TL;DR: RO-DBT, developed by Thomas R. Lynch, targets maladaptive overcontrol, meaning excessive self-restraint, rigid rule-following and inhibited emotional expression, the opposite of the undercontrol standard DBT was built for. It has been studied mainly in restrictive anorexia. The evidence is small and mostly uncontrolled: one inpatient series, two small outpatient studies, and a randomized trial that has not yet reported.
A patient, a composite drawn from several patients, finishes a full course of standard DBT for restrictive anorexia and reports, accurately, that she used the skills. She rode out distress without acting on it, tolerated hunger and urges and the discomfort of a meal plan, did everything the protocol asked, and the restriction held through all of it.
Lynch’s account reads an outcome like this as a mismatch between a treatment and the problem it was pointed at: standard DBT was designed for undercontrol, and he treats restrictive anorexia as a disorder of overcontrol. Radically Open Dialectical Behavior Therapy, or RO-DBT, was built for that difference.
Overcontrol and undercontrol
Thomas R. Lynch, the clinical psychologist who developed RO-DBT and published its treatment manual in 2018, describes self-control as a spectrum with two failure points, where the usual clinical assumption is that a patient has too little of it.
Undercontrol is the familiar failure: impulsivity, mood-driven behavior, difficulty inhibiting urges, and relationships that swing between closeness and rupture. Standard DBT, developed by Marsha Linehan for chronic suicidality and borderline personality disorder, teaches distress tolerance and emotion regulation so that patients can wait out intense emotion and urges instead of acting on them.
Overcontrol runs the other way: excessive self-restraint, rigid rule-following, perfectionism, inhibited emotional expression, a preference for structure over spontaneity, and guarded relationships. People with overcontrol are often high-achieving and conscientious, and in Lynch’s account they are also frequently lonely, since their habitual restraint keeps other people at a distance.
Lynch’s central claim is that overcontrol is kept going by social exchanges as well as by what happens inside the person. When someone masks distress, holds a flat expression and rarely shows vulnerability, others read the signals as distance and respond in kind, which lowers social connection and deepens emotional loneliness, which in turn reinforces the overcontrol. He calls the behaviors that carry this loop social signaling.
Why anorexia is treated as a case of overcontrol
Hempel, Vanderbleek and Lynch, writing in the journal Eating Disorders in 2018, described anorexia nervosa as a “prototypical” overcontrolled disorder, one in which patients show little openness or flexibility, inhibit their emotional expression, have limited awareness of their own emotions and few close social ties, and live with high emotional loneliness as a result. In their account, self-starvation and the flattened social signals that accompany it work as strategies for reducing negative feeling, and the treatment goal shifts from emotional regulation to social connection.
The framing fits restrictive presentations as clinicians in this field describe them. Rigid rules around food, a high capacity to override hunger and fatigue, perfectionism, inhibited emotional display, and social withdrawal that deepens as the illness goes on are the features of overcontrol expressed through eating. The frame also fits the ego-syntonic quality of the illness, the sense that the restriction belongs to the self instead of intruding on it, which is covered in why anorexia can feel like part of who you are.
Anorexia and other eating disorders occur at every body size, and the overcontrol frame describes temperament and behavior, so it says nothing about a patient’s weight. The frame is also specific to restrictive presentations: binge-eating disorder and bulimia nervosa involve loss of control over eating, and for those presentations standard DBT and its emotion-regulation skills are the usual match. The broader eating disorders archive covers those diagnoses separately.
Where standard DBT skills can go wrong in restriction
In clinical practice, standard DBT distress-tolerance skills can become more tools for restraint when the patient’s core problem is already too much restraint. A skill built to help a dysregulated person wait out an urge to self-harm works just as well on an urge to eat, so in a restrictive presentation it may serve the restriction instead of loosening it.
A second pattern seen in practice is that emotional loneliness, more often than dysregulation, sits behind the rigidity. Patients who mask feeling in group and answer every check-in with “fine” may be holding their feelings back from the people in the room, and skills that teach them to contain more will not reach that loneliness. Hempel and colleagues name the same loneliness as the treatment target.
For families sorting through DBT options for a teenager, the distinction between DBT-A and standard DBT concerns age and format, while the distinction between standard DBT and RO-DBT concerns which direction the self-control problem runs. A teen whose anorexia comes with self-harm and emotional crises may need the standard skills, and when FBT is not enough covers that case; a teen who is rigid, rule-bound and composed is the one the overcontrol frame describes.
What RO-DBT teaches instead
Standard DBT organizes its skills into four modules: mindfulness, distress tolerance, emotion regulation and interpersonal effectiveness. The aim is to help a dysregulated person slow down, tolerate intense states and choose a response.
RO-DBT retains a behavioral, skills-based structure but aims the skills in the opposite direction, at three targets Lynch sets out: openness, flexible responding, and social connection built on shared vulnerability. In practice that means learning to notice when one is shutting down or masking, to signal emotion outward in ways others can read, to loosen rigid rule-following, and to risk the vulnerability that lets closeness form. Hempel, Vanderbleek and Lynch’s 2018 paper lists body posture, gestures and facial expression among the skills, on the premise that changing how a person communicates an inner state is what breaks the loneliness that keeps overcontrol in place.
