Clinicians working with adolescents face several challenging situations daily. For example, treating young people with eating disorders is complex because of the interplay among physical, psychological, and social constructs that can help precipitate and perpetuate eating disorders. Practitioners must keep in mind that eating disorders have the highest mortality of any of the psychiatric diagnoses, and death can result from medical complications or from suicide. Adding to the complexities of adolescent treatment, often adolescents will present with more than one complaint or problematic behavior. One common combination is adolescents presenting with an eating disorder and other maladaptive behaviors, including nonsuicidal self-injury (NSSI).
A significant number of young people with an eating disorder also self-injure. Adolescents who present with a history of NSSI can create anxiety in even the most seasoned clinicians for a variety of reasons. These can include clinicians’ own countertransference to self-inflicted pain and awareness of adolescents’ resistance to change a maladaptive coping strategy that has provided a transient solution to a personal problem. A family’s reaction to the self-injury, as well as reactions from other professionals like teachers and physicians, can be challenging and may cause additional distress in adolescents, which may lead to increased frequency of self-injury.
There is a sense that NSSI has increased—or at least the identification of it has increased— over the past several years. This increase in NSSI led to a proposal for the DSM-5 to include a separate diagnosis of self-harm. A summary of the proposed criteria was as follows:
Engaging in self-injurious behavior at least on five days without suicidal intent. The self-injurious behavior is associated with negative feelings of thoughts, preoccupation with the behavior before it occurred, frequent urges to self-injure, and the behavior is engaged with a purpose which could be a relief from a negative feeling or a thought or to gain interpersonal responses or a feeling (Gilman, 2012).
The proposed DSM-5 criteria looked at behavior and the motivation behind the behavior in order to classify it as self-injurious behavior. Ultimately the diagnosis of self-harm was not included in the DSM-5, but the fact that experts in the field considered self-injurious behavior as a stand-alone diagnosis demonstrates the significance and frequency of occurrence. Even without a formal diagnosis in DSM 5, self-injury behavior requires treatment.
What is Nonsuicidal Self-Injury?
NSSI is defined as “direct and deliberate destruction of body tissue in the absence of any observable intent to die” (Nock, 2010). Clinicians are seeing an increased number of patients who report at least one event of NSSI. A recent study from the Victoria Health Youth Survey, a population survey of residents of Victoria British Columbia ages fourteen to twenty-one years old, showed that over 16 percent of the youth had self-harmed (Klonsky, 2011).
The mean age of onset of NSSI was 15.2 years old and only 56 percent of the people ever sought help for NSSI (Nixon, Cloutier, & Jansson, 2008). Another study of high school students found that 23 percent of them had self-harmed in the past. In this study the average age of onset was younger—at fourteen years old—and the majority did not seek help for NSSI (Muehlenkamp & Gutierrez, 2007). In early adolescents, girls self-harm more frequently than males—five to six times more often, in fact—but the sex difference levels off in later adolescence (Hawton et al., 2012).
Other research has looked at the incident of borderline personality disorder (BPD) in adolescents in a community setting ranging from 0.9–1.3 percent, with risk factors for development being abuse and maternal inconsistency. The researchers also note that at this age there is more flexibility in the symptoms and that this may be a good time to intervene (Chanen, McCutcheon, Jovev, Jackson, & McGorry, 2007). There may be an association of some of the same factors that leads to the development of an eating disorder and BPD; one common link may be attachment style. Attachment behaviors are ways that people cope when they feel distress, particularly in regards to relationships. Typically when people feel interpersonal distress it causes them to reach out to individuals who provide comfort. In childhood these attachment figures are typically parents, and as people get older this shifts to other family members, friends, and romantic partners (Zachrisson & Skårderud, 2010).
There is some research showing that children with an insecure attachment are more likely to develop BPD and eating disorders. One study showed that adults who displayed an insecure attachment style and had high levels of early separation anxiety were associated with more body dissatisfaction, and with developing anorexia and bulimia (Troisi et al., 2006). It also found that patients with an insecure attachment style might be more influenced by the mainstream and the anorexia sub-culture than a person with a secure attachment style.
