deals with settling on the main class, relating it to other classes, confirming the relationships, and removing any incomplete classes (Bryant and Charmaz 2019). Selective coding is the process by which all categories are unified around a "core" category, and categories that need further explication are filled-in with descriptive detail (Corbin and Strauss 1990). The core category represents the central phenomenon of the study. It is identified by asking questions such as: “What is the main analytic idea presented in this research? If my findings are to be conceptualized in a few sentences, what do I say? What does all the action/interaction seem to be about? How can I explain the variation that I see between and among the categories?” (Corbin and Strauss 1990). Therefore, in the present work, during selective coding, the researcher attempted to find relationships between classes to formulate a theory of adolescent self-injury.

Validity and reliability

In the organizational field, reliability and validity of research and its results are central factors in determining the research quality. No common standards for determining the validity of qualitative studies are recognized due to differences in their approaches and methods. As qualitative research should not be the results of a single test or a phase in the research, they should follow a systematic validity approach (Hayashi Jr et al., 2019). In qualitative studies, validity may have various meanings, including appropriateness, trustworthiness, rigor, and even quality, being defined in different terms (Golafshani 2003).

In the present study, the following methods were used to achieve validity and reliability. First, interviews were carefully recorded and transcribed in detail, which allowed for accuracy of the original data prior to interpretation through coding. Second, the researcher used member checking to verify the analysis (Lincoln and Guba 1985). The data analysis and results were presented to a number of participants, and their reactions were used for results validity. In this regard, the participants did not agree on the meaning units related to the concepts of anxiety, peer inhibition, judgment of others and blame and after discussing each of the disputed categories, the cases were modified. The rest of the categories were approved. Furthermore, in open coding, the participants checked the units of meaning and according to their reactions (the discrepancies in wording of units) corrections were made in the wording of units. In the axial coding, the results was approved by all but in the selective coding, participants declared to state the family as the most important factor, which was modified. Third, researcher reflexivity was utilized (Pandey and Patnaik 2014), wherein a researcher self-review was performed repeatedly during the data collection and analysis processes. And fourth, the contents of the interviews and the coding were provided to three experts for review, who provided an audit of the coding. Combined, these measures were intended to reduce the potential influence of research bias on data analysis.

Results

Data analysis was performed so that in the first stage, the texts were read one by one, and the concepts contained in it were identified. In this step, open coding, a conceptual code was assigned to each smallest significant unit of textual content. Thirty-two concepts were observed, which were named according to the researcher's inference and closely related to the data. The open coding continued until no new information was obtained from the texts, and it was ensured that all the concepts in the texts came under the title of a code (i.e., data saturation). Appendix B presents the results of open coding, including the code enumeration. The codes (number of codes) are: Self- deprecation (12), Identity (1), Religion (3), Feeling lonely (7), Feeling misunderstand (6), Temporary catharsis (19), Depression (19), Anger (18), Anxiety (7), Hate (5), Repression of emotions (19), Coping strategy (6), Attention seeking (5), Manipulative behavior (3), Imitation (13), interpersonal pressure (5), Prohibition (12), Interpersonal difficulties (3), Communication with family (18), Disagreement or separation from parents (5), Disagreement and separation from boyfriend (9), Abuse (1), Others reaction (7), Others judgment (11), Fear of hurting others (4), Picture (8), Video (8), Music (5), Blame (6), Fear (1), Habitation (1), and Working and financial issues (5).

In the second phase, axial coding, several concepts or codes with a common meaning and theme were named under a more comprehensive title or category. These categories were considered sub-categories. Axial coding has a higher abstraction and searches for patterns embedded in the text (Corbin and Strauss 1990); therefore, in this process, a deeper description of the text was obtained, and finally, the sub-categories were summarized in the main categories in the current work. Axial coding results are presented in Appendix C. The sub-categories (number of them) are: Thoughts (16), Emotions (100), Behaviors (14), Peers factors (33), Family factors (23), Communication with different gender (10), Communication with others (22), Media and cyberspace (21), School (7), Economic factors (6). The main categories (number of them) are: Individual or Psychological factors (130), and Social factors (122).

