healthmedicinet

This article describes the study protocol of an RCT comparing group CBT and group ST for patients with SAD and comorbid APD. Evidence of the effectiveness of group ST on PDs other than borderline personality disorder (BPD) is sparse [69], and to the best of our knowledge, this is the first study on group ST for this specific patient group. Therefore, this trial can provide valuable information on the effectiveness of ST offered in a group format as a treatment option for these patients. It will also extend our knowledge on prolonged CBT, which is the advised treatment for patients with APD according to the Dutch treatment guidelines for treatment of personality disorders [13].

Research on the optimal treatment of APD is scarce. Most studies have investigated treatment of SAD and compared the outcome of patients with and without comorbid APD [70]. APD is in general relatively understudied [71], although it is one of the most prevalent personality disorders across community and clinical samples and is known by its significant functional impairment [71–73]. Since APD is often comorbid with SAD, with data on comorbidity varying between 32–50 % [3], this study can provide clinically relevant information for a substantial group of patients with APD.

The current study has several strengths. The chosen design, an RCT, is recognized to be the best and most definitive way of demonstrating that an intervention is effective [74]. Both treatments are well documented [17, 24, 75] and have been implemented in scientific research before in treatment studies for GCBT (e.g. [76–79]) and individual and group ST (e.g. [19, 43, 80, 81]).

Treatments will be provided by therapists with ample experience in anxiety treatment in specialized mental health care. Besides being trained in the treatment protocol, therapists will be supervised by experienced CBT and ST supervisors.

Both treatment protocols will be suitable for use in a clinical setting. The semi-open character of the groups, in which influx of patients is allowed every 10 weeks, diminishes waiting time and is therefore an ethical solution in case of long waiting lists as a consequence of lengthy treatments for patients with severe psychopathology. Also, from an organizational perspective, in which limitation of waiting time is considered important, this format can be attractive and might possibly even offer a cost-effective option for treatment of APD [13]. The semi-open character enables formation of groups with comparable psychiatric morbidity, allowing for contact with fellow patients. The size of the intended study sample will be sufficient to reliably detect differential treatment effects of a medium effect size or larger, and the relatively limited waiting time is expected to have a positive effect on the influx of patients. The selected measuring instruments are scientifically sound, and the repeated measurements allow for a close monitoring of changes in time, as well as mediation analyses of putative working mechanisms. The follow-up period of one year enables us to monitor how patients fare after the study treatment.

However, there are also some limitations to consider, given the chosen research design. The decision to compare GST with GCBT, with GCBT as treatment as usual (TAU), instead of a no-treatment or attention control condition might affect internal validity, since it sets boundaries on the possibility to control for other potentially influencing factors not directly related to the therapeutic context. When the effects of both treatments prove to be not significantly different, the absence of an inactive condition implies that one cannot exclude the possibility that factors such as passage of time or attention may have critically influenced outcomes observed. The influence of the factor time is, however, expected to be limited, since APD, like other personality disorders, is by definition assumed to be persistent [8], and recent studies show that APD symptoms are relatively stable over time [71, 82, 83]. Moreover, the addition of a control group would, in our opinion, have some serious ethical consequences, because it would deprive help-seeking patients of treatment for too long a period of time.

Also one might argue the validity — for this specific patient group — of GCBT as TAU, because of the relative paucity of scientific evidence for prolonged CBT for APD. Although we subscribe the need for additional scientific evidence, we feel that prolonged GCBT can be considered as TAU based on Dutch guidelines [13] and earlier studies on SAD with comorbid APD [16] or APD with or without comorbid SAD [84–86].

Another point to consider is that both treatments consist of two individual sessions followed by 30 group sessions. A ST treatment of thirty sessions is relatively short. Most published RCTs investigating ST treatments thus far comprise 50 or more sessions covering time spans of more than one year [19, 81, 87]. A longer duration of treatment might possibly show greater improvements on the outcome measures. On the other hand, the RCT of Farrell et al. [80] compared the effectiveness of adding 30 group sessions of ST to TAU and TAU alone in patients with BPD and found significantly greater reductions in BPD symptoms and global severity of psychiatric symptoms in the GST. So we expect that this treatment period will be long enough to detect differential effects of treatment if present.

The semi-open structure of the group treatments might have disruptive effects on the course of therapy because of patients entering and leaving at several time points. The possible difficulties experienced by the newcomers are addressed by adding two individual sessions and by reframing something that might be frightening into something positive (e.g. for CBT to facilitate exposure, and for ST to provoke dysfunctional modes, which can subsequently be addressed). One cannot rule out that this treatment format might influence effectiveness and that this effect is different for both treatment modalities.

Furthermore, execution of the multicentre RCT in two large cities in the Netherlands and under tightly controlled conditions might have implications for generalizability and external validity. When treatment results are promising, implementation studies of ST under more naturalistic conditions (benchmark studies) are needed to obtain further evidence for the treatment modalities.

In conclusion, this RCT will extend our knowledge on the effectiveness of both GCBT and GST and will give us an opportunity to examine if GST is a valuable option for patients suffering from SAD in combination with APD. Moreover, the results will inform us if the effects of GST exceed those of existing practices mainly consisting of prolonged CBT interventions.

Since people with this double diagnosis report a low quality of life and are seriously impaired in social interactions and fulfilling social relationships with a negative impact on their well-being, developing and testing an additional and easily applicable treatment option will have both scientific, clinical and individual value, taking us a step forward in treating these severely disabled patients.