Thinking Well: A randomised controlled feasibility study of a new CBT therapy targeting reasoning biases in people with distressing persecutory delusional beliefs

Abstract

Background and objectives

Delusional beliefs with persecutory content are common in psychosis, but difficult to treat. Interventions targeting hypothesised causal and maintaining factors have been proposed as a way of improving therapy. The current study is a feasibility randomised controlled trial of the ‘Thinking Well (TW)’ intervention: This novel approach combines the recently developed Maudsley Review Training Programme (MRTP), with additional, focussed cognitive-behavioural therapy sessions.

Methods

31 participants with distressing persecutory delusions and schizophrenia spectrum disorders were randomised to TW or to treatment as usual in a 2:1 ratio. Participants completed outcome assessments at 0 (baseline), 1 (post-MRTP), 6 (post-TW) and 8 (follow-up) weeks. Key outcomes included belief flexibility, paranoia, and delusional conviction and distress. Participants allocated to TW completed the MRTP package and four CBT sessions with a clinical psychologist.

Results

Recruitment proved feasible. Participants reported the intervention was relevant and had resulted in positive changes in thinking and mood, which they could use in everyday life. Treatment effects were moderate-large for key outcomes including belief flexibility, paranoia conviction and distress. The additional TW sessions appeared to confer benefits over MRTP alone.

Limitations

Assessments were not carried out blind to treatment condition. Recruitment was opportunistic, from an identified pool of research participants. Finally, a few participants had already completed the MRTP as part of a previous study.

Conclusions

The TW intervention appears to be feasible and acceptable to participants, and the effects of treatment are promising. A fully powered randomised controlled trial of the intervention is warranted.

Highlights

• •

  The Thinking Well intervention was feasible to deliver and feedback was positive.

• •

  Analyses suggest large effects post intervention, with confidence intervals favouring the intervention.

• •

  Effects on state paranoia were not sustained; future work should focus on maintenance.

• •

  A larger study powered to detect changes in key outcomes is warranted.