There are three main types of insecure attachment:
- Anxious/Resistant: Characterized by attempts to perform a significant number of proximity-seeking behaviors
- Avoidant: Characterized by attempts to be self-reliant and not express needs or wants to others
- Disorganized: Is usually a mixture of the two above styles that comes from caregivers providing inconsistent and unreliable care in early childhood
It has been shown that those patients with a resistant/anxious insecure attachment style are more likely to have higher levels of neuroticism on personality inventories, as well as higher rates of disordered eating (Eggert, Levendosky, & Klump, 2007). Interventions that help decrease insecure attachment organization may help lower symptomology in eating disorder pathology and NSSI.
Risk Factors for NSSI
A study looking at risk factors for youth that self-injure found that risk factors included female gender, depressive symptoms for females, feelings of hopelessness for males, running away from home for both sexes, and maladaptive dieting for females (Taliaferro, Muehlenkamp, Borowksy, McMorris, & Kugler, 2012). Another study in Australia looked at youth between the ages of fourteen to nineteen years old and found that 8 percent had self-injured in the past. This study found that the presence of anxiety, depression, alcohol misuse, cannabis use, tobacco use and antisocial behavior were associated with self-injury (Moran et al., 2011). Other risk factors include adolescents having a negative cognitive style and parents having a past or current depressive episode. Having multiple stressful life events, more depressive symptoms, less supportive relationships, and more negative interactions also predicted more likelihood of self-injury (Hankin & Abela, 2011). Nock has found that NSSI was associated with a history of abuse, poor verbal skills, poor problem-solving skills, poor distress tolerance, genetic predisposition for high emotional and cognitive reactivity, and identification with Goth subculture (2010). Over 90 percent of adolescents that self-injure meet criteria for a psychiatric diagnosis with the most common being depression, substance abuse and dependence, conduct disorder, antisocial personality disorder, phobias, and multiple diagnoses (Ougrin et al., 2012).
Why Teenagers Self Injure
Matthew Nock has proposed four key reasons for why adolescents self-injure. One reason is to regulate affect. Some people will self-injure to get rid of an overwhelming negative feeling, while others self-injure because they feel numb and self-injury will help them feel something. Nock also proposes that self-injury can function interpersonally (2010). It can let others know how distressed someone is feeling and signal for help, or it can get them out of social roles.
Adolescents often engage in eating disorder behaviors for some of the same reasons that they use NSSI. Most often these behaviors are tools to help regulate their emotions.
Eating Disorders and NSSI
In adolescents with eating disorders the rates of NSSI are even higher, with one study showing that over 40 percent of youth on an inpatient eating disorder unit had self-injured (Peebles, Wilson, & Lock, 2011). The rates of self-injury tend to peak in adolescence and typically decrease in adults, which is estimated to be between 2 and 5 percent of the population. In patients with eating disorders, their rates of NSSI remain high: 12 to 46 percent of adult patients with an eating disorder self-injure (Peebles et al., 2011). Self-injury appears in patients with eating disorders as a maladaptive coping mechanism, but does not taper off as it does in the general psychiatric population. Self-injurious behaviors in adolescents with eating disorders are more likely to occur when patients have bulimia, anorexia nervosa, binge and purge subtype, and/or impulsive behaviors (Peebles et al., 2011).
Risk factors for developing NSSI include poor distress tolerance, poor problem-solving skills, depression, and anxiety, which are also several of the risk factors for developing an eating disorder (Peebles et al., 2011). In patients with eating disorders, risk factors for self-injury include the aforementioned factors as well as binging and purging, using more than one method to purge, and those that have the most extensive treatment histories both at the inpatient and outpatient level of care (Peebles et al., 2011). In one study of 376 women on an eating disorder inpatient unit, results showed that 34.6 percent of the women had engaged in NSSI in their lifetimes and 21.3 percent had engaged in NSSI in the past six months. Compared to patients with eating disorders who did not engage in NSSI, those with NSSI had higher rates of trauma, higher rates of dissociation, more obsessive thoughts and compulsive behaviors, and higher rates of impulsivity (Paul, Schroeter, Dahme, & Nutzinger, 2002).