In the selective coding stage, the researchers, based on their understanding of the adolescent self-injury phenomenon, tried to deduce the relationships between the categories and express them in the form of a theory. Based on the results obtained in this study, the factors related to adolescent self-injury could be divided into two general categories of the individual or psychological factors (Code number: 130) and social factors (Code number: 122). Individual or psychological factors include any factor affecting NSSI that may be related to one of the respondent’s individual characteristics. For instance, the interviewee's response, which attributes NSSI to feeling lonely, is considered as an individual factor (emotional factor). Individual or psychological factors are divided into three subsets: (a) thoughts, (b) emotions, and (c) behaviors. Based on the agreement between the authors, the factor or category of thoughts refers to the cognitions and beliefs related to NSSI, which can trigger such an action on the part of the adolescent (for instance self-deprecation belief). The category of emotions refers to feelings related to NSSI. Emotions that trigger or reinforce behaviors that leads to NSSI (feelings such as anger). The category of behaviors includes behaviors that are related to self-injury actions. These behaviors can include both self-injury actions and behaviors that facilitate or prevent these actions (for instance the manipulative behavior). As observed, the main role is assigned to emotions in adolescent self-injury.

Individual or psychological factors

Mental activity produces thoughts, which consist of an aim-oriented flow of concepts and relations, resulting in a reality-oriented conclusion. Thinking enables people to interpret, represent, or model the world they experience, thus predicting world-related affairs (Rajvanshi 2010). Based on the results of the interviews in this research, adolescent self-injury-related thoughts include three cases: (a) self-deprecation (i.e., developing negative thoughts and perceptions of self after self-injuring); (b) identity (i.e., questioning oneself and one’s purpose in life); and (c) religious beliefs (i.e., feeling guilty and remorseful after self-injuring).

Cabanac (2002) stated there is no consensus in the contexts on a description of emotion; the word is taken for granted in itself. Emotions are biological and mental states that lead to psychological and physical changes affecting people's behavior. In this research, participant statements revealed that feeling lonely, misunderstood, depressed, angry, or hateful can contribute to self-injury in adolescents, as can repression of emotions. Moreover, although the feeling of emptiness and temporary relaxation after self-injury can be a positive reinforcement for self-injuring, anxiety after self-injury can be a deterrent.

From a behavioral perspective in this research, self-injury is associated with the concepts of coping, attention seeking and manipulative behavior, both of which reinforce self-injury. Adolescents in this sample report harming themselves because they feel under pressure, and self-injuring is the most effective means they have found to cope with the pressure. Adolescents in this sample may also think that they can attract the attention and love of others or otherwise get what they want by engaging in NSSI.

Social factors

Based on an analysis of the data in this study, the social factors are associated with adolescent self-injury. Based on the agreement of the authors, social factors include any factor affecting NSSI that may be related to the respondent's social environment that includes family factors, peer factors, communication with different gender, communication with others, media and cyberspace, school and economic factors. For example, the interviewee's response, which attributes the NSSI to divorce or parental conflict, falls into the category of social factors (Family factors). According to the results of this study, relationships with peer groups can be both motivating and inhibiting. Imitation, interpersonal pressure, and interpersonal difficulties are ways that peer groups affect adolescent self-injury in this sample as motivating factors. Prohibition is a way that peer groups affect adolescent self-injury as an inhibiting factor. Adolescents in this sample reported that the type of relationship they have with their family is a very important and determining factor. In this research, tensions in the family, divorce, rejection, and lack of communication skills can cause stress and self-injury in adolescents. Additionally, results of communication with different gender in the form of infidelity or traumatic sexual experiences (rape) can cause stress and, thereby, trigger adolescents’ self-injury in this sample. The factor of communication with others is related to self-injury outcomes. Participants in this study reported that their self-injury causes others to feel upset and angry, and others blame adolescents for NSSI. However, participants also reported feeling concerned about others' negative perceptions and judgments about them, which contributes to participants’ tendency to hide their wounds so that others do not discovery their self-injury. Seeing photos and videos or listening to music in cyberspace can also be a motivating factor or trigger for self-injury. Participants also reported school factors as results of self-injury, which might include issues such as being reprimanded or intimidated by school staff. Further, participants discussed economic factors, including unemployment and income, as stressors that contribute to their self-injury.

