Treatment outcomes across subgroups of stroke survivors receiving psychological treatment for anxiety and depression

Résumé fourni par la source

BACKGROUND: Stroke increases the risk of developing depression and anxiety. While psychological therapies are a recommended first-line treatment, outcomes vary between stroke survivors. Identifying those at risk of poorer treatment outcomes could improve treatment selection and patient outcomes. This study aimed to identify classes of stroke survivors based on stroke-related clinical characteristics and examine their psychological treatment outcomes. METHODS: NHS Talking Therapies for anxiety and depression (NHS TTad) data, linked to secondary care health records, were used to identify stroke survivors referred to TTad services across England from 2012 to 2019 (N = 6349). Latent class analysis was used to identify patient subgroups based on stroke type, time from stroke diagnosis to psychological treatment, presence of hemiplegia or paraplegia, and physical comorbidities. A distal categorical outcome approach was used to assess treatment across classes. Logistic regression models were used to examine outcomes by treatment intensity within each class. RESULTS: Four statistically distinct classes were identified, primarily differentiated by the type of stroke and time to access psychological treatment. Reliable recovery rates varied from 50 to 55% across classes, with deterioration rates ranging from 5 to 8%. High-intensity treatment was associated with reduced dropout rates in three classes. CONCLUSION: There were limited differences in psychological treatment outcomes between classes of stroke survivors. Stroke-related variables used in this study appear insufficient on their own to meaningfully stratify stroke survivors by psychological treatment prognosis. Future research should incorporate additional patient-level data, such as measures of cognition and language, to develop more clinically informative subgrouping.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.