Pilot data suggests that working professionals and college students routinely use alarms and snooze. Alarm usage and snoozing is associated with several negative health biomarkers including lighter sleep, higher resting heart rate, and reduced sleep duration. It is unclear when this behavior is established, but it is likely in the teenage years when chronic sleep restriction begins to effect a large percentage of Americans. We will ask teens about psychological traits (e.g. personality) and snoozing behavior in a repeated measures design. In addition, we will implement a smartphone based intervention which notifies teens when they are awake past their minimum bedtime for adequate sleep. throughout the study, we will monitor sleep and heart-rate via wearable. From this data, we will establish the prevalence of alarm and snoozing behaviors in teens. We will determine what demographic, psychological, and behavioral traits predict snoozing, and if there are any differences in health biomarkers (e.g. sleep duration, resting heart rate)between snooze and/or alarm users. We will use data from the wearables and smartphones to generate features that can detect snoozing, and will validate them against self-report. Finally, we seek to determine if alarm and snoozing behavior can be reduced via a smartphone intervention aimed at increasing sleep duration.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
OTHER
Masking
NONE
Habitual bed and wake times will be detected using a wearable for 4 weeks. Using habitual wake time, a minimum bedtime to achieve adequate sleep will be calculated (8 hours). If phone usage is detected after this minimum bedtime, participants will receive a notification informing them they are awake past their minimum bedtime and recommend they go to bed to obtain adequate sleep.
Sleep Duration
Reduced sleep duration is associated with negative health outcomes. We will compare average baseline and intervention sleep durations (measured by wearables) using 1-way ANCOVA. Covariates will be significant factors from a bivariate logistical regression predicting the proportion of days woken to an alarm or snoozing from total answered survey using demographic, behavioral, psychological, and physiological factors as predictors. If the ANCOVA reveals a significant effect of intervention, a followup ANCOVA will compare average sleep duration for the first two weeks of intervention to the last two weeks of intervention using same covariates to determine if intervention effects change over time. Another follow correlation will compare the number of times the intervention occurred with the average difference of baseline and intervention sleep duration.
Time frame: 12 weeks, daily sleep measured passively via wearable
Resting Heart Rate
Higher resting heart rate (RHR) rate is associated with negative health outcomes. We will compare average baseline and intervention RHR (measured by wearables) using 1-way ANCOVA. Covariates will be significant factors from a bivariate logistical regression predicting the proportion of days woken to an alarm or snoozing from total answered survey using demographic, behavioral, psychological, and physiological factors as predictors. If the ANCOVA reveals a significant effect of intervention, a followup ANCOVA will compare average RHR for the first two weeks of intervention to the last two weeks of intervention using same covariates to determine if intervention effects change over time. Another follow correlation will compare the number of times the intervention occurred with the average difference of baseline and intervention RHR.
Time frame: 12 weeks, Heart rate measured every minute passively via wearable
Alarm and Snooze Self Report
Alarm and Snoozing (AS) behaviors are associated with negative health outcomes. We will compare average baseline and intervention AS proportions using 1-way ANCOVA. Covariates will be significant factors from a bivariate logistical regression predicting the proportion of days woken to an alarm or snoozing from total answered survey using demographic, behavioral, psychological, and physiological factors as predictors. If the ANCOVA reveals a significant effect of intervention, a followup ANCOVA will compare AS proportion for the first two weeks of intervention to the last two weeks of intervention using same covariates to determine if intervention effects change over time. Another follow correlation will compare the number of times the intervention occurred with the average difference of baseline and intervention AS proportion.
Time frame: 6 weeks via survey. 2 weeks at beginning of study (BASELINE), 2 weeks at beginning of intervention, 2 weeks at end of intervention
Reaction Time
Sleep inertia is a cognitive impairment measured using reaction time (RT) following sleep. We will compare average baseline and intervention RTs using 1-way ANCOVA. Covariates will be significant factors from a bivariate logistical regression predicting the proportion of days woken to an alarm or snoozing from total answered survey using demographic, behavioral, psychological, and physiological factors as predictors. If the ANCOVA reveals a significant effect of intervention, a followup ANCOVA will compare RTs for the first two weeks of intervention to the last two weeks of intervention using same covariates to determine if intervention effects change over time. Another follow correlation will compare the number of times the intervention occurred with the average difference of baseline and intervention RTs.
Time frame: 6 weeks via survey. 2 weeks at beginning of study (BASELINE), 2 weeks at beginning of intervention, 2 weeks at end of intervention
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.