Summary

In general, the results of this study showed that NSSI in adolescents is complex and a multidimensional phenomenon, influenced by various factors simultaneously. Among the influential individual factors (thoughts, emotions, and behavior), the role of emotions is fundamental. In particular, depression, anger, and repression of emotions tend to lead to self-injury, and temporary catharsis can be an important reinforce for subsequent self-injury behavior. Among the influential social factors (peers, family, relationships with the different sex, relationships with others, media and cyberspace, school, and economic), the role of the family is prominent. Almost all participants reported communication problems with family members, especially parents. In most cases, despite a healthy relationship between parents together and an appropriate atmosphere in the family, conflict or disagreement is formed between parents and adolescents due to the application of special parental restrictions due to the puberty characteristics of adolescents. Therefore, it seems that among all individual and social factors, family-related issues play a significant role in the development and perpetuation of self-injury within this sample.

Discussion

The results of this study demonstrate that factors affecting NSSI are divided into two general categories of individual or psychological factors (thoughts, emotions, and behaviors) and social factors (peers, family, communication with the different sex, communication with others, media and cyberspace, school, and economic). Among the individual and social factors, the role of emotions and the family are important, respectively. Adolescents’ relationship with their family can be a motivating and maintaining factor for NSSI.

Nock (2009) integrated theoretical model, Klonsky and Glenn (2009) motives of NSSI, and Han (2019) statements are consistent with the results of this study regarding the impact of individual and social factors on NSSI. Nock (2009) expressed that both intrapersonal and interpersonal factors are effective on NSSI. Klonsky and Glenn (2009) identified 13 functions of NSSI that categorized into interpersonal and intrapersonal functions. Han (2019) stated self-injury in adolescents is influenced by individual, social and cultural factors.

The results of our research on the effect of individual factors (thoughts, emotions, and behaviors) in adolescent NSSI are consistent with previous studies. The factor of thoughts involved self-deprecation (negative thoughts and perceptions of self), identity (questioning oneself and one’s purpose in life) and religion (feeling guilty and after self-injuring). Earlier studies have suggested the role of low self-esteem in NSSI (Forrester et al. 2017; Laye-Gindhu and Schonert-Reichl 2005; Madge et al. 2011) that are consistent with self-deprecation. Taliaferro et al. (2020) identified that a sense of positive identity prevents NSSI, which is consistent with the results of the present study regarding the role of identity in NSSI. Borrill et al. (2009) mentioned that rumination in NSSI is important and it is consistent with the result of this study regarding the role of thoughts in contributing to NSSI.

The results of the present study demonstrate that feeling lonely, misunderstood, depressed, angry, hateful or repression of emotions contribute to self-injury. The feeling of emptiness and temporary relaxation after self-injury can be a positive reinforcement for self-injuring, whereas anxiety after self-injury can be a deterrent. The research literature has pointed to the role of emotional regulation (Andover and Morris 2014; Gratz and Roemer 2004; McKenzie and Gross 2014; Perez et al. 2012), emotion recognition (Borrill et al. 2009), avoiding negative emotions (Chapman et al. 2006; Lloyd-Richardson et al. 2007), alexithymia (Borrill et al. 2009; Jacobson and Gould 2007), emotional dysregulation (Brown and Plener 2017; Rojas-Velasquez et al. 2020), depression (Hankin and Abela 2011; Jacobson and Gould 2007; Madge et al. 2011), anger (Chapman and Dixon-Gordon 2007; Laye-Gindhu and Schonert-Reichl 2005; Peterson et al. 2008), and anxiety or stress (Edmondson et al. 2016; Jacobson and Gould 2007; Madge et al. 2011; Moran et al. 2012) in NSSI.