Increased Risk of Suicide
The Treatment of Resistant Depression in Adolescents (TORDIA) found that youth that self-harm are at greater risk of attempting suicide compared to those youth that have previously attempted suicide, and a history of NSSI at baseline was a stronger predictor of a suicide attempt than a history of suicide attempts during the twenty-four-week study period. Similarly, an adolescent depression antidepressant and psychotherapy trial also showed that a history of NSSI predicted a suicide attempt during the twenty-eight-week study; however, a history of prior suicide attempts did not (Asarnow et al., 2011).
Another study showed that people that engage in NSSI have a thirty-fold increased risk for completing suicide (Hankin & Abela, 2011). One team of researchers conducted clinical interviews on an adolescent inpatient unit and found that 70 percent of the adolescents that had engaged in NSSI had at least one previous suicide attempt (Boxer, 2010). As suicide is the third leading cause of death among adolescents, preventing NSSI may be a way to prevent suicide.
It is also known that people with eating disorders are at an increased risk of dying by suicide. In patients with anorexia, it has been thought that because the patients are physically frail, less lethal means of suicide may result in death because of their weakened bodies. However, a study looked at this and found those with anorexia that died by suicide did so by means that also would have been lethal for physically healthy individuals, such as jumping in front of a train (Holm-Denoma et al., 2008).
One way to make sense of the link between NSSI, eating disorders and suicide is the interpersonal theory of suicide. Thomas Joiner notes that this theory has three components necessary for people to attempt suicide, including:
- Feeling like a burden
- Not feeling like they belong
- A habituation to physically painful and/or fear inducing situations
It is thought that part of the increased risk of death from suicide in anorexia nervosa is due to increased pain tolerance that comes with the pain of self-starvation. It may be that the combination of self-starvation and NSSI increases the capacity to tolerate physical pain and lessens the natural fear of death that is evolutionarily valuable (Van Orden et al., 2010).
Treatment of NSSI in Patients with Eating Disorders
There are practice parameters available from the American Psychiatric Association and American Academy of Child and Adolescent Psychiatry on the treatment of eating disorders, but no such guidelines exist for the treatment of NSSI. Clinicians will often see that when people have both an eating disorder and NSSI, when the eating disorder becomes under better control the self-injury behavior increases. One explanation for this could be that both are maladaptive coping skills and as the treatment team takes away one form of coping, patients begin to increase the use of the other preferred coping skill. It is clear that if patients have both an eating disorder and self-injury, both need to be addressed in treatment.
In order to address self-injury, it is recommended that clinicians screen all of their patients for self-injury. Just like we should be mindful of our responses to suicidal ideations, we should maintain our own affect and maintain a curious attitude to elicit more information. Often when patients disclose self-injury to a nonprofessional, the nonprofessional exhibits an intense emotional reaction that could elicit shame in patients and make them less likely to disclose more information. Or, on the opposite end of the spectrum, the reaction may become reinforcing for patients in terms of the amount of attention received as a result of the behavior.
With any behavior that we want to modify, we need to fully understand it by doing a behavioral analysis, looking at what factors, feelings, and thoughts led up to the self-injury and any consequences from the self-injury. Oftentimes parents or loved ones will assume that patients are self-injuring for attention. Although it is possible that patients are seeking attention, a more common reason that people self-injure is that they are experiencing intense emotions and self-injury provides emotional regulation. Alternately, people are not feeling any emotion or are dissociating and self-injury allows them to feel something.
In the short-term, treatment can introduce replacement behaviors, which seek to substitute patients’ thoughts and urges to self-injure with more adaptive behaviors, activities, and skills (Walsh, 2012). Replacement skills can be thought of on a continuum: initially people may need to hold ice or pull a rubber band on their wrist, which will cause pain, but not tissue damage. But these types of skills are not the most socially accepted skills, and over time patients might move to journaling, walking their dogs or seeking support from loved ones.
