Evidence-based review of psychotherapeutic treatments for NSSI
The literature search described above was repeated using the terms “self-injury” or “self-harm” combined with the terms “treatment” or “therapy” in PsychINFO®, PubMed, and ClinicalTrials.gov databases. Results of this refined search indicate that despite an increased interest in NSSI in the literature, few psychotherapeutic treatments have been designed and evaluated specifically for NSSI [
11]. Of grave concern is that no treatments have been designed and evaluated specifically for NSSI among adolescents. The dearth of interventions for NSSI among adolescents may be due to the relatively recent interest and recognition of the problem of NSSI among this age group [
12], and may improve with the adoption of NSSI as a psychiatric disorder in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorder [
13].
The lack of empirically supported treatments for NSSI, however, presents a dire situation for the clinician who is left to treat youth with NSSI without reference to evidence-based strategies. Guidance on how to treat adolescents presenting with NSSI may be obtained from studies of adults with NSSI, as well as studies of adolescents and adults with related conditions or disorders. For example, a handful of studies of have evaluated psychosocial interventions for
deliberate self-harm (DSH). DSH typically refers to self-injury that can be suicidal and/or non-suicidal [
1].
Cognitive and behavioral therapies (CBT) show the most promise in treating NSSI across various settings [
14]. A form of CBT, Problem-Solving Therapy [
15], was one of the first treatments for DSH to be evaluated using randomized controlled trials. Problem-Solving Therapy involves training in the skills and attitudes necessary to promote active problem solving [
16]. Treatment with Problem-Solving Therapy focuses on accomplishing the following goals: (1) Developing or enhancing a positive problem orientation and decreasing a negative orientation; (2) Training in rational problem solving (i.e., defining and formulating the problem, generating alternative solutions, making a decision, and solution verification); and (3) reducing avoidance of problem solving, as well impulsive and careless decision making [
17]. Within this model, NSSI is conceptualized as a dysfunctional solution to problems, with improved problem solving attitudes and skills leading to decreased reliance on NSSI to cope.
Evaluations of Problem-Solving Therapy with patients with DSH suggest promise as a treatment, but with limitations. An early meta-analysis found a trend toward reduction of DSH with therapies focused on problem solving, but when compared to control conditions, the difference was not statistically significant [
18]. A later meta-analysis of six randomized controlled studies, four of which included at least some older adolescents (15–17 years old), found Problem-Solving Therapy to be superior to control conditions in reducing depression and hopelessness, and in improving problem solving [
19]. Unfortunately, this meta-analysis did not directly examine the effects of these treatments on reduction of actual DSH. A recent study of group-based Problem-Solving Therapy for adult females who engaged in self-poisoning also found preliminary evidence for improvement with depression, hopelessness, suicidal ideation, and social problem solving, but also failed to show a significant difference between the control and treatment group; indeed, neither group evidenced DSH during the two-month follow-up [
20].
The lack of consistent results of Problem-Solving Treatments for DSH has led some to argue that treatment must integrate strategies beyond problem solving skills and attitudes to be effective in treating DSH [
14]. Manual-Assisted Cognitive-behavioral Therapy (MACT) for DSH was developed as just such a treatment. MACT is a brief therapy for DSH that integrates problem-solving therapy with cognitive techniques and relapse prevention strategies. An early pilot study of MACT, which included some adolescents, found a lower rate of DSH for MACT when compared to treatment as usual [
21]. Consistent with several of the other problem-solving therapies, however, the difference was not statistically significant. A multi-site randomized controlled study of MACT was subsequently conducted with 480 people, including some adolescents (ages 16 and 17). Although the results supported the cost-effectiveness of MACT over treatment as usual, no significant effect was found on repeated DSH [
22]. Further analyses indicated that for participants with borderline personality disorder, MACT was associated with
increased costs when compared to treatment as usual [
23]. A newly developed version of MACT, the “Cutting Down” program, has recently been developed and piloted specifically for adolescents with DSH [
24]. Although findings from this single-group open trial study of 24 adolescents suggest promise in reducing DSH, these results need to be tested under more rigorous conditions (e.g., control or treatment comparison group, randomized assignment, larger sample), especially given the disappointing findings of prior MACT treatments when evaluated in randomized controlled trials [
22,
23].