Regarding the behavioral factors obtained in our study, adolescents are injuring themselves in order to cope with pressure, attract the attention and love of others, and/or get what they want. This aligns with Klonsky et al. (2014) noted function of NSSI to achieve interpersonal influence or communication. Additional research literature has referred to role of coping styles (Borrill et al. 2009; Wan et al. 2020) and attention seeking (Jacobson and Gould 2007; Lloyd-Richardson et al. 2007) in NSSI.

The results of current study on the impact of social factors (peers, family, relationship with different sex, communication with others, media and cyberspace, school, and economic) in adolescent NSSI are also consistent with previous studies. Imitation, interpersonal pressure, and interpersonal difficulties are ways that peers affect adolescent self-injury in this sample as motivating factors. Prohibition is way that peers affect self-injury as an inhibiting factor. Further, the results of this study suggest that tensions, divorce, rejection, and lack of communication skills in the family can cause stress and self-injury. Traumatic sexual experiences or rape are significant stressors that were implicated in NSSI in the current sample. Further, participants talked about the impact of their self-injury on others, causing their loved ones to become upset and anger, and to blame the adolescent engaging in NSSI. Adolescents reported feeling concerned about others' negative perceptions and judgments about them. Participants also talked about seeing photos and videos or listening to music in cyberspace as a motivating factor or trigger for self-injury. Additionally, they reported that being reprimanded or intimidated by school staff and experiencing unemployment or income issues are stressors that contribute to their self-injury. In earlier studies, the social factors include the role of neglect (Baldwin et al. 2019), family communication patterns, specifically conversation orientation and competence (Victor et al. 2019), sibling relationships consist bullying (Baldwin et al. 2019), family factors like parent/child interactions, inter-parental relationships, family dynamics, family satisfaction, flexibility, … (Fortune et al. 2016; Halstead et al. 2014; Hasking et al. 2020; Taliaferro et al. 2020), interpersonal effects like interpersonal stressors, adverse interactions, lack of social support, … (Brown and Plener 2017; Edmondson et al. 2016; Hankin and Abela 2011), poor relationships with friends and peers socialization effects (Jarvi et al. 2013; Prinstein et al. 2010; Victor et al. 2019), sexual abuse or sexual violence (Jacobson and Gould 2007), support social for prevent NSSI (Rojas-Velasquez et al. 2020), and media and internet (Baldwin et al. 2019; Jacob et al. 2017; Jarvi et al. 2013).

We believe that cultural factors have a major impact on adolescent NSSI. In particular, cultural differences between Western and Eastern countries can be noted. As far as the culture of Iranian adolescents is concerned, we can refer to cultural influences such as the family parenting system, restrictions imposed by culture and lifestyle (for example, on how to dress, permission to go out with friends, etc.), the type of relationships adolescents have with the different sex (these relationships aren’t free in Iran. These relationships already exist in secret. Family and community prohibit the adolescents from this relationship), Iranians' attitudes toward emotions and their tendency to suppress and ignore them, and religious inhibitions. It seems that Iranian culture with an emphasis on emotional inhibition has a significant effect on NSSI. According to the findings of our study, emotional factors including emotional suppression or, conversely, temporary catharsis are closely related to the NSSI. Also, the religious context of Iranian society and the religious inhibitions emphasized by the family have been effective in preventing the occurrence of NSSI. However, being religious can also exacerbate NSSI by creating guilt in adolescents with NSSI.

