Enriching Cognitive-Behavioral Therapy With Emotion Regulation Training in Patients With Chronic Multiple Somatoform Symptoms (ENCERT): A 3-Year Follow-up of a Randomized Controlled Trial

Somatic symptoms not caused by a known biomedical condition ("somatoform disorders") are highly prevalent, involve a high risk of chronicity, are one of the major reasons for doctor visits, and are a tremendous burden for health care systems. Despite the economic relevance, research activities are disproportionately low. The only intervention with an evidence grade I is cognitive behavioral therapy (CBT). However, average effect sizes for CBT in somatization syndromes are only moderate (Cohen's d < 0.5), and more powerful interventions are needed.

There is convincing evidence that patients with somatoform disorders have emotion regulation deficits, which are not addressed by current CBT approaches. We conducted a randomized controlled trial in which we compared a conventional CBT for patients with chronic multiple somatoform symptoms with CBT complemented with emotion regulation techniques (ENCERT; see ClinicalTrials.gov ID: NCT01908855). Previous interventional studies have mainly focused on acute effects and rarely include follow-up periods > 2 years after the end of therapy. The primary goal of the proposed project is to investigate whether effects that were demonstrated for both treatments in our previous randomized controlled trial can be maintained until 3 years after the end of therapy.

This study is based on the intention-to-treat (ITT) sample (N=255 patients) of our previous randomized controlled trial (ClinicalTrials.gov ID: NCT01908855). Patients were primarily recruited via referrals by primary care doctors. After a screening phase baseline assessments with different self- and clinician rating scales (see Outcome Measures) followed. Participants who fulfilled the eligibility criteria were randomized to one of the two study arms: cognitive behavioral therapy vs. cognitive-behavioral therapy complemented with emotion regulation strategies. The post assessment took place after the 20th therapy session and at a follow-up at six months after post treatment. In the current project all randomized patients will be reassessed 3 years after the end of therapy.

Different methods were applied to prevent bias and to assure a high quality level of the current study. Data handling, data monitoring and statistical analyses were supervised by the Coordinating Center for Clinical Trials (KKS) of Philipps-University of Marburg; data quality and safety principles were applied. Additionally an independent Data Safety Monitoring Board was nominated. Randomisation occured and was controlled centrally through the randomisation's central office in KKS Marburg. The current study qualifies as a single-blinded trial. Assessment interviews are conducted and analyzed by people blinded to the treatment condition. Treatments in the previous randomized controlled trial were manualized, and therapists received an intense training. Treatment fidelity/integrity were analyzed with rating schemes for an priori defined proportion of randomly selected videotaped treatment sessions. Allowed additional treatments during study inclusion were thoroughly monitored and analyzed.

The sample size calculation was based on the primary outcome variable "somatization severity index" of the Screening of Somatoform Disorders (SOMS-7T). Based on results of the main validation study of SOMS-7T, metaanalytic estimations, and results of a pilot study of the efficacy of ENCERT, the power calculations yield a necessary total sample size of N=194 to detect a clinical relevant difference of 4 points symptom reduction on SOMS-7T between the two treatments with a power of 0.80 and an alpha=.05. With regard to an estimated drop-out rate of 25%, N=255 patients were recruited.

As main efficacy analysis the primary outcome shall be analyzed with linear mixed-effect models according to the ITT principle. Multilevel longitudinal mediation analyses will be conducted in order to test whether the effect of treatment condition on intraindividual changes in somatic symptom severity can be explained by intraindividual changes in emotion regulation skills.