Introduction

The pervasive use of screen-based digital media devices, during the day and evening hours, is accompanied by a high prevalence of insufficient sleep, affecting a majority of adolescents [1]. In this review article, we provide an updated overview of the current research on screens and sleep among adolescents. Due to space limitations and the changing nature and usage patterns of digital technology, we focus on articles published between 2015 and 2019. We begin by summarizing the high rates of both screen use and insufficient sleep among teenagers. Next, we review up-to-date evidence of the associations between screen use and sleep outcomes using observational studies. We follow with an appraisal of the more limited recent experimental studies on screens and sleep. We conclude with directions for future research, interventions, and policies.

High Levels of Screen Use and Poor Sleep Health Among Adolescents

During the last decade, the screen-based media landscape has changed markedly. According to the Pew 2018 report, 95% of American teens (aged 13–17) own or have access to a smartphone, which is a more than 20 percentage point increase from 73% in 2014–2015 [2]. As a result, 45% of teens reported being “almost constantly online.” Data from a 2016 nationally representative study of over 2600 US youth indicated that 47% of tweens and 57% of teens have TVs in their bedrooms [3]. Additionally, parent-reported data from the 2014 Sleep in America (SIA) poll by the National Sleep Foundation revealed that 75% of youth keep at least one type of electronic screen-based device in their bedroom [4].

Coinciding with the rise in digital media devices, two recent studies using US population-based data on teens show that compared with 1991, 58% more adolescents reported insufficient sleep in 2015. The authors concluded that the decreased sleep duration was primarily driven by use of portable screen devices [5••, 6]. Others have also argued that the ubiquitous prevalence of screen-based media devices (e.g., smartphones and tablets) is associated with poorer adolescent sleep health [7]. For example, the SIA poll found that the 57% of teenagers who leave an electronic device on in their bedroom after bedtime obtained less total sleep and had lower sleep quality compared with those who did not have a device in their room [4]. In fact, a recent US-based study found that 87% of 553 teenagers self-reported high levels of screen activities as a barrier to good sleep [8].

Screen Media Use and Sleep Outcomes in Observational Data

Several meta-analyses and reviews have documented a robust inverse association between screen-based media activities and adolescent sleep health outcomes [8,9,10,11,12,13]. Across a diverse range of socio-cultural backgrounds and levels of economic development, consistent empirical results have emerged from countries around the world, including the USA [6, 14], France [15], Mexico [16], Turkey [17], Brazil [18], and India [19]. Overall, these studies have shown that screen-based media device use is associated with a range of adverse sleep outcomes, including later bedtimes [20,21,22,23], later sleep midpoints [16, 24], longer sleep latencies [25, 26], insufficient sleep duration [27, 28], more insomnia symptoms [29••, 30], reduced sleep quality [31,32,33], lower sleep efficiency [34], and excessive daytime sleepiness [14, 35,36,37,38]. For example, one study using two US adolescent national datasets from 2010 to 2015 revealed a linear exposure-response association between screen devices use and increased odds of short sleep duration [5••]. Likewise, a Canadian study found a dose-response relationship between social media use and greater risk of short sleep duration, with much higher odds of insufficient sleep being observed among teens using social media for 5 or more hours per day compared with those using it for only 1 h per day [21]. In addition, parent-reported data from the 2016 National Survey of Children’s health found that every hour of digital screen time was associated with 3–8 fewer minutes of sleep duration and more irregular bedtimes during the weeknights [39]. While 3 to 8 min per hour may not seem like a large association, given the many hours that children spend in front of screens per day, this can add up to 40 min or more of shorter sleep duration if the child uses screens for 5 h per day or 80 min less sleep if the child uses screens for 10 h per day.

