Evidence for efficacy and effectiveness
DBT places a strong emphasis on empirical data to establish both the efficacy of the treatment and its effectiveness in clinical practice. Although a focus on empirical data does not differentiate DBT from most other cognitive-behavioural treatments, it is a defining characteristic of the treatment. This chapter summarizes briefly the current evidence for the efficacy of DBT and outlines principles for clinicians to consider in the evaluation of effectiveness in clinical practice.
Efficacy of DBT
Linehan developed DBT to treat adult women with a diagnosis of BPD and a recent history of parasuicidal behaviour. The initial randomized trial compared one year of DBT to treatment-as-usual (TAU) in the community (Linehan et al., 1991; Linehan et al., 1994). After one year, recipients of DBT had significantly fewer parasuicidal acts, less medically severe parasuicides, fewer psychiatric inpatient days, lower anger, higher social and global functioning, and higher treatment-retention rates (DBT = 83% versus TAU = 42%). Although all participants showed improvements in depression and suicidal ideation, the changes in the two groups on these variables were equivalent. At one-year follow-up, treatment gains were generally maintained, if less marked (Linehan et al., 1995).
Subsequent to this initial study, six RCTs have been published using comprehensive DBT to treat suicide attempts and non-suicidal self-injury in the context of BPD (Carter et al., 2010; Clarkin et al., 2007; Koons et al., 2001; Linehan et al., 2006; McMain et al., 2009; Verheul et al., 2003) and a further three studies have applied comprehensive DBT to the treatment of clients with personality disorder traits combined with self-injurious behaviours (Mehlum et al., 2014; Pistorello et al., 2012; Priebe et al., 2012). In summary, in these subsequent studies participants in the DBT arm of the trials continued to demonstrate reductions in the frequency of suicide attempts and non-suicidal self-injury. In some of the more recent studies with rigorous comparator treatments and adherence standards, significant differences between DBT and the comparator treatment were not found on all of the suicidal behaviour outcomes (Linehan et al., 2006; McMain et al., 2009). The reasons for these findings are unclear. Miga et al. (submitted) suggest that clinician expertise in treating suicidal behaviour and clients with multiple problems across conditions may be a factor in attenuating the outcomes in DBT. A recent review from the Cochrane Collaboration concluded that DBT was the only comprehensive treatment for BPD with sufficient data to meta-analyze and that data from the analysis indicated that DBT was a helpful treatment for individuals with BPD with its effects including a reduction in inappropriate anger, self-harm and an improvement in general functioning (Stoffers et al., 2012).
In addition to these efficacy studies with the population for whom Linehan originally developed the treatment, there have been several studies examining the efficacy of adaptations of DBT to other specific client populations or settings. The adult, outpatient populations include clients who have comorbidity for BPD and either substance abuse or dependence (Linehan et al., 1999; Linehan et al., 2002), and clients diagnosed with a mood disorder (Lynch et al., 2003; Lynch et al., 2007). Other populations include inpatients who meet criteria for PTSD (Bohus et al., 2013) and adolescents with personality disorder traits and a recent history of suicide attempts or non-suicidal self-injurious behaviour (Mehlum et al., 2014). All of these studies demonstrated significant clinical reductions in the targeted focus of the intervention. Caution is warranted, however, as with the exception of the studies in substance dependence and with adolescents, the comparator conditions were not robustly controlled. Furthermore, testing of the adaptation has been limited to a single trial in most cases. Thus these trials indicate areas of promise rather than proof for the adaptations of DBT.
Given the comprehensive nature of DBT and the associated expense of its use, there has been much interest in whether a less comprehensive version of the treatment may be equally efficacious. A recently published dismantling study addressed this question (Linehan et al., 2015). The study endeavoured to evaluate the importance of the Skills Training Component of the treatment as this element is perhaps the most frequently used as a stand-alone intervention although evidence for the efficacy of this is sparse. The trial compared comprehensive DBT with DBT Skills Training plus case management and with DBT individual therapy with an activity-based class. All conditions had their suicidal crises managed by clinicians trained with the DBT Suicide Crisis Protocol. All treatment conditions resulted in similar improvements in the frequency and severity of suicide attempts, suicide ideation, use of crisis services due to suicidality and reasons for living. Intervention arms that included DBT Skills Training had better outcomes for non-suicidal self-injury and for depression. There were some indications that comprehensive DBT may have had some advantages in the follow-up year in terms of reduction in suicide attempts, hospital admissions and emergency department visits. These differences did not reach statistical significance. This study provides further evidence of the benefit of DBT Skills Training although as both arms of this trial that included skills training also had individual work and DBT crisis management as well it would be premature to conclude that interventions with skills training alone are sufficient for treating suicidal individuals with a BPD diagnosis.