A group therapy for DSH, Developmental Group Psychotherapy, has also been extensively evaluated with adolescents. This therapeutic approach combines problem-solving skills training with aspects of Dialectical Behavior Therapy (described below) and psychodynamic therapy. An initial evaluation of this treatment found a reduction in repeated DSH when compared to treatment as usual [
25]. A replication of this treatment, however, failed to find improvement in DSH over treatment as usual [
26]. An additional large-scale (n = 366 adolescents) replication of the Developmental Group Psychotherapy treatment also failed to show the superiority of the treatment over treatment as usual, nor was it cost-effective over treatment as usual [
27].
The Treatment of SSRI-Resistant Adolescent Depression (TORDIA) study is the only study that we found that evaluated NSSI separately from suicidal self-injury as a treatment outcome [
28]. The TORDIA study included adolescents, ages 12–18, who had a diagnosis of major depressive disorder but did not respond to a selective-serotonin uptake inhibitor (SSRI). Treatment arms included antidepressant medication (venlafaxine or a different SSRI), with or without CBT. The CBT arm included cognitive restructuring and behavior activation components, skills training in emotion regulation, social skills, and problem-solving, as well as parent–child sessions to improve support, decrease criticism, and improve family communication and problem-solving. Approximately one-third of the sample also had a history of NSSI [
28]. As such, this represents a unique study in that it is the only treatment study for adolescents that did not collapse NSSI and suicidal self-injury into DSH.
Overall results of the TORDIA study at 12 weeks of treatment indicated that a combination of CBT with medication (either venlafaxine or a different SSRI) was more efficacious in reducing depressive symptoms than just switching to a different medication [
28]. Problem-solving and social skills appeared to be the most effective components of the CBT intervention in this study [
29]. The superiority of CBT and medication over medication alone, however, was not sustained at the 24-week follow-up [
30]. Further, there were no differences in the rate of NSSI events across the various treatments, including medication and CBT [
31]. The findings of this study suggest that treatments that may be effective for a condition related to NSSI may not adequately address NSSI [
31].
More promising findings are found in a study examining the efficacy of a 12-session CBT intervention for DSH [
32]. This study included 82 individuals who engaged in DSH, including adolescents (age 15–17) and adults, randomized to either an adjunctive CBT intervention or to treatment as usual. In contrast to the TORDIA study, this adjunctive CBT intervention was developed to specifically identify and modify the mechanisms that maintain DSH. Specifically, this CBT treatment directly assessed the most recent episode of DSH, examined emotional, cognitive, and behavioral contributions to the maintenance of DSH, and addressed these maintaining factors using cognitive and behavioral strategies. This focused, adjunctive CBT intervention for DSH was found to be superior to treatment as usual in reducing episodes of DSH at the 9-month follow-up. The authors suggest that CBT was effective in reducing DSH because it actively targeted the depressive symptoms, suicidal cognitions, and problem-solving deficits that maintained DSH.
The form of CBT with that has the most evidence supporting a reduction in DSH is Dialectical Behavior Therapy (DBT). DBT combines skills-training, exposure and response prevention, contingency management, problem-solving training, and cognitive modification strategies with mindfulness, validation, and acceptance practices [
33]. It is important to note, however, that DBT was not designed to treat DSH, but instead was designed to treat borderline personality disorder, which often includes DSH. Randomized and non-randomized studies indicate that DBT is effective in adult patients with borderline personality disorder for a range of outcome variables, including DSH [
34‐
36].