One meta-analysis investigated the association between portable screen-based media devices and sleep outcomes [13]. Merging results from 20 studies and over 125,000 youth, the authors consistently found that bedtime media usage is associated with insufficient sleep duration (OR = 2.17, p < 0.001), poor sleep quality (OR = 1.46, p < 0.01), and excessive daytime sleepiness (OR = 2.72, p < 0.01). The results of this meta-analysis also indicated that the mere presence of a portable screen-based media device in the bedroom has adverse associations with several sleep outcomes. Other recent studies support the case that the presence (not the use) of screen devices in the bedroom is associated with increased sleep problems [13, 40,41,42]. One study compared three groups of teens and found that longtime owners of smart phones slept significantly less and had poorer sleep quality than non-owners and new owners at the baseline [40]. Further, access to a smartphone in a teenagers’ bedroom is associated with shorter sleep duration [15], lower sleep quality, and less efficiency [43].

Although most of the aforementioned studies relied upon self-reported or parent-reported data, which are more prone to recall errors or bias, studies using objective measures of screen-based activities and assessing sleep via actigraphy [44, 45••, 46] found similar results. One recent study found that electronic media use accounted for 30% of all variance in adolescent sleep efficiency, as measured by actigraphy [45••]. Additionally, one Swiss study using data from mobile phone records revealed that teens who suffered from mobile phone related nocturnal awakenings had 3 to 5 times higher odds of restless sleep and insomnia symptoms [47].

Several longitudinal cross-lagged analyses have revealed a reciprocal relationship between screen media use and sleep problems. Results from 3 studies highlighted that more screen time at baseline was associated with shorter sleep duration, more sleep problems, and more daytime impairments in the follow-up wave, and baseline sleep problems were linked with higher consumption of screen time at follow-up 1 year later [47,48,49]. Adding evidence to support bidirectionality, adolescents report using media at bedtime to “help” them fall asleep [50]. Thus, there may be reverse causality in which the individuals who have trouble falling asleep turn to media devices as a strategy to either help them fall asleep or for entertainment.

Although the vast majority of recent observational studies find an inverse relationship between screen time and sleep health, several studies have failed to find an association. Studies examining screen time and sleep duration among both Korean high school boys [37] and Australian girls (8–16 years old) found that high-frequency social media interactions before bedtime were not significantly linked with reduced time in bed [51]. More research is needed to understand why these studies failed to find an association.

Does It Matter What Time of Day the Screen Is Used?

Although most studies examined total daily screen time as a predictor, data from studies assessing both daytime and pre-bedtime screen use indicate associations with poor sleep outcomes [32, 52]. However, the greatest effects on sleep come from studies assessing nighttime media use in the bedroom (i.e., in the 1–2 h before bedtime) [53, 54] or screen activities after lights-out [55] and the use of violent media at any time of the day [56, 57]. For example, one lab study found that video game use prior to bedtime leads to sleep loss and next day attention deficits among Australian adolescents [58]. Another experimental study revealed that prolonged violent gaming before bedtime caused decreases in objective sleep efficiency and total sleep time, as well as an increase in subjective sleep onset latency [59].

Are There Differences Across Types of Device?

There are varying results regarding whether the type, size, or how interactive a teen is with screens affects sleep outcomes. Some studies have found that interactive screen media use (e.g., video games and mobile devices) may have a greater impact on sleep than passive use, such as watching television [20, 60,61,62,63,64]. For example, one study found no association between TV, game console or computer use, and sleep problems, and only cell phone and MP3 player use in the bedroom (which are less subject to parental control) predicted teens’ sleep problems [32]. One meta-analysis found no association between television watching and sleep duration, while computer use was associated with shorter total sleep duration.