Dialectical Behavior Therapy
Finally, there are evidence-based treatments that have been found to decrease self-injury in adolescents and adults, and may be helpful for those patients that have both an eating disorder and self-injure. Dialectical behavior therapy (DBT) improves emotional regulation, distress tolerance, and interpersonal effectiveness skills; in all studies, DBT was found to significantly decrease NSSI in adolescents (Nock, Teper, & Hollander, 2007). DBT looks at NSSI and eating disorder behaviors as ineffective problem-solving behaviors when emotional pain becomes unbearable. DBT is designed to help adolescents create a life worth living, help them problem solve, and manage their emotions and distress in a more adaptive manner. DBT is a mix between cognitive behavioral therapy and Eastern practices, primarily including mindfulness. The dialectical philosophy helps to highlight the multiple tensions involved in working with an adolescent with NSSI and eating disorders. The primary dialectic in DBT is acceptance and change, hence the adolescents are accepted exactly where they are, but also need to change (Miller & Smith, 2008).
In DBT there are several treatment targets. The target behaviors of DBT are those that are life threatening, such as suicidal ideations and behaviors, NSSI, and eating disorder behaviors that could result in death. Other targets of DBT are therapy-interfering behaviors, such as treatment nonadherence or missing appointments; quality of life-interfering behaviors, such as eating disorder behaviors, substance misuse, school truancy and other psychiatric conditions; and finally, increasing skills.
A comprehensive DBT program for adolescents and their families will typically include: a skills group for adolescents where they learn the skills; a family skills group where the families learn the skills to help reinforce the adolescents’ skills, which will help to generalize the skills; and individual DBT sessions with a therapist where adolescents can discuss the use of skills and get feedback. DBT therapists are also usually available for phone coaching, in which adolescents may call when they are stuck and receive brief coaching on which skills to try in the situation.
There are five functions of a comprehensive DBT program:
- Increase adolescents’ motivation to change
- Increase the number and effectiveness of the behavioral skills that the adolescents and their families have available to them
- Help adolescents generalize their skills in all environments
- Help structure the environment in a way that supports recovery
- Provide the skills necessary to families to support the adolescents in recovery (Miller & Smith, 2008)
The core of DBT is to help adolescents create a life worth living that takes maladaptive coping skills off of the table, such as suicidal behaviors, self-harm, and eating disorder behaviors. There are four core modules in DBT: mindfulness, which helps with self-regulation; interpersonal effectiveness, which helps adolescents develop and maintain healthy relationships; emotional regulation, which helps adolescents set themselves up to have good days emotionally; and distress tolerance, which helps adolescents learn coping skills to deal with stressful events when problem solving is not an option (Miller & Smith, 2008).
Another promising treatment for adolescents who self-injure and have eating disorders is mentalization-based treatment (MBT). MBT is a psychodynamic therapy that is hypothesized to work by increasing awareness of patients’ own minds, and recognizing that others have minds of their own that cause them to act. This awareness helps increase emotional regulation and relationships. Mentalizing is thinking about people’s own internal mind state and behaviors and being curious about other people’s behaviors. It is also knowing that others’ minds are connected to internal thoughts and feelings that we cannot know without asking about another’s mind state.
Mentalizing develops in childhood in the context of attachment relationships. Children learn to mentalize others by being mentalized by their loved ones. Children will cry out and good caregivers will notice the signal, mark it with an appropriate response, and then work to fix the difficulties. This process leads to a secure attachment pattern and establishes the beginnings of the children’s ability to mentalize, in which they will notice emotional changes in caregivers or peers and practice marking those changes and having an attuned response. Later children will learn about their own complex emotional states that others cannot just notice about them, but rather must ask about, and they will learn to ask others about their mental states. If as infants or children their signals were not met by caring, nonthreatening adults that had appropriate contingent responses, the ability to think about mental states and give a narrative to relationships is disrupted, which can have long lasting effects.
