Dialectical behavior therapy (DBT) is a structured form of psychotherapy that combines behavioral methods with acceptance, validation, and mindfulness. It was developed by psychologist Marsha M. Linehan and colleagues, initially for adults with recurrent suicidal behavior and borderline personality disorder. Its central therapeutic balance is that a person’s experiences and responses can be understandable in context while particular behaviors still require change.
Comprehensive DBT is a coordinated treatment program rather than a single technique or skills class. Its standard outpatient form includes individual therapy, group skills training, between-session coaching, and a consultation team for therapists. Treatment follows an explicit hierarchy that gives priority to life-threatening behavior, followed by behavior that interferes with treatment and problems that impair everyday life.
DBT has an established research base for selected outcomes in borderline personality disorder and for self-harming adolescents, with a more variable evidence base for other conditions and adaptations. Its development also influenced wider clinical practice by integrating systematic attention to suicide risk, practical skills acquisition, and the therapeutic relationship within a behavioral framework.
Origins and development
DBT emerged from behavioral treatment research at the University of Washington during the 1980s. Linehan’s research program addressed a population often excluded from psychotherapy trials: people with repeated suicidal behavior, severe emotional distress, and substantial difficulty remaining in treatment. Standard behavioral approaches emphasized identifying problems and changing behavior, but this emphasis could be experienced as criticism or as evidence that the therapist did not understand the person’s suffering. Approaches centered only on acceptance, conversely, could seem insufficiently responsive to the need for concrete change.
The resulting treatment combined cognitive behavioral therapy with deliberate acceptance strategies. Validation became a central clinical method rather than merely an introductory expression of empathy. Mindfulness practices, influenced partly by Zen traditions within Buddhism, were translated into skills taught in a secular clinical setting. The term “dialectical” described both a conceptual stance—holding apparently opposed truths together—and an active therapeutic method for moving beyond rigid either-or positions.
An early randomized controlled trial, published in 1991, compared DBT with treatment as usual in women with borderline personality disorder and chronic self-harming behavior. It reported advantages for DBT in outcomes including the frequency of self-harming acts, treatment retention, and psychiatric inpatient use. The study was small, and its comparison condition did not control all differences in treatment organization or intensity. Nevertheless, it helped establish that a highly structured outpatient treatment could be studied systematically in this population.
In 1993, Linehan published Cognitive-Behavioral Treatment of Borderline Personality Disorder and an accompanying skills-training manual. These works specified the treatment’s theoretical model, behavioral targets, intervention strategies, and organizational requirements. A substantially revised skills manual appeared in 2015, expanding the teaching materials and incorporating developments in clinical practice. The manuals distinguish a comprehensive treatment system from the use of individual DBT techniques.
Subsequent research included comparisons with expert community psychotherapy, studies of individual treatment components, and adaptations for adolescents and other clinical groups. DBT came to be grouped among “third-wave” cognitive and behavioral therapies because of its prominent use of mindfulness and acceptance. That classification does not replace its behavioral foundations: detailed analysis of behavior, reinforcement, exposure, problem solving, and repeated practice remain central. Its history is therefore both the development of a particular therapy and an example of how treatment design changed in response to difficulties encountered with a high-risk clinical population.
The biosocial model and dialectical framework
DBT’s principal explanatory framework is the biosocial model of emotion regulation. In its original formulation, severe and persistent emotional dysregulation develops through transactions between biological vulnerability and an invalidating environment. Emotional vulnerability can involve heightened sensitivity to emotional cues, intense responses, and a relatively slow return to baseline. The model does not require that every person show all these features, nor does it identify a single biological abnormality as the cause of borderline personality disorder.
An invalidating environment is one in which internal experiences are repeatedly dismissed, misinterpreted, punished, or responded to inconsistently. Examples include treating distress as deliberate misconduct or recognizing suffering only when it becomes extreme. In the model, such responses can interfere with learning how to identify emotions, communicate needs, and regulate arousal. The relationship is transactional rather than simply one-directional: a person’s intense responses can alter how others react, and those reactions can further intensify the person’s distress.
This account is a treatment-guiding model, not a finding that a particular family caused a disorder. Invalidating interactions can occur in families, schools, health services, and broader social settings. Biological vulnerability also encompasses more than inherited characteristics; development and experience affect emotional functioning. The model offers a way to formulate an individual case without treating either biology or environment as a complete explanation.
