Goal content and attainment in digitally supported CBT for psychosis: A mixed-methods analysis

Cognitive behavioural therapy for psychosis (CBTp) is recommended as an evidence-based intervention for individuals experiencing psychosis (NICE, 2014). Evaluations of CBTp have traditionally prioritised standardised symptom-based outcomes. While symptom change is clinically important, recovery-oriented approaches to mental health emphasise that meaningful progress is defined not only by symptom reduction but also by improvements in safety, autonomy, relationships, wellbeing, and participation in valued activities. There remains limited empirical examination of both the types of goals individuals prioritise within CBTp and the extent to which progress is made toward these personally defined outcomes, particularly within routine clinical settings. This study is a mixed-methods analysis of goal-based outcome data collected within an ongoing hybrid type 2 implementation–effectiveness study of digitally supported CBTp (SloMo; ClinicalTrials.gov identifier: NCT06568081). SloMo integrates face-to-face CBTp sessions with an online therapy platform and mobile application designed to support formulation, reflection, and between-session practice. The digital platform incorporates the Goal-Based Outcomes (GBO) measure (Law et al., 2015), which enables the systematic recording of collaboratively defined free-text therapy goals and self-rated progress on a 0–10 scale. The GBO is included as one of the outcome measures within the broader implementation–effectiveness study. While the parent study examines change in GBO scores alongside other clinical outcomes, the present analysis focuses specifically on goal-based outcomes. In addition to examining change over therapy, this study explores the content of collaboratively defined goals and associations between participant characteristics and goal attainment. The purpose of the present study is threefold: 1. To describe the types of goals individuals set during digitally supported CBTp. 2. To examine whether goal-based outcome scores change from early to late therapy. 3. To explore whether baseline demographic and clinical characteristics are associated with the magnitude of change in goal-based outcome scores. To address the first objective, a qualitative content analysis will be conducted on baseline goal statements recorded within the embedded GBO measure. Individual goal statements constitute the unit of analysis. An iterative coding process will be used to identify and refine goal domains that capture common themes across participants. Participants may contribute to multiple goal categories where applicable. Frequencies of goal categories will be reported at the participant level to characterise the distribution of therapy priorities. To address the second objective, change in goal-based outcome scores will be examined by comparing baseline and end-of-therapy GBO ratings among participants with available data at both time points. Change will be operationalised as the difference between baseline and post-therapy scores. Paired-sample statistical tests will be used to evaluate mean change, and effect sizes will be calculated to quantify the magnitude of improvement. In addition, clinically meaningful improvement will be defined pragmatically as an increase of two or more points on the GBO scale, and the proportion of participants meeting this threshold will be reported. To address the third objective, exploratory univariable linear regression analyses will be conducted to examine associations between baseline demographic variables (e.g., age, gender, ethnicity) and baseline clinical characteristics (paranoia severity and wellbeing) and change in GBO scores. Change in GBO score will serve as the dependent variable. These analyses are exploratory and aim to examine whether improvement in personally defined goals varies across participant characteristics within routine clinical practice. This study is situated within routine NHS services and reflects real-world implementation conditions rather than a tightly controlled efficacy trial. It therefore provides an opportunity to examine both the content of service-user-defined goals and goal attainment within everyday clinical settings. It is anticipated that participants will identify a broad range of goals extending beyond symptom reduction, including domains related to safety, independence, social participation, emotional wellbeing, and self-understanding. It is further expected that goal-based outcome scores will show improvement from early to late therapy. Associations between baseline demographic or clinical characteristics and goal change are examined without strong directional hypotheses, given limited prior research in this area.