# Digital Cognitive Behavioural Therapy for Insomnia Delivered Within a Crenotherapy Setting: Results from a Multicentre Proof-of-Concept Randomised Controlled Trial

## TL;DR

This study tested whether adding a spa treatment to digital insomnia therapy helps more than the therapy alone, finding it may benefit younger people and those with anxiety.

## Contribution

The study explores the integration of digital CBT-I with crenotherapy, identifying potential subgroup benefits for younger individuals and those with anxiety.

## Key findings

- Both groups improved in insomnia severity and psychological symptoms, but no overall benefit from crenotherapy.
- Younger participants (<60) in the crenotherapy group showed greater insomnia symptom improvement.
- Crenotherapy-delivered dCBT-I led to greater anxiety reduction in participants with baseline anxiety symptoms.

## Abstract

Background/Objectives: Insomnia disorder is highly prevalent and disabling, yet access to cognitive behavioural therapy for insomnia (CBT-I), the recommended first-line treatment, remains limited. Digital CBT-I (dCBT-I) offers scalable alternative; however, treatment outcomes vary according to intervention format and delivery context. This study evaluated whether delivering dCBT-I within a structured, medically supervised crenotherapy context improved insomnia symptom severity compared with stand-alone dCBT-I. Methods: In this multicentre proof-of-concept randomised controlled trial, 66 adults with insomnia disorder were allocated to receive either stand-alone dCBT-I (n = 38) or dCBT-I delivered within a 3-week standardised crenotherapy programme (medically supervised thermal spa treatment; n = 28). The primary outcome was change in Insomnia Severity Index (ISI) scores from pre- to post-treatment. Secondary outcomes included subjective sleep parameters (e.g., sleep efficiency and sleep onset latency), sleep-related functioning, pre-sleep arousal, anxiety and depressive symptoms. Engagement and satisfaction were assessed as additional descriptive outcomes. Results: Both groups showed significant improvements in insomnia severity, sleep parameters, and psychological symptoms. However, the primary between-group comparison did not demonstrate a statistically significant additive effect of crenotherapy on insomnia severity. ISI outcomes did not differ between the crenotherapy-delivered and stand-alone dCBT-I groups. Nevertheless, post hoc exploratory subgroup analyses suggested that, among participants younger than 60, delivery of dCBT-I within a crenotherapy care setting was associated with greater improvements in insomnia symptoms compared with stand-alone dCBT-I (mean ISI change: 10.4 vs. 5.4, p = 0.030). In a separate subgroup analysis, among participants with baseline anxiety symptoms, dCBT-I delivered within a crenotherapy care setting was associated with a greater reduction in anxiety compared with stand-alone dCBT-I (p = 0.030). Engagement and satisfaction were high in both groups, with no significant differences. Conclusions: Delivering dCBT-I within a crenotherapy context appears feasible and may offer specific benefits for specific subpopulations, particularly younger individuals and those with comorbid anxiety. These findings support further investigation of context-sensitive digital models to improve personalisation and accessibility of insomnia treatment.