The dialectical framework organizes the therapist’s response to these difficulties. Acceptance and change are the best-known pair, but other tensions include independence and the need for support, consistency and flexibility, and the person’s immediate wishes versus longer-term goals. A dialectical intervention seeks a workable synthesis rather than choosing one side absolutely. For example, a therapist can acknowledge that withdrawing from an overwhelming interaction brought short-term relief while examining how repeated withdrawal damages relationships.
Validation means identifying what is understandable, accurate, or contextually coherent in an experience or response. It does not mean agreeing with every interpretation or endorsing harmful action. A therapist may validate fear and anger while questioning the conclusion that another person intended harm. Different levels of validation range from attentive listening to recognizing how a response makes sense given learning history or current circumstances.
Behavioral principles provide the complementary account of persistence and change. Self-injury, avoidance, or substance use may recur because they rapidly reduce distress or alter an interpersonal situation. This is a functional explanation, not an assumption that the person consciously planned those consequences. DBT therefore asks both why behavior makes sense in context and what new responses could perform its functions with fewer harmful consequences.
Organization of comprehensive treatment
Standard outpatient DBT is organized around four interacting delivery modes: individual therapy, group skills training, between-session coaching, and a therapist consultation team. These modes serve different functions. Individual sessions strengthen motivation and address personally significant behavioral targets; groups teach capabilities; coaching supports use of those capabilities in everyday situations; and consultation supports therapists’ effectiveness and adherence to the model.
Individual therapy commonly occurs once a week. The therapist coordinates the treatment plan, reviews monitoring records, and selects session targets according to a predetermined hierarchy. Sessions are not limited to discussing the week’s most emotionally salient event. A recent life-threatening act, a pattern of missed appointments, or a serious obstacle to housing may take precedence over other topics because of its place in that hierarchy. Treatment planning connects these immediate targets to the person’s own reasons for seeking a more sustainable life.
Skills groups are generally educational rather than open-ended process groups. Meetings often last about two hours or longer, commonly involve two facilitators, and include homework review, teaching, and rehearsal. The standard curriculum covers mindfulness, distress tolerance, emotion regulation, and interpersonal effectiveness. Programs frequently repeat the curriculum so that participants encounter skills more than once. A full course often lasts approximately a year, although research protocols and clinical programs differ in duration.
Between-session coaching is intended to help a person apply skills when the relevant situation is actually occurring. Traditionally delivered by telephone, it may use other agreed communication methods in contemporary services. Coaching is distinct from unrestricted access to a therapist and from a comprehensive emergency service. Programs specify availability, purposes, and boundaries. Its behavioral purpose is to make skillful help-seeking and real-world practice more accessible, rather than to make the therapy office the only setting in which difficulties can be managed.
The consultation team usually meets weekly. Team members discuss treatment problems, their own responses, and questions about applying DBT principles. The consultation structure reflects the model’s recognition that clinicians can become discouraged, rigid, or inconsistent when working with recurrent crises. It is not simply an administrative meeting, and it does not replace individual professional accountability or formal supervision where that is required.
These four modes support five broader functions: improving capabilities, strengthening motivation, generalizing learning, maintaining therapist motivation and competence, and structuring the environment. Distinguishing modes from functions matters when services adapt delivery. A change in format may preserve a function, but a skills class alone does not automatically reproduce the full treatment. Coordination with other mental health services, medication management, and crisis provision may also be necessary, without making those services distinctive components of DBT itself.
Target hierarchy, behavioral analysis, and treatment stages
DBT uses an explicit hierarchy to determine what receives attention first. Within the initial treatment stage, life-threatening behaviors have priority. These include suicide attempts and other behaviors posing serious danger. Next come therapy-interfering behaviors, such as repeated nonattendance, withholding essential information, or clinician responses that undermine collaboration. Quality-of-life-interfering behaviors follow, including problems involving substance use, unstable housing, severe avoidance, or destructive relationship patterns. Skills acquisition is pursued in relation to these targets rather than as an unrelated educational exercise.