In consideration of these findings, the following implications for counseling practice are noted. Professional counselors may work with adolescents using cognitive based techniques to address negative or self-deprecating though patterns that lead to or trigger NSSI. Existential techniques may be applied to address adolescents’ identity and sense of purpose in life. As documented in previous literature and highlighted in the present study, professional counselors may also work with adolescents who self-injure to improve non-harming coping skills, utilize problem-solving skills, and practice emotion regulation. Considering the prominent role of social and familial factors in NSSI within this sample, professional counselors may work closely with adolescents to improve interpersonal functioning and assertive communication skills (behavioral approach). Further, adolescents should be supported in attempts to develop or enhance a healthy support system. Finally, when working with Iranian adolescent females or clients from traditional cultures where emotional inhabitation is expected, cross-gender interaction is prohibited, and/or adherence to religious protocols is expected, counselors may help clients with using cognitive interventions (i.e., changing thoughts and attitudes), acceptance and commitment techniques (e.g., identifying emotions and accepting them, setting values and goals, feeling responsible for achieving them), Gestalt interventions (e.g., paying attention to feelings, conflicts, and gestalt unfinished), and reality therapy interventions from Glasser’s model (e.g., appropriate communication with others, identifying needs and plans, feeling responsible).

Limitations and implications for further research

Since most qualitative studies aim at identifying a phenomenon and presenting a theory about it based on qualitative data, the generalization of findings to other samples is not possible, which can be considered one of the main limitations of this research. Conducting face-to-face interviews is important in establishing communication, gaining trust, and obtaining information. Due to the spread of the coronavirus, restrictions have been placed on conducting face-to-face interviews; thus, a number of them (7 interviews) were conducted by telephone, which could have affected the participants’ interview response.

Based on this research results, it is suggested to design interventions that mainly focus on emotional regulation, as well as family relationships and functions, in adolescents who self-injure. By identifying the function of self-injury for each adolescent, a counselor can more appropriately adapt therapeutic interventions to meet individual needs. The findings of this study also help to more elucidate counselors’ conceptualizations of NSSI, which included considering it as a functional and adaptive emotional coping skill. The current study provides information relevant to the social issues include peers, family, media/cyberspace and school factors. Accordingly, it is suggested that family counselors and school psychologists in intervention programs address the effects of these factors on NSSI. Counselors may choose to treatment strategies which are rooted in family and school dynamics. Research with different communities and samples can also be performed. Finally, one of the cultural phenomena different from Western societies dedicated to Iranian culture is issues related to the relationships with the different sex. This is a factor that affects the experience of NSSI at a high level and may be a significant area for future research.

Appendix A

See Table 1.

Table 1.

Interview protocol

1. Have you ever hurt yourself?

2. What has changed since you first did this (location, method, frequency of behavior)?

3. What made you do this (what happened), and what kind of experience was it?

4. Please tell me about your feelings and thoughts before, during, and after the injury?

5. Did others understand? How did self-injury change your relationships with other people (family, friends, relatives, school)?

6. What was the effect of self-harm on you?

7. What affected your identity and self-perception? How did it affect the impression of others about you?

8. Do you do this again?

9. What do you think is necessary, or what should happen to prevent you from doing this again?

10. What else do you think about the experience of hurting yourself that is important, and I need to know?

Appendix B

See Table 2.

Table 2.

