JMIR Publications

ABSTRACT

Background:

A third of adults in Western countries have impaired sleep quality and widely accessible sleep-interventions are limited. One solution is distribution of sleep aids through smartphone applications, but most empirical studies are limited to small pilot trials in distinct populations (e.g. soldiers) or individuals with clinical sleep disorders so general population data are required. Furthermore, recent research shows sleep app users desire a personalised approach, offering an individually tailored choice of techniques. One such aid is ‘Peak Sleep’, a smartphone application based on scientifically validated principles for improving sleep quality, such as mindfulness meditation and cognitive behavioural therapy.

Objective:

To test the impact of the smartphone app ‘Peak Sleep’ on sleep quality and collect user experience data to allow for future app development.

Methods:

This study was a two-arm pilot randomised control trial. Participants were general population UK adults (>18 years old) who were interested in improving their sleep quality, not undergoing clinical treatment for sleep disorder or using sleep medication >1 per week. Participants were individually randomised to receive the intervention (3 months of app use) versus control. The intervention involved free access to Peak Sleep – an app that offered a choice of behavioral techniques to support better sleep (mindfulness, cognitive behavioural therapy, acceptance commitment therapy). The primary outcome was sleep quality assessed using the Insomnia Severity Index, at 0, 1, 2 and 3 month follow-ups. Objective sleep data collection was planned (Oura ring) however, the COVID-19 pandemic lockdowns began just after recruitment started, so this was not possible. Participant engagement with the app was assessed using the Digital Behaviour Change Intervention Engagement Scale and qualitative telephone interviews with a subsample.

Results:

101 participants were enrolled in the trial and 21 were qualitatively interviewed. Sleep quality improved in both groups over time, with ISI scores of the intervention group improving by a mean of 2.5 and control by 1.6, but there was no significant effect of group (p>0.05). App users’ engagement was mixed, with qualitative interviews supporting this view of a polarised sample who either really liked or disliked the app.

Conclusions:

In this trial, self-reported sleep improved over time in both intervention and control arms, with no impact by group, suggesting no effect of the sleep app. Qualitative data suggested polarised views on liking vs not liking the app, features that people engaged with, and areas for improvement. Future work could involve developing the app features integrating these, then testing the app using objective measures of sleep in a larger sample .This study found no impact of a sleep app on self-reported sleep quality in a general population sample of poor sleepers. Important user experience were collected and the app’s tailored nature was well-received. Therefore, future research could evaluate the app’s efficacy after integration of suggested improvements and using objective measures of sleep. Clinical Trial: ClinicalTrials.gov (ID: NCT04487483)