The hierarchy distinguishes urgency from importance. Employment, relationships, and personal fulfillment can be central treatment goals, but an immediate threat to life takes precedence in a particular session. Therapy-interfering behavior is addressed early because an ineffective or collapsing treatment cannot reliably support other goals. The category applies to therapists as well as patients, which makes the treatment relationship itself a legitimate subject of analysis.
A principal assessment tool is the diary card, a brief repeated record of relevant experiences and actions. Depending on the treatment plan, it may track urges, self-harm, emotional intensity, substance use, and skills practice. Its purpose is not merely to generate a symptom score. It helps identify patterns, improves recall between sessions, and provides information for selecting targets. What is recorded can change as the person’s difficulties and goals change.
Behavioral chain analysis examines a specific episode in detail. The therapist and patient identify vulnerabilities, the prompting event, and successive links involving thoughts, emotions, bodily sensations, actions, and environmental responses. Consequences are examined over both short and longer time frames. For example, an argument might be followed by an interpretation of rejection, increasing physiological arousal, withdrawal from supportive contact, and a harmful act that briefly reduces distress. The analysis tests the sequence rather than assuming that one diagnosis or emotion explains the entire episode.
Solution analysis then identifies where the chain could change. Interventions can include rehearsing a communication skill, changing an environmental cue, testing an interpretation, using exposure to reduce avoidance, or arranging different consequences. Commitment strategies help connect these tasks to the person’s goals. Rather than treating motivation as a fixed prerequisite, DBT treats it as something that can fluctuate and can itself be addressed behaviorally.
The broader model describes stages beyond initial behavioral stabilization. Stage 2 concerns severe emotional suffering and, where relevant, traumatic experiences; stage 3 addresses ordinary life problems and personal goals; stage 4 concerns greater fulfillment and connectedness. These later stages are less standardized and less extensively researched than stage 1. They are not a rigid timetable or a requirement that every participant pass through an identical sequence. Trauma-focused work, in particular, depends on a specific clinical formulation and treatment protocol rather than simply the passage of time.
The four skills modules
DBT skills training translates broad therapeutic principles into named, practicable behaviors. Skills are taught through explanation, demonstration, rehearsal, homework, and review of obstacles. The curriculum is not intended to eliminate emotion or substitute positive thinking for painful realities. It aims to increase the range of responses available when existing patterns are ineffective.
Mindfulness provides a foundation across the curriculum. The “what” skills are observing, describing, and participating; the “how” skills are acting nonjudgmentally, one-mindfully, and effectively. These distinctions separate noticing an experience from labeling it and from engaging fully in an activity. “Wise mind” is a teaching concept for integrating emotional information with reasoned understanding. Mindfulness in DBT can involve brief attention exercises and ordinary activities rather than extended meditation alone.
Distress tolerance addresses situations in which intense discomfort cannot immediately be removed or in which attempts to escape it would create greater harm. Crisis-survival skills include pausing before acting, redirecting attention temporarily, self-soothing, and evaluating consequences. Other methods seek to alter high physiological arousal. Acceptance skills, including radical acceptance, concern acknowledging the facts of a situation without adding a continuing struggle against their existence. Acceptance is distinguished from approval, agreement, or abandoning efforts to change what remains changeable.
Emotion regulation focuses on identifying emotions, understanding their functions, reducing vulnerability, and changing responses when appropriate. The curriculum examines prompting events, interpretations, bodily changes, action urges, and consequences. “Check the facts” asks whether the emotional response fits available evidence. Opposite action involves acting against an emotion-driven urge when that response is not justified by the facts or would be ineffective. When an emotion accurately signals a solvable problem, problem solving may be more appropriate than attempting to reverse it. The module also addresses patterns of sleep, activity, and other everyday conditions that influence emotional vulnerability.
Interpersonal effectiveness distinguishes three goals that may conflict: obtaining an objective, maintaining a relationship, and maintaining self-respect. The mnemonic DEAR MAN organizes assertive requesting and refusal; GIVE emphasizes relationship-preserving behavior; FAST concerns self-respect. These frameworks do not promise that a request will succeed. They provide a way to communicate clearly and evaluate choices while recognizing differences in power, safety, and context.