There are two main types of mentalizing errors. Hypomentalizing presents itself in people who are not curious about their own mental state or those in others, commonly seen in conditions like autism spectrum disorders. Then there is hypermentalizing, where people make assumptions about another person’s intent or mental state but does not validate them, living as if the presumption is reality. This not only leads to difficulties in interpersonal relationships, but people who hypermentalize also often have a difficulty regulating their own emotions—which can lead to maladaptive coping techniques such as self-harm and eating disorder behaviors.
The aim of MBT is to reinstate the ability to mentalize when it is lost. In people that hypermentalize, this typically occurs when they are suffering from conditions that have elevated emotions. Uncontrolled emotions initially lead to impulsive behaviors that provide patients with a feeling of short-term control over eating disorder behaviors and self-injury. In reinstating the ability to mentalize, they will gain better control of their emotions. The first step of MBT is to help patients achieve better control of the expression of their emotions, because without the ability to control their emotions they cannot begin to consider their internal states.
The primary intervention in MBT is the therapeutic relationship and giving patients a different relational experience. The therapeutic stance in MBT is one of curiosity and “not knowing,” taking time to mark differences in perspective and legitimizing different perspectives. It also involves being very curious with patients about their experiences, and asking for descriptions of their experiences (“What are you feeling?”) rather than explanations (“Why are you feeling that?”). Eventually patients will get better at recognizing when their mentalizing abilities get off line and learn how to reestablish their mentalizing capacities. The reestablishment of mentalization allows patients to regulate their thoughts and feelings which allows them to self-regulate and build healthier relationships (Bateman & Fonagy, 2010).
A recent study compared high quality treatment as usual to MBT for adolescents. MBT was studied in 80 adolescents that had depression and engaged in self-harm, and results showed that MBT was better than treatment as usual in terms of decreasing self-injury and depression scores (Rossouw & Fonagy, 2012).
It is a common occurrence to have adolescents with both an eating disorder and NSSI. When doing an assessment, clinicians should always ask about NSSI. If patients answer that they do self-injure, remain respectfully curious and try to get as much information about when, how, and the function of the self-injury. Remember that there is an increased risk of suicide in both eating disorders and NSSI in adolescents, so take a careful suicide assessment.
Most adolescents use the eating disorder and NSSI to serve core functions of emotional regulation, so both the eating disorder and self-injury must be addressed at the same time. Otherwise, one will get better while the other gets worse as they continue to struggle to regulate their emotions. With one of their main coping mechanisms being limited, they will rely on the other form more. It may even be beneficial to go over some replacement behaviors with patients immediately to help give them some different coping skills to try. Finally, there are evidence-based treatments for NSSI, which include DBT and mentalization-based treatment.
Asarnow, J. R., Porta, G., Spirito, A., Emslie, G., Clarke, G., Wagner, K. D., . . . Brent, D. A. (2011). Suicide attempts and nonsuicidal self-injury in the treatment of resistant depression in adolescents: Findings from the TORDIA trial. Journal of American Academy of Child and Adolescent Psychiatry, 50(8), 772–81.
Bateman, A., & Fonagy, P. (2010). Mentalization-based treatment for borderline personality disorder. World Psychiatry, 9(1), 11–5.
Boxer, P. (2010). Variations in risk and treatment factors among adolescents engaging in different types of deliberate self-harm in an inpatient sample. Journal of Clinical Child & Adolescent Psychology, 39(4), 470–80.
Chanen, A. M., McCutcheon, L. K., Jovev, M., Jackson, H. J., & McGorry, P. D. (2007). Prevention and early intervention for borderline personality disorder. Medical Journal of Australia, 187(7), S18–S21.
Eggert, J., Levendosky, A., & Klump, K. (2007). Relationships among attachment styles, personality characteristics, and disordered eating. International Journal of Eating Disorders, 40(2), 149–55.
Gilman, S. L. (2012). How new is self-harm? The Journal of Nervous and Mental Disease, 200(12), 1008–16.
Hankin, B. L., & Abela, J. R. Z. (2011). Nonsuicidal self-injury in adolescence: Prospective rates and risk factors in a 2½ year longitudinal study. Psychiatry Research, 186(1), 65–70.