Summary of open coding results from interviews with adolescents

Meaning units	Open coding
	Concepts	Number of codes
I'm a weak person. I'm weak and can't handle my problems. I'm not strong in hardships. I'm not good, and the others don't love me. I'm unfortunate that soon fall in love a boy. I'm a sad person who can never live well. I'm a miserable person	Self- deprecation	12
Who am I? Why do I suddenly change and become like this?	Identity	1
I'm religious and feel guilty a lot that I punched the wall and hurt my body. I cut my hand, and I have to answer on doomsday. I feel guilty about hurting my body	Religion	3
I feel lonely. No one likes me, and I am far away from everyone. I don't want to be with people, and I don't trust them. There is no one to sympathize with. If there was someone that I could talk to, I didn't hurt myself, but there wasn't. My mom or friends couldn't sympathize with me	Feeling lonely	7
If there was a companion who listened to me, and I could confidently tell my feelings without fear, I didn't hurt myself. Why does no one understand me while I understand others? Talking to my family, they don't understand me, they deride me, or my mom sometimes beat me. My parents haven't understood me since I finished elementary school	Feeling misunderstand	6
I wanted to be empty and calm. It causes pain and ignition, but it calms me down. I felt ignition and comfortable. I was calming down with blood and pain. I get rid of nervousness and anger, and I am released. It takes the burden off my shoulders. I become carefree and neutral. It calms me down for two or three days. My grief decreased. My discomfort decreased, but it came back. At first, I felt empty, peaceful, and comfortable, but then I felt remorse. At first, I was relieved, but then I said it hadn't worth it. I calm down, but I'm not happy	Temporary catharsis	19
I'm so miserable. I am totally depressed and crying. I have no hope in life. I was very sad. I cried a lot. I am very tired. I am not happy. I felt defeat. I am very sorrowful. There is sadness on my heart. My heart was hurt, and it was broken to pieces. I feel pain and sadness. I'm bored. I want to die and be comfortable. I wish death	Depression	19
I like to drub someone. When I hurt myself, I thought that I was punching my parents and brother, who had made me angry. I was very angry. I was nervous. I was angry with my parents	Anger	18
What happens if my mother put her threats into action. I'm afraid of the future. I am very scared to lose my life. When I was hurting myself, I was afraid of losing my darlings. I was worry that my parents finding out my cousin and I are friends	Anxiety	7
I hate myself and my family. I hate others. I hate everyone. I hate my father	Hate	5
I can't tell others how I feel; thus, I repress and hurt myself. I don't cry in front of others because they get upset, and I get upset too. I kept my feelings in myself and emptied out on my hand. I can't express my feelings by talking. I get angry and sad, then keep in myself. So far, I have not been able to share my feelings with anyone	Repression of emotions	19
I am a timid and stressed person that I hurt myself because of quarreling with my parents. I don't blame myself; I'm not guilty and have the right to hurt myself because I am under much pressure. I have much pressure and endure a lot. I repeat self-injury because nothing can calm me down	Coping strategy	6
It attracted attention, and my friends hugged me and calmed me down. My friends treated me better and asked not to do it again; they said, " if you do, we'll be angry or beat you." I hid, but I would like my parents to understand as I did it due to their separation	Attention seeking	5
My parents took my phone because of the lessons; thus, I hurt to get the phone. I wanted to get my friend's opinion to reconcile with me. When my family found out, they gave me more freedom to prevent me from doing it again	Manipulative behavior	3
My cousin and friends have done it. I saw some kids (both outside and inside the school) had injured themselves. I learned to punch from peers. I had seen the kids (in the upper class) that I learned. The first time, I saw my friends did it, but they did not encourage me. I saw one of my friends had done it; thus, I did it to make a chummier relationship with her. My friends were doing it, and I wanted to test too	Imitation	13
My friend gave me a razor. Some of my friends encouraged me to do it. I wanted to prove my courage to my cousin	Interpersonal pressure	5
My friends forbade me. Most of my friends said, "do not do it." My friends beat me to prevent me. My friends left me when I hurt	Prohibition	12
My friends' personal problems break my nerves. I suffer a lot from my friends' unfaithfulness, and I am distrustful. I am very dependent on my friend, and she sulks over me, and I cry a lot	Interpersonal difficulties	3