DBT has been adapted for use with adolescents with numerous problem behaviors, including NSSI and suicidal self-injury [
37‐
40]. Studies have also examined the adaptation of DBT for incarcerated male [
41] and female [
42] adolescents, as well as for children [
43]. Despite over a decade of articles on DBT for adolescents, there have been no randomized control studies of DBT in adolescents [
44]. Indeed, a 2009 review [
45] found only three non-randomized studies of DBT with adolescents that included a comparison group [
42,
46,
47]. Available evidence from quasi-experimental and pre-post designs suggests that DBT for adolescents may be helpful in reducing hospitalization, suicidal ideation, and treatment dropout; however, support for reducing NSSI is limited [
38,
45]. For example, a feasibility study of DBT on an inpatient unit found that DSH decreased for the DBT group as well as for the treatment as usual group at follow-up [
46]. In summary, DBT is an effective form of treatment for NSSI and suicidal self-injury among adults with borderline personality disorder, and therefore holds great promise for treatment of NSSI among adolescents [
48]. Empirical support for the application of DBT to adolescents with NSSI, however, remains limited.
Other variations of CBT and non-CBT treatments for DSH have also been explored in the literature. For instance, multisystemic therapy has been evaluated as an alternative to hospitalization for youth engaging in DSH. Multisystemic therapy was originally developed as a treatment for antisocial youth [
49] and has been adapted for use with youth in emotional and behavioral crises [
50]. Multisystemic therapy is a family-based treatment that is grounded in a social-ecological model, focusing interventions on the multiple systems that maintain youths’ problem behavior [
49]. In a randomized trial of youth presenting in psychiatric crisis, multisystemic therapy demonstrated superiority to hospitalization in decreasing DSH, as rated by parents on the Child Behavior Checklist [
51]. Because the MST group had higher DSH at baseline than the hospitalization group, however, it wasn’t possible to rule out regression to the mean as an explanation for the treatment effect. Further, no treatment effect of MST was found for depressive affect, hopelessness, and suicidal ideation.
Other variations of treatments for NSSI and DSH have been evaluated with adults, but not adolescents. Emotional regulation group therapy [
52], a 14-week adjunctive therapy for NSSI uses strategies from DBT and Acceptance and Commitment therapy. This group treatment has been shown to reduce NSSI in adult women with subthreshold or threshold BPD [
53,
54], although more studies are needed to confirm the findings. Psychodynamic approaches, including interpersonal psychodynamic psychotherapy [
55], mentalization-based therapy [
56], object-relations psychodynamic psychotherapy [
57], and transference-focused psychotherapy [
58] have also been studied for adults with DSH. Interpersonal Therapy for Depressed Adolescents, an efficacious treatment for depressed adolescents [
59], has been adapted for use with adolescents with NSSI (ClinicalTrials.gov Identifier: NCT00401102), although results from the randomized controlled trial have yet to be published.
Another treatment currently under evaluation is the Treatment for Non-Suicidal Self-Injury in Young Adults (T-SIB; ClinicalTrials.gov Identifier: NCT01018433). The 9-session T-SIB intervention was designed specifically to treat NSSI among young adults, ages 18 to 29 years, and includes motivational enhancement pre-treatment strategies, functional analysis, and skill training for problem-solving, distress tolerance, cognitive distortions, and interpersonal skills. Although this study is ongoing and no findings have been published, preliminary results support the feasibility, acceptability, and efficacy of the time-limited T-SIB intervention for young adults who engage in NSSI [
60].
Finally, preliminary evidence suggests that exercise may be a promising treatment for addressing the urges to engage in NSSI behavior. Exercise or participation in sports has been reported as one of the most helpful strategies to resist urges to engage in NSSI [
61]. A single-case, quasi-experimental study of a young adult with a 13-year history of NSSI demonstrated that urge and frequency of NSSI significantly declined with the use of physical exercise [
62]. Further research is needed to understand the efficacy of exercise and physical activity among adolescents with NSSI.
In summary, little research has examined the efficacy of treatments designed specifically for adolescents with NSSI. Most of the available studies have focused on DSH instead of NSSI, making it difficult to understand what exactly the treatment is addressing. Further, many of the studies have examined adolescents along with adults; only a handful of studies have focused specifically on adolescents. While variations of CBT enjoy the greatest support in the literature, that support is not consistent when focusing on adolescents with NSSI.