Hawton, K., Bergen, H., Kapur, N., Cooper, J., Steeg, S., Ness, J., & Waters, K. (2012). Repetition of self-harm and suicide following self-harm in children and adolescents: Findings from the Multicentre Study of Self-harm in England. Journal of Child Psychology and Psychiatry, 53(12), 1212–9.
Holm-Denoma, J. M., Witte, T. K., Gordon, K. H., Herzog, D. B., Franko, D. L., Fichter, M., . . . Joiner, T. E. Jr. (2008). Deaths by suicide among individuals with anorexia as arbiters between competing explanations of the anorexia-suicide link. Journal of Affective Disorders, 107(1–3), 231–6.
Klonsky, E. D. (2011). Nonsuicidal self-injury in United States adults: Prevalence, sociodemographics, topography, and functions. Psychological Medicine, 41(9), 1981–6.
Miller, A. L., & Smith, H. L. (2008). Adolescent nonsuicidal self-injurious behavior: The latest epidemic to assess and treat. Applied and Preventive Psychology, 12(4), 178–88.
Moran, P., Coffey, C., Romaniuk, H., Olssosn, C., Borschmann, R., Carlin, J. B., Patton, G. C. (2011). The natural history of self-harm from adolescence to young adulthood: A population-based cohort study. The Lancet, 379(9812), 236–43.
Muehlenkamp, J. J., & Gutierrez, P. M. (2007). Risk for suicide attempts among adolescents who engage in non-suicidal self-injury. Archives of Suicide Research, 11(1), 69–82.
Nixon, M. K., Cloutier, P., & Jansson, S. M. (2008). Nonsuicidal self-harm in youth: A population-based survey. Canadian Medical Association Journal, 178(3), 306–12.
Nock, M. K., Teper, R., Hollander, M. (2007). Psychological treatment of self-injury among adolescents. Journal of Clinical Psychology, 63(11), 1081–9.
Nock, M. K. (2010). Self-injury. Annual Review of Clinical Psychology, 6, 339–63.
Ougrin, D., Zundel, T., Kyriakopoulos, M., Banarsee, R., Stahl, D., & Taylor, E. (2012). Adolescents with suicidal and nonsuicidal self-harm: Clinical characterisitcs and response to therapeutic assessment. Psychologic Assessment, 24(1), 11–20.
Paul, T., Schroeter, K., Dahme, B., Nutzinger, D. O. (2002). Self-injurious behavior in women with eating disorders. American Journal of Psychiatry, 159(3), 408–11.
Peebles, R., Wilson, J. L., & Lock, J. D. (2011). Self-injury in adolescents with eating disorders: Correlates and provider bias. Journal of Adolescent Health, 48(3), 310–3.
Rossouw, T. I., & Fonagy, P. (2012). Mentalization-based treatment for self-harm in adolescents: A randomized controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 51(12), 1304–13.
Taliaferro, L. A., Muehlenkamp, J. J., Borowksy, I. W., McMorris, B. J., & Kugler, K. C. (2012). Factors distinguishing youth who report self-injurious behavior: A population-based sample. Academic Pediatrics, 12(3), 205–13.
Troisi, A., Di Lorenzo, G., Alcini, S., Nanni, R. C., Di Pasquale, C., & Siracusano, A. (2006). Body dissatisfaction in women with eating disorders: Relationship to early separation anxiety and insecure attachment. Psychosomatic Medicine, 68(3), 449–53.
Van Orden, K. A., Witte, T. K., Cukrowicz, K. C., Braithwaite, S., Selby, E. A., & Joiner, T. E. Jr. (2010). The interpersonal theory of suicide. Psychology Review, 117(2), 575–600.
Walsh, B. W. (2012). Treating self-injury: A practical guide (2nd ed.). New York, NY: The Guilford Press.
Zachrisson, H. D., & Skårderud, F. (2010). Feelings of insecurity: Review of attachment and eating disorders. European Eating Disorders Review, 18(2), 97–106.