Skills selection depends on the behavioral function of a problem. Temporary distraction can help prevent a dangerous action during an acute crisis, yet repeated distraction can reinforce avoidance when a problem requires engagement. Assertiveness may be useful in one relationship and insufficient in a coercive environment. DBT therefore combines skills instruction with case formulation and feedback. Knowing a skill’s name is not equivalent to using it reliably, and lack of improvement can reflect difficulties with selection, practice, context, or the treatment plan rather than a simple lack of effort.
Clinical evidence and its interpretation
DBT’s most developed evidence base concerns adults with borderline personality disorder, especially those with recurrent suicidal or self-injurious behavior. Trials have examined suicide attempts, nonsuicidal self-injury, hospital use, treatment retention, symptoms, and functioning. These outcomes are related but not interchangeable. A reduction in self-injury does not necessarily demonstrate a reduction in suicide deaths, and improved retention does not by itself establish improvement in every clinical domain.
A 2006 trial by Linehan and colleagues randomly assigned 101 women with borderline personality disorder and recent suicidal or self-injurious behavior to DBT or community treatment by experts. It reported advantages for DBT in suicide attempts and several measures of crisis-service use and retention. The active comparison strengthened the study beyond a simple comparison with loosely defined usual care. Nevertheless, its sample and treatment setting limit direct generalization to all people with the diagnosis.
Not all comparisons show DBT to be superior to another organized treatment. A 2009 trial involving 180 patients compared DBT with general psychiatric management, a structured approach combining disorder-specific case management and other clinical interventions. Both groups improved substantially, without significant between-group differences on the principal outcomes reported. Such findings indicate that DBT is one evidence-supported structured treatment, not that its unique superiority over all alternatives has been established.
A 2015 component trial involving 99 women compared standard DBT with versions organized around individual therapy or skills training, while retaining a suicide-risk assessment and management protocol across conditions. All conditions showed improvement in suicide-related outcomes. Conditions containing skills training had advantages on some measures, including nonsuicidal self-injury. Component studies help identify potentially important ingredients, but they do not establish that an isolated skills course is equivalent to comprehensive DBT across populations and outcomes.
Evidence for adolescents includes a 2018 trial of 173 participants at high risk for suicide, comparing adolescent DBT with individual and group supportive therapy. DBT showed advantages on suicide attempts, nonsuicidal self-injury, and total self-harm at the end of treatment; between-group differences weakened during follow-up. This distinction between immediate and sustained effects is important when interpreting treatment efficacy.
Systematic reviews generally find benefits for some self-harm and borderline symptom outcomes, while emphasizing variation in study quality and certainty. Psychotherapy trials are difficult to blind, comparator treatments vary, and samples can be too small to estimate uncommon outcomes such as suicide death reliably. Attrition, inconsistent outcome definitions, and differences in treatment fidelity further complicate synthesis. Intention-to-treat analysis and clearly specified clinical endpoints help reduce some interpretive problems, but cannot remove all uncertainty.
Evidence is therefore most accurately described by population, treatment format, comparator, outcome, and follow-up period. Reduced crisis behavior, improved health-related quality of life, and sustained social functioning are distinct achievements. A therapy can have evidence for one without having equally strong evidence for the others.
Adaptations for adolescents and other clinical populations
Adolescent DBT, often abbreviated DBT-A, modifies comprehensive treatment to address developmental needs and family relationships. Adaptations associated with the work of Alec Miller, Jill Rathus, and colleagues include caregiver participation, multifamily skills groups, family interventions, and a “Walking the Middle Path” module. This module addresses validation, behavioral change, and polarized interactions between adolescents and caregivers. Caregiver involvement is intended to improve the environment in which skills are used, rather than to assign responsibility for the adolescent’s difficulties.
Adolescent protocols vary in length and organization. The presence of self-harm is not equivalent to a borderline personality disorder diagnosis, and adolescent studies do not necessarily require the same diagnostic criteria as early adult trials. Developmental stage, confidentiality, safeguarding, and the roles of caregivers all affect implementation. Evidence from a particular adolescent protocol should therefore not be generalized automatically to every school-based skills curriculum or youth program using the DBT name.
DBT has also been adapted for substance use disorders, including treatment for people with co-occurring borderline personality disorder. These adaptations add substance-specific targets and strategies for maintaining engagement after lapses. “Dialectical abstinence” combines an abstinence-oriented goal with an acceptance-based, problem-solving response when substance use occurs. The intention is to avoid both minimizing a lapse and allowing it to become a reason for abandoning treatment. The supporting research is smaller and more specialized than the principal evidence base for standard DBT.
