A feasibility study of a cross-diagnostic, CBT-based psychological intervention for acute mental health inpatients:...

Acute psychiatric inpatient wards are characterized by minimal provision of therapeutic activities and high readmission rates. Implementation of a comprehensive inpatient psychological intervention service has been recommended to overcome these problems; however, whether this is feasible or effective remains unclear.

This non-randomized parallel cluster feasibility trial examined the feasibility of delivering and evaluating cross-diagnostic psychologically informed acute psychiatric care the Edinburgh-Acute Psychological Inpatient Therapy Service (EDAPTS) and gathered preliminary clinical outcome data. Patients able to consent and complete questionnaires were recruited from two adult acute wards (i.e., clusters) and received either EDAPTS plus TAU or TAU.

Between October 2015 and October 2016, 96 inpatients were recruited. Findings suggested that there were good data completion rates for several clinical outcomes, that several EDAPTS components were successfully delivered, and that some initial effects appeared to favour the intervention, depending on outcome. However, difficulties relating to the recruitment process were also identified, as well as problems relating to adequate delivery of group therapies, participant engagement in some intervention components, and data completion at follow-up.

These issues, and the feasibility of randomization and rater-blinding, have important implications for the design of future trials. Overall, this study provides an important insight into the challenges and complexities of developing and evaluating a comprehensive psychological intervention service in an acute psychiatric setting.

Individual therapy sessions can be delivered in the acute environment. The EDAPTS intervention showed some promise on outcomes of distress and self-efficacy. Delivery of nurse-led groups was challenging and may need to be embedded into routine clinical practice to increase intervention and outcome reach. More parameters, for example, randomization at cluster level, should be tested before progressing to an adequately powered, single-blind, definitive cluster RCT.