Early response in cognitive-behavior therapy for syndromes of medically unexplained symptoms

Maria Kleinstäuber, Michael J. Lambert, Wolfgang Hiller

Early dramatic treatment response suggests a subset of patients who respond to treatment before most of it has been offered. These early responders tend to be over represented among those who are well at termination and at follow-up. Early response patterns in psychotherapy have been investigated only for a few of mental disorders so far. The main aim of the current study was to examine early response after five therapy-preparing sessions of a cognitive behavior therapy (CBT) for syndromes of medically unexplained symptoms (MUS). In the context of a randomized, waiting-list controlled trial 48 patients who suffered from ≥3 MUS over ≥6 months received 5 therapy-preparing sessions and 20 sessions of CBT for somatoform disorders. They completed self-report scales of somatic symptom severity (SOMS-7 T), depression (BDI-II), anxiety (BSI), illness anxiety and behavior (IAS) at pre-treatment, after 5 therapy-preparing sessions (FU-5P) and at therapy termination (FU-20 T). The current analyses are based on data from the treatment arm only. Repeated measure ANOVAs revealed a significant decrease of depression (d = 0.34), anxiety (d = 0.60), illness anxiety (d = 0.38) and illness behavior (d = 0.42), but no change of somatic symptom severity (d = -0.03) between pre-treatment and FU-5P. Hierarchical linear multiple regression analyses showed that symptom improvements between pre-treatment and FU-5P predict a better outcome at therapy termination for depression and illness anxiety, after controlling for pre-treatment scores. Mixed-effect ANOVAs revealed significant group*time interaction effects indicating differences in the course of symptom improvement over the therapy between patients who fulfilled a reliable change (i.e., early response) during the 5 therapy-preparing sessions and patients who did not reach an early reliable change. Demographic or clinical variables at pre-treatment were not significantly correlated with differential scores between pre-treatment and FU-5P (-.23 ≤ r ≤ .23). Due to several limitations (e.g., small sample size, lack of a control group) the results of this study have to be interpreted cautiously. Our findings show that reliable changes in regard to affective-cognitive and behavioral variables can take place very early in CBT of patients with distressing MUS. These early changes seem to be predictive of the outcome at therapy termination. Future studies are needed in order to replicate our results, and to identify mechanisms of these early response patterns in somatoform patients. ISRCTN. ISRCTN17188363 . Registered retrospectively on 29 March 2007.