گروه CBT برای روان پریشی: آزمایش کنترل شده طولی با بیماران بستری در بیمارستان

Individual cognitive behaviour therapy for psychosis (CBTp) is a recommended treatment in the acute phase and beyond. However, less is known about the effectiveness of group CBTp in acute care. This mixed methods study explored the implementation and effectiveness of brief group CBTp with inpatients. This prospective trial compared inpatients who received either a four week group CBTp program or treatment as usual (TAU). Participants (n = 113 at baseline) completed self-report measures of distress, confidence and symptoms of psychosis at baseline, post-intervention and one month follow up. CBTp group participants also completed a brief open-ended satisfaction questionnaire. Using complete case analysis participants who received CBTp showed significantly reduced distress at follow up compared to TAU and significantly increased confidence across the study and follow up period. However, these effects were not demonstrated using a more conservative intention-to-treat analysis. Qualitative analysis of the satisfaction data revealed positive feedback with a number of specific themes. The study suggests that brief group CBTp with inpatients may improve confidence and reduce distress in the longer term. Participants report that the groups are acceptable and helpful. However, given the methodological limitations involved in this ‘real world’ study more robust evidence is needed.

Cognitive behaviour therapy for psychosis (CBTp) has been widely researched over the last 20 years and there is considerable evidence that it is an effective intervention (Wykes, Steel, Everitt, & Tarrier, 2008). Guidelines for professionals recommend individual CBTp in the treatment of schizophrenia (American Psychological Association, 2004 and Canadian Psychiatric Association, 2005) and some recommend that this should start in the acute phase (Royal Australian and New Zealand College of Psychiatrists (RANZCP), 2005 and National Institute for Health & Clinical Excellence, 2010). Mental health service providers must consider how best to offer treatment within the financial constraints of the current economic climate (World Health Organisation, 2013). Group therapy is a practical way of streamlining therapy and several randomised controlled trials (RCTs) have been conducted comparing group CBTp with treatment as usual (Barrowclough et al., 2006 and Wykes et al., 2005), group psycho-education (Bechdolf et al., 2004 and Bechdolf et al., 2010), social skills training (Lecomte et al., 2008) or enhanced supportive therapy incorporating emotional support and non-symptom related counselling (Penn et al., 2009) with mixed findings. There is some evidence that long term group CBTp can be more effective than individual CBTp if used as an early intervention (Saska, Cohen, Srihari, & Woods, 2009) or for those with less severe symptoms (Lockwood, Page, & Conroy-Hiller, 2004). While a review of the literature on CBTp found no differences in effect sizes between group and individual therapy, it suggested that clustering effects in group therapy may improve treatment efficacy (Wykes et al. 2008). Unfortunately, there is considerable heterogeneity amongst the type and length of therapy interventions used in these studies (e.g. ranging from 8 to 24 sessions, and based on different CBTp manuals) and the type of measures used to assess change (e.g. positive and negative symptom scale (PANSS; Kay, Fiszbein, & Opler, 1987), psychotic symptoms rating scales (PSYRATS; Haddock, Mc Carron, Tarrier, & Faragher, 1999), Beliefs about voices questionnaire (BAVQ; Chadwick, Lee & Birchwood, 2000), brief psychiatric rating scale (BPRS; Ventura, Green, & Shaner, 1993) and many more) making direct comparisons difficult. Moreover, the majority of this research has only studied outpatient populations. Group therapy in inpatient settings is challenging in a number of ways. First, the timing of intervention, because service users are currently experiencing crisis, there is considerable uncertainty regarding length of stay in hospital, and a common increase or change in medication at the time of admission. A recent systematic review concluded that there are positive signs that group CBTp in inpatient settings may be effective, but more robust evidence is needed (Owen, Speight, Sarsam, & Sellwood, 2014). There are similar difficulties in the outpatient literature regarding heterogeneity in type and length of therapy, and the plethora of assessment measures used to assess change. In addition, inpatient research has often used small sample sizes (Haddock, Tarrier, et al., 1999) or lacked treatment as usual (TAU) control groups (Dannahy et al., 2011 and Pinkham et al., 2004). But research has shown positive findings in terms of service users' experiences of participating in groups (Bickerdike & Matias, 2010) and general wellbeing (Drinnen, 2004). Several studies have started to move away from pure CBTp manuals and include elements of person based therapy (Dannahy et al. 2011), acceptance and commitment therapy (Gaudiano & Herbert, 2006) or mindfulness (Chadwick et al., 2005 and Drinnen, 2004). There is also some encouraging evidence that incorporating CBTp groups into routine practice in acute inpatient care can reduce readmission rates (Svensson et al., 2000 and Veltro et al., 2006). In line with this evidence and calls from service users for more choice of treatment in hospital (James, 2001), UK government initiatives for best practice on inpatient wards include the provision of talking therapy groups (Bright, 2006 and Department of Health, 2007). One example of this in clinical practice comes from Clarke and colleagues who designed an inpatient therapy group adopting a recovery approach based on CBTp and mindfulness, encouraging normalisation of symptoms and education on emotional coping skills, arousal management and problem solving (Hill, Clarke, & Wilson, 2009). They ran the group in four weekly sessions and measured participants' levels of distress, perception of control over their mental health, their goals regarding their mental health and their experiences of the group (Phillips, Clarke & Wilson, unpublished). Due to the small sample size no statistically significant changes were found but the feedback from service users about their experiences of the group were positive, particularly regarding increased wellbeing and decreased isolation. Unfortunately, this study did not have a control group so it is not possible to determine whether the findings were due to the group intervention or some other variable. Further research with a larger sample size and a control group is necessary in order to provide more robust evidence for the positive effects of such a group.