Cognitive Behavioural Therapy and Adjunctive Physical Activity for Functional Movement Disorders (Conversion Disorder): A Pilot, Single-Blinded, Randomized Study

Functional movement disorders (FMD) are one of the commonest conditions seen in neurological practice [1], thus having a remarkable impact on the health care system. Nonetheless, there are no treatment recommendations, and their prognosis is generally unfavourable [1]. Evidence supporting cognitive behavioural therapy (CBT) in FMD is scarce, and two reviews concluded that caution is needed in assuming that CBT is necessarily beneficial [2,3]. Owing to preliminary suggestions that adjunctive physical activity (APA) might also be beneficial in FMD [2,3], we here aimed to test the feasibility and efficacy of CBT and APA in FMD.Thus, 29 patients with FMD (conversion disorder [4]) were randomized using a block procedure (block size = 4 with balanced combinations) to receive CBT alone (90-min session, once a week) or CBT + APA (60-min session, twice a week) (fig. 1). As a control group, we enrolled 8 patients under standard medical care. All patients were diagnosed elsewhere, but they were engaged at enrolment in a discussion about their diagnosis and its implications. Both treatments ran for 12 weeks. The CBT protocol was adapted from previous studies on conversion disorder [5]. Thus, CBT focused on identifying somatic misinterpretations, negative thoughts, and illness beliefs maintaining FMD and low mood or anxiety, and patients were engaged in establishing alternative hypotheses for the bodily sensations. Distraction techniques were developed based on the current view on FMD [1]. Hence, as most of our patients were affected with functional tremor (online suppl. table 1; see www.karger.com/doi/10.1159/000446660 for all online suppl. material), motor strategies were implemented (for instance tapping at a certain frequency with a non-tremulous body region) to be used to reduce/interrupt the specific FMD. The approach was formulation based, and any issues raised during the sessions were addressed. Sessions also incorporated the planning and review of homework, which included completing records about FMD, avoiding behaviours, solving techniques, etc. The APA consisted of structured low/moderate-intensity walking to be conducted as a group at a country track or individually when the subject was unable to attend the collective session. Details on this type of physical activity have been described elsewhere [6].Patients were assessed at baseline (T0) and at the end of the 12-week programme (T1; 9.2 ± 1.8 weeks for the control group). All subjects were videotaped at T0 and T1 using the Psychogenic Movement Disorder Rating Scale (PMDRS) [7]. A single rater (R.E.), blinded to the aims of the study and time evaluation (e.g. T0, T1), completed the PMDRS (severity subscore) on videotape in a randomized order. Patients completed the PMDRS functional impact subscore, based on self-reported incapacitation. Total PMDRS score (severity + functional impact) represented the primary outcome, whereas the two PMDRS subscores as well as the Hamilton Depression Scale, the Beck Anxiety Inventory, and the Patient Health Questionnaire-15 (PHQ-15) represented secondary outcomes.Eight patients (27.6%) withdrew their consent; hence, 21 patients in the active groups are described here (fig. 1). Table 1 summarizes the examined variables at T0 and T1. Specifically, ANOVA tests for repeated measures showed a significant improvement of the primary and all secondary outcomes over time for both active groups (all p < 0.001) but not for the control group (all p > 0.05). Moreover, there were no differences between the two active groups (all p > 0.05). Then, multiple regression analyses were run to explore whether certain baseline features could influence the primary outcome. Higher motor severity at baseline (p = 0.045) and longer disease duration (p = 0.043) were significantly associated with a poorer outcome (R2 = 0.45; p < 0.005). Moreover, a trend was observed for PHQ-15 (p = 0.057) when the outcome was normalized for disease duration.While our results are preliminary, a few clear considerations would be possible. First, both active groups significantly improved over time (vs. the control group), with no differences between the two of them. This indicates that CBT is feasible (compliance over 70%) and effective in improving the specific motor symptoms in patients with FMD. Very limited knowledge exists about CBT in this group of patients, but there is evidence showing the efficacy of CBT in patients with psychogenic non-epileptic seizures [2,5] supporting our findings. In this context, it is noteworthy that none of our patients had overlapping psychogenic non-epileptic seizures, thus suggesting that CBT was effective in improving the specific motor phenotype of our patients.Second, CBT was also effective in ameliorating depressive symptoms and anxiety. However, the improvement in the primary outcome (e.g. the specific motor disorder) was not directly driven by the amelioration of affective symptoms. This finding reflects the conceptual shift that has been put forward in the current DMS-5, where the criteria for the diagnosis of FMD discard the presence of causative psychological and psychiatric factors and emphasize the crucial importance of the neurological examination [4]. The only factors influencing the outcome were a severer motor phenotype at baseline and longer disease duration. The latter, in line with previous results [8], strongly reflects the need for avoiding delay in diagnosis and referral to appropriate treatments. While our results show that CBT is effective in chronic FMD (mean disease duration of our patients was in fact about 1.5 years), it remains to be assessed whether the magnitude of therapeutic response would be even greater in de novo patients (only 2 of our patients had disease duration of - and received their first diagnosis from - less than 6 months to be considered de novo and to directly test this hypothesis). Worthy of note is the fact that when the outcome was normalized for disease duration a trend (which we would advocate was due to the small sample size) was observed for higher baseline PHQ-15 scores being associated with higher PMDRS after treatment. PHQ-15 is a self-administered screening scale capturing the presence and severity of a number of somatic symptoms (neurological or otherwise) [9]. This reinforces the concept that FMD represent a multifaceted entity, where the central motor disorder can (or not) occur in the context of other somatic and/or affective symptoms, and poses the question of what should be considered as the ‘best' outcome to depict the whole picture of FMD.Third, we failed to find a beneficial effect of APA on any outcomes. Previous studies supporting APA in the management of FMD often focused on multi-disciplinary treatment (including CBT) [2,3], thus hampering the understanding of the relative contribution of each approach. We do not believe the small sample size played a major role here. To explain such a discrepancy, we would rather consider the specific motor phenotype of our patients. Indeed, they were mostly affected with tremor (approx. 75%), whereas a previous cohort included far more patients with dystonia and gait disturbances [2,3], on whom APA would seem, intuitively, more likely to work. Hence, the possible confounding role of the specific motor phenotype should be considered in future trials.We acknowledge the pilot nature of our study and the fact that the control group was re-evaluated approximately 9 weeks (rather than 12) after the baseline evaluation. However, while preliminary, our results clearly show that CBT is highly effective in FMD, arguing for a collaborative effort to overcome the discourse between neurologists, psychiatrists, and psychologists regarding these patients. It is in fact paradoxical for such a seemingly effective treatment as CBT that, although half of the patients with functional symptoms are identified by neurologists at the first contact, only 1.5% are referred to a psychologist [10].The authors have no conflicts of interest to disclose.