The standard outpatient protocol in Lynch’s 2018 manual runs about 30 weeks, with a weekly individual session of about an hour and a weekly skills class of about two and a half hours. The anorexia studies described below used 32-week and 40-week versions.
| Standard DBT | RO-DBT | |
|---|---|---|
| Target problem | Undercontrol: impulsivity, emotion dysregulation (Linehan) | Overcontrol: rigidity, inhibited expression, social withdrawal (Lynch) |
| What the skills aim at | Tolerating distress, regulating emotion, inhibiting urges | Openness, flexible responding, social connection |
| Proposed mechanism | Emotion regulation | Social signaling; less emotional loneliness (Hempel, Vanderbleek and Lynch, 2018) |
| Eating-disorder presentations studied | Bulimia nervosa (Safer, Telch and Agras, 2001); binge-eating disorder (Telch, Agras and Linehan, 2001) | Restrictive anorexia (Lynch et al., 2013; Isaksson et al., 2021; Ejdemyr et al., 2025) |
| Format, as sourced | Weekly individual therapy, skills group, phone coaching, consultation team (Linehan) | About 30 weeks: weekly hour-long individual session plus 2.5-hour skills class (Lynch, 2018 manual) |
| Evidence state in eating disorders | Randomized trials against wait-list controls in bulimia (31 women) and binge-eating disorder (44 women) | Three studies without a comparison group, 83 patients in all; one randomized trial registered, not yet reported |
What the evidence shows
The most cited eating-disorder data come from Lynch et al. (2013), published in BMC Psychiatry: an uncontrolled study of 47 adults with restrictive-type anorexia treated in an inpatient program, with an average stay of about 22 weeks. Seventy-two percent completed the program, and among completers the authors reported large improvements in eating-disorder symptoms, quality of life and psychological distress. There was no comparison group, and the authors described the findings as preliminary support that called for a randomized trial.
Isaksson, Ghaderi, Ramklint and Wolf-Arehult (2021), in the Journal of Behavior Therapy and Experimental Psychiatry, followed 13 adult women with mild to moderate anorexia through a 40-week outpatient RO-DBT program using a multiple-baseline single-case design, in which each patient is measured weekly for four to six weeks before treatment begins and then throughout it, so that each serves as her own comparison; the study was registered as ISRCTN47156042. Eight of the 13 completed, improvements began after RO-DBT started rather than during the baseline weeks, and the authors said the result was preliminary and needed replication.
Ejdemyr and colleagues (2025), also in Eating Disorders, ran a 32-week standard RO-DBT program for 23 outpatients with anorexia in Sweden, registered as NCT06244576 through Umeå University. Nineteen of the 23 stayed to the end, treatment satisfaction was high, and eating-disorder symptoms fell, with an effect size of d = 0.90 on eating-disorder psychopathology. The study had a single arm and no control group, so it cannot say how much of the change RO-DBT caused, and the authors concluded that the treatment was feasible, acceptable and potentially effective.
A randomized trial is under way. Avila-Parcet and colleagues published its protocol in BMC Psychiatry in April 2025: at Hospital de la Santa Creu i Sant Pau in Barcelona, weight-restored adults with anorexia are randomized to treatment as usual or treatment as usual plus a 30-week RO-DBT skills group, with quality of life at six and 12 months as the primary outcome (NCT06050421). The registry entry, last verified in September 2023, now lists the trial’s status as unknown, with recruiting as its last known status and completion estimated for mid-2026, and no results had been published when this post was written.
Where RO-DBT fits
RO-DBT is a targeted option for the overcontrolled restrictive patient, delivered by programs whose clinicians have trained in it specifically, since its skills and its individual-therapy structure differ from standard DBT. For adolescents with anorexia, family-based treatment carries the strongest evidence, and the reasons FBT alone is not always enough are covered separately. For adults, protocols such as enhanced cognitive behavioral therapy, CBT-E, have been tested in randomized trials in anorexia, which RO-DBT has not.
RO-DBT for ARFID is earlier still. ARFID is a separate diagnosis in which restriction is not driven by a wish to be thin, and the better-studied adult treatment is CBT-AR, the Thomas and Eddy protocol covered in CBT-AR for adults. The hypothesis that RO-DBT could help the subset of ARFID patients whose restriction rests on rigidity and perfectionism has not been tested in any study.
Some situations call for a physician before any protocol question. Fainting or near-fainting, chest pain, a very slow or irregular heartbeat, dizziness on standing, signs of dehydration, or a day or more without eating or drinking: contact a physician the same day. For chest pain or fainting, call 911 or go to an emergency room. In a mental-health crisis, or if you are thinking about suicide, call or text 988, the Suicide and Crisis Lifeline.
Before choosing a protocol, the useful question is diagnostic: whether the restriction is driven by too much control or too little. The answer decides which treatment fits, and sorting it out is part of what a first session is for. More on the diagnosis itself is in the anorexia archive. The composite patient who finished standard DBT unmoved had practiced every skill she was taught, and the restriction had held through all of it.