Experimental Data Provide Mixed Results

Although observational studies demonstrate direct associations between adolescents’ electronic media use before bed and insufficient sleep, relevant experimental research is limited and offers mixed results. Results from studies among young adults report that, compared with reading printed material, reading from a light-emitting e-book or tablet before bed results in greater melatonin suppression, less self-reported sleepiness before bed, later bedtimes, and later sleep onset times as measured with polysomnography [65••, 66, 67]. Additionally, a recent review of experimental research on the effects of video game exposure on sleep in adolescents and adults concluded that playing video games, particularly in the evening, negatively impacts sleep duration, sleep onset latency, and sleep architecture [68]. One possible mechanism underlying the relationship between screen-based media use and insufficient sleep is the alerting effect of short-wavelength blue light commonly emitted by mobile devices. In one study, young adults using a computer for 2 h in the evening had greater melatonin suppression, disrupted sleep, and daytime sleepiness when the computer emitted more short wavelength light [69]. In another study, adolescents showed more sensitivity to short wavelength light than adults, demonstrating greater melatonin suppression after exposure to light containing more blue light, while adults showed no differences in suppression to the different lighting conditions [70]. In contrast, teens using a tablet to read or complete puzzles for 2 h prior to bedtime demonstrated no differences in sleep quality or sleep duration compared with performing the same activities on paper [71], suggesting participants were unaffected by the light of the device. Additionally, one study with adults demonstrated that exposure to bright light during the day eliminates the impacts of nighttime use of light-emitting devices on melatonin [72]. Further experimental research is needed, however, to examine whether interactive contents on mobile devices, such as video games or social media, have a greater impact on adolescents’ sleep compared with more passive content, such as puzzles and reading. One recent study suggests the content of media use does play a role in whether sleep is disrupted [73]. Young adults assigned to using a smartphone in the evening to perform a breathing exercise through an application, compared with those browsing social media, reported improvements in subjective sleep quality.

From the sparse experimental findings currently available, it cannot be determined whether increased media use causes the observed sleep deficiencies reported in observational studies, or whether media use displaces poor sleepers’ time spent awake. In fact, in one study among preadolescents, time in bed was decreased by 1.5 h per night, and as a result, children reported watching significantly more television [74], suggesting the relationship between media use and sleep is bi-directional.

A variety of experiments have explored ways to limit adolescent nighttime media use to reduce negative impacts on sleep. An intervention instructing high school athletes not to use any electronic media after 22:00 had no measured impact on sleep timing or efficiency [75]. This finding is perhaps due to the exclusive sample of high school athletes studied, who may have already followed more routine and efficient sleep than non-athletes [76]. In addition, participants in the intervention group had an average weekday bedtime of 22:10, suggesting the intervention was too close to bedtime to be effective. Indeed, a different study demonstrated that adolescents who stopped using their mobile phones 1 h prior to average weekday bedtime turned off their lights earlier and slept longer [77••]; however, only 26% of students invited to participate contributed data, suggesting low motivation to restrict their phone use. Additionally, a recent focus group of adolescents and young adults reported that although subjects expressed willingness to change their behaviors to improve sleep, they find it difficult to disconnect from their devices at night, due in large part to social media use and fear of missing out (FOMO) [78, 79]. For example, a recent study had found that FOMO is one of the pathways that drive teens’ late night social media use and poor sleep [26].

For teens reluctant to reduce their media use, wearing “blue-blocker” glasses to filter out short wavelength light mitigated melatonin suppression and alertness when using light-emitting devices at night, although no differences in polysomnography or self-reported sleep quality were observed [80••, 81]. In addition, using a tablet for 1 h before bedtime that emits light with short wavelengths filtered out, compared with dim or bright light, leads to no differences in sleep onset latency or architecture [82].

Several studies have explored the efficacy of using school-based interventions to provide students with information on the importance of sleep and good sleep hygiene habits. One such intervention leads to a significant decrease in nighttime electronic media use, but no impact on self-reported sleep duration or daytime tiredness [83]. In contrast, another study reported that their intervention resulted in improvements in objective sleep quantity and quality, but no differences in technology use before bedtime [84]. Yet, a follow-up study with the same intervention yielded no impact on self-reported media use before or during sleep or sleep hygiene [85]. Together, these studies highlight the need for further research to determine what information such interventions could provide to be most effective for adolescents.

Potential Impacts of Screen Time on Sleep

Screen-based media use has been linked to a range of adverse developmental outcomes among adolescents, including poor academic performance [86, 87], low school satisfaction [88], and physical, behavioral, and mental health problems [27, 28, 89]. For example, national survey data of US 8th and 10th graders revealed that watching TV on a weekday was negatively associated with academic outcomes, and students with lower grades used technology and interactive social media more often than their peers with