Participants in this condition will receive an access code to download the mobile application and conduct an orientation to SHIFT and the study procedures with a study team member. They will then enter four weeks of study directed use of SHIFT where they are trained on a daily procedure of opening SHIFT at the beginning of their day and planning their day in accordance with the app recommendations. App usage (4 times per week) will be incentivized with weekly bonuses added to study compensation. Following the four weeks of study-directed use, participants will continue with self-directed use. At four months, participants will then complete a booster session with two additional weeks of study-directed use, followed by self-directed use until the end of the study. The structure with the booster session has evidence for maintaining treatment gains and is also in alignment with the commercialization plan.

Participants in this condition will only complete questionnaires for the initial eight months. Those who remain shift workers will have the option of receiving the SHIFT app and the option to complete follow-up surveys in the same manner as the SHIFT App condition.

SHIFT is a mobile application designed to improve sleep in night shift workers. The SHIFT mobile application is used to collect data from an Apple Watch to assess an individual shift worker's body-clock timing and make personalized recommendations of light exposure schedules that are designed to align the body-clock with the night shift work schedule.

Aim 1a. Measure the effect of SHIFT on work productivity and satisfaction compared to waitlist control using the Job Satisfaction Index. Effect will be tested using a mixed-effects linear regression model with participants as the random effect and Time, Condition, and the Time × Condition interaction term as the fixed effects.

Aim 1b: Measure the effect of SHIFT on global health compared to waitlist control using the NIH PROMIS Global Health questionnaire. Effect will be tested using the same method as Aim 1a.

Aim 1c. Measure the effect of SHIFT on turnover compared to waitlist control, measured at 8-month follow-up. Turnover will be operationalized as an individual who has either terminated the position they were in at baseline or is no longer engaged in shift work as operationalized in the study. Effect will be determined using a generalized mixed-effects regression with turnover as a dichotomous outcome.

Aim 2a. Measure user experience of Android and iOS versions of SHIFT using the User Experience Questionnaire (UEQ). The following ranges of clinical indifference will be used: ± 3 points on the User Experience Questionnaire based on the bin size of 6 for each of the thresholds (bad, neutral, and good user experience).

Aim 2b. Measure accuracy of predicted circadian misalignment (CM), sleep, and depression in Android and iOS versions. CM will be indexed with the outputs of the biomathematical model of the circadian system. Sleep will be measured using the Insomnia Severity Index and sleep diaries. Depression will be measured using the Quick Inventory of Depressive Symptomatology. The following ranges of clinical indifference will be used: 1) predicted CM = ± 3 hours based on approximately 2x the absolute mean error of our model predictions, 2) insomnia severity = ± 6 points, and 3) depression = 28.5% of the QIDS-SR16 score.

A series of semi-structured interviews will be used for thematic analysis, and a comprehensive roadmap for future app updates based on user feedback. The semi-structured interviews will utilize the interview-guide approach following the CFIR framework selected for this study. Six phases will be followed for thematic analysis: (1) data familiarization, (2) generating initial codes, (3) searching for themes, (4) reviewing themes, (5) defining and naming themes, and (6) producing the report. We will combine deductive and inductive techniques to increase the accuracy of thematic analyses.