Introduction: Treatment non-response occurs regularly, but psychotherapy is seldom examined for such patients. Existing studies targeted single diagnoses, were relatively small, and paid little attention to treatment under real-world conditions. Objective: The Choose Change trial tested whether psychotherapy was effective in treating chronic patients with treatment non-response in a transdiagnostic sample of common mental disorders across two variants of treatment delivery (inpatient and outpatient). Methods: The controlled nonrandomized effectiveness trial was conducted between May 2016 and May 2021. The study took place in two psychiatric clinics with N = 200 patients (n = 108 inpatients and n = 92 outpatients). Treatment variants were integrated inpatient care versus outpatient care based on acceptance and commitment therapy (ACT) for approximately 12 weeks. Therapists delivered individualized and non-manualized ACT. Main outcome measures were symptoms (Brief Symptom Checklist [BSCL]); well-being (Mental Health Continuum-Short Form [MHC-SF]), and functioning (WHO Disability Assessment Schedule [WHO-DAS]). Results: Both inpatients and outpatients showed decreases in symptomatology (i.e., BSCL: d = 0.68) and increases in well-being and functioning (MHC-SF: d = 0.60 and WHO-DAS: d = 0.70), with more improvement in the inpatients during treatment. Both groups maintained gains 1 year following treatment, and the groups did not significantly differ from each other at this timepoint. Psychological flexibility moderated impact of stress on outcomes. Conclusions: Psychotherapy as practiced under routine conditions is effective for a sample of patients with common mental disorders, a long history of treatment experience and burden of disease, in both inpatient and outpatient settings. Trial registration: This study was registered in the ISRCTN registry on May 20, 2016, with the registration number ISRCTN11209732.