I have a lot of disagreements with my family; they don't understand me and limit me a lot. I fight with my parents, and they care a lot about people's words. I am angry about the restrictions and doubts of my parents. My parents are suspicious of me. I argue a lot with my mother; she says, "do this and don't do this"; it makes me very busy and breaks my nerves. I dispute with my parents because they don't let me talk to my friends or go out; they restrict me about dressing, and I can't do what I love. The severities and restrictions of the family bother me. My parents are religious and strict, especially my mother. I am tired of my family. I want to run away from home because I am prisoner and feel gloom. My parents are good together but not with me, and I can't tell them everything because they have a lot of negative reactions. My parents are good together, but they argue with me. My father disputes a lot with my brother, but mom and dad are good together. I have unpleasant childhood memories of my parents. My relationship with my parents has been deteriorated since the sixth grade, especially my father who has a mental illness and addiction. I am afraid of my father finding out about self-injury because he constantly blames me. I hate my father because he restricts and blames me. I felt discriminated. My relationship with my parents, especially my mother, is inappropriate, there are a lot of disputes and discrimination or rejection; also, I dispute with my younger brother. I have a lot of problems with my brother and I feel discriminated. We have family problems and contention with relatives. I have an inappropriate relationship with my stepfather, who is a troubled and addicted person	Communication with family	18
After the divorce of parents, I live with my father and stepmother; I have a lot of problems with my father and quarrel together. There is a history of self-injury in my mother's life. Parents relationship is not good. My parents dispute with each other. I injured myself severely (I went to the hospital) to prevent my parents from separating	Disagreement between parents or divorce	5
My boyfriend left me. I was in a relationship with my cousin, and I sulked and hurt myself. I am very upset and angry with my boyfriend's betrayal. I want to take revenge. My boyfriend died. I remembered the memories with my boyfriend, and I was upset. My boyfriend's words were repeated in my mind	Disagreement and separation from boyfriend	9
I have abused by my cousin and another boy	Abuse	1
My mom was scared and didn't let me be alone, my friends quarreled and got angry with me, and my boyfriend quarreled with me and said not to do. My parents were sad, angry, and distrustful; they didn't talk to me that tormented me. My brother was very upset. My parents were very unhappy and displeased. My father's cheerlessness tormented me. I distanced from my friends because they said don't do it. My mom understood, but the same bad conditions existed. My mom was upset, and my friends spread a rumor. My mom was very angry, and my friends were worried and told the counselor to talk to me	Others reaction	7
I'm sorry and afraid it will remain in hand, and it will be bad for courtship and marriage. I am afraid of happening something that my father understands; it can get worse. I hid it, but my mom found out. I covered it so that others did not understand. First, I hid it, then my parents understood and blamed me and were upset. I wore long sleeves so that others did not understand. I hid, but the school understood and told my parents. I hide from the others because they say, "how are you and how do you dare." The others' opinions about me will be negative, and it will bother me. Others will say what a weak person you are or your family has trouble. Why did you do it? What did you lack? Or Whom are you influenced by? She is a very bad girl	Others judgment	11
I won't repeat it because my mother can be upset. When I was hurting myself, I was afraid of discomforting my darlings	Fear of hurting others	4
Looking at self-injury pictures in cyberspace, Instagram and Telegram channels	Picture	8
Watching self-injury in dep video, the Like app, the WhatsApp status of peers, clips on the phone	Video	8
Sad context and songs (singing and shouting)	Music	5
The school officials blamed and criticized me. The manager argued with me	Blame	6
I'm afraid of sending me to the school management committee	Fear	1
We didn't find a house to rent, and we had habitation problems	Habitation	1
My father has job problems. We have economic concerns. My father is always at work, and we don't have fun	Working and financial issues	5

Appendix C

See Table 3.

Table 3.

Summary of axial coding results

Open coding		Axial coding
Concepts	Number of codes	Sub-categories	Number of codes	Main categories	Number of codes
Self-deprecation	12	Thoughts	16	Individual or Psychological factors	130
Identity	1
Religion	3
Feeling lonely	7	Emotions	100
Feeling misunderstood	6
Temporary catharsis	19
Depression	19
Anger	18
Anxiety	7
Hate	5
Repression of emotions	19
Coping strategy	6	Behaviors	14
Attention seeking	5
Manipulative behavior	3
Imitation	13	Peer factors	33	Social factors	122
Interpersonal pressure	5
Prohibition	12
Interpersonal difficulties	3
Communication with family	18	Family factors	23
Disagreement between parents or divorce	5
Disagreement or separation from boyfriend	9	Communication with different gender	10
Abuse	1
Others’ reaction	7	Communication with others	22
Others’ judgment	11
Fear of hurting others	4
Picture	8	Media and cyberspace	21
Video	8
Music	5
Blame	6	School	7
Fear	1
Habitation	1	Economic factors	6
Working and financial issues	5

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.