For post-traumatic stress disorder, distinct approaches integrate DBT principles with trauma-focused treatment. DBT with a prolonged exposure protocol adds structured trauma processing to comprehensive DBT under specified conditions. DBT-PTSD is a separate, multicomponent treatment developed for complex clinical presentations associated with childhood abuse. These approaches should not be conflated with ordinary DBT skills training. Their evidence concerns defined protocols, eligibility criteria, and clinician competencies.
Other adaptations target eating disorders, particularly presentations involving binge eating, or broader difficulties associated with emotion dysregulation. Some trials and clinical studies report improvement, but samples, comparator treatments, and intervention formats differ substantially. The fact that emotional dysregulation occurs in many diagnoses does not establish that DBT is equally effective for all of them, or that it replaces disorder-specific treatment.
Radically open dialectical behavior therapy, or RO DBT, requires a further distinction. It is a separately developed treatment aimed at maladaptive overcontrol, including excessive inhibition, rigidity, and difficulties with social signaling. Standard DBT primarily developed around severe dysregulation and undercontrolled harmful behavior. Similar terminology does not make the two treatments interchangeable. Across adaptations, the relevant question is which defined intervention was tested, for which population, and with what outcomes—not simply whether “DBT” appears in its title.
Implementation, access, and unresolved research questions
Delivering comprehensive DBT requires coordination, training, and protected clinical time. Weekly individual sessions, skills groups, coaching arrangements, and consultation meetings create demands that differ from those of a single-session-format therapy. Services also need procedures for assessment, documentation, crisis coordination, and transitions between levels of care. A program’s ability to provide these functions depends partly on the surrounding health system, not solely on an individual therapist’s skills.
Treatment fidelity concerns whether the delivered intervention retains the model’s specified strategies and functions. Research programs may use recorded sessions and structured adherence ratings. In routine services, training certificates or the use of worksheets do not by themselves establish adherence. Conversely, adapting appointment schedules or communication methods does not necessarily remove the treatment’s essential functions. Clear descriptions of components, staffing, dose, and deviations are important for interpreting both service evaluations and clinical studies.
Access is affected by workforce availability, cost, geography, language, and the practical demands of attending multiple appointments. These barriers can intersect with social determinants of health, including unstable housing, employment constraints, and caregiving responsibilities. An apparent difficulty with attendance may therefore reflect structural obstacles rather than lack of motivation. Cultural adaptation also requires more than translating acronyms: assumptions about emotional expression, autonomy, family obligations, and acceptable assertiveness can influence how skills are understood and used.
Remote delivery can reduce travel demands and extend geographic reach, but it changes aspects of treatment. Group privacy, reliable internet access, coaching boundaries, and coordination during crises require explicit arrangements. Digital worksheets and applications may support practice, yet evidence for a particular digital tool cannot be inferred from evidence for therapist-delivered comprehensive DBT. Use of electronic platforms also raises data privacy questions because monitoring records can contain highly sensitive information about self-harm, relationships, and substance use.
Research on mechanisms examines whether improvement is mediated by increased skills use, changes in emotional responding, reduced avoidance, validation, or features shared with other psychotherapies. Associations between skills use and better outcomes support plausible explanations, but do not establish that skills alone caused the change. Treatment components interact, and therapeutic alliance, expectations, structure, and clinician expertise can contribute alongside specifically DBT-defined methods.
Economic evaluations face a related problem of perspective. A resource-intensive outpatient program may reduce hospital or emergency-service use, but financial results depend on local costs, baseline service use, and the comparison treatment. Findings from one service cannot be assumed to apply elsewhere. Longer-term studies also need to distinguish reduced crisis expenditure from broader gains in functioning and participation.
Further uncertainties concern optimal treatment duration, which participants need the full package, and how benefits can be maintained after treatment ends. Early research often recruited predominantly adult women, leaving less direct evidence for some other demographic groups. Adverse experiences and treatment-related deterioration have also been reported less systematically than symptom improvement. Better reporting of these outcomes, together with functional recovery and patient-defined goals, remains important for evaluating DBT beyond short-term crisis reduction.