JMIR Publications

ABSTRACT

Background:

Shift workers are at a high risk of developing sleep disorders such as shift worker sleep disorder or chronic insomnia. Cognitive behavioral therapy (CBT) is the first-line treatment for insomnia, and emerging evidence shows that internet-based CBT is highly effective with additional features such as continuous tracking and personalization. However, there are limited studies on internet-based CBT for shift workers with sleep disorders.

Objective:

The aim of this study was to evaluate the effect of a 4-week physician-assisted internet-delivered CBT program for shift workers that implements machine learning–based well-being prediction; this study focused on the sleep duration of shift workers at high risk for sleep disorders. We evaluated these outcomes using an app and fitness trackers in the intensive care unit (ICU).

Methods:

Participants were eligible if they were on a three-shift schedule and had a Pittsburgh Sleep Quality Index of >5. Prior to the study, participants were asked to complete questionnaires regarding the subjective evaluation of sleep, burnout syndrome, and mental health. The baseline period was 1 week, followed by an intervention period of 4 weeks. The internet-delivered CBT program included “well-being prediction”, “activity/sleep chart”, and “sleep advice”. A job-based multitask and multilabel convolutional neural network–based well-being prediction model was used. Participants were asked to wear a consumer fitness tracker to track their daily activities and sleep. The primary endpoint of this study was sleep duration. For continuous measurements (sleep duration, steps, etc.), the mean baseline and mean of the fourth week of intervention were compared. The paired t-test or Wilcoxon signed-rank test was performed depending on distribution of the data.

Results:

A total of 61 shift workers in the ICU or emergency department aged 21 to 55 years (mean 32.9 years, SD 8.3 years) participated. In the fourth week of intervention, the mean daily sleep duration for 7 days (6.15 h) showed a statistically significant improvement compared to baseline (5.86 hours, P = .023). Subjective sleep quality, as measured by the Pittsburgh Sleep Quality Index, also showed remarkable improvement from baseline (8.74) after the intervention (7.84, P = .003). There was no significant improvement in the subjective well-being scores. Feature importance analysis for all 45 variables in the prediction model showed that sleep duration had the highest importance.

Conclusions:

The physician-assisted internet-delivered CBT program targeting shift workers with a high risk of sleep disorders showed significant improvement in sleep duration as measured by wearable sensors along with subjective sleep quality. This study implies that sleep improvement programs using an app and wearable sensors are feasible and may play an important role in preventing shift-work-related sleep disorders.