University of Twente Student Theses

Login

Online Treatment of adolescent sleep problems: A text mining analysis of written feedback from CBT-I therapists

Filmer, Hazel (2023) Online Treatment of adolescent sleep problems: A text mining analysis of written feedback from CBT-I therapists.

This is the latest version of this item.

[img] PDF
423kB
Abstract:Chronic insomnia in adolescents presents challenges to their well-being. Cognitive Behavioural Therapy for Insomnia (CBT-I) has shown effectiveness in adults, but its application in adolescents is still being explored. Internet-based CBT-I offers accessibility, potentially helping reluctant adolescents seek help. However, the impact of therapist feedback in this setting is unclear. This study investigates therapist feedback in adolescent Internet-based CBT-I, exploring topics, sentiments, and therapist differences. Data from seven Dutch-speaking CBT-I therapists for adolescents were analysed using text mining techniques. The dataset comprises feedback from 78 adolescent clients in six weekly CBT-I consultations, totalling 586 messages. Term Frequency-Inverse Document Frequency (TF-IDF) analysis and a word cloud revealed frequent terms. A three-topic model was chosen due to higher interpretability for human raters. Latent Dirichlet Allocation (LDA) was used for topic modelling and identified three topics: sleeping duration and quality, sleep habits and patterns, and sleep tracking. Topic identification was supported by ChatGPT. Sentiments were analysed using the NRC emotion lexicon to match relevant text excerpts. Therapists predominantly addressed fear in their feedback, with some also expressing anticipation, joy, and surprise, possibly while addressing clients' fears and concerns related to sleep. Variations among therapists highlighted individualized approaches. Future research should investigate AI feedback, treatment outcomes, and protocol-based vs. free-form feedback in Internet-based CBT-I for adolescents to personalize therapy and enhance outcomes. This research highlights the importance of text mining techniques, such as LDA and LiLaH, in analysing therapists' written feedback in online CBT-I for adolescents with insomnia.
Item Type:Essay (Master)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:70 social sciences in general
Programme:Psychology MSc (66604)
Link to this item:https://purl.utwente.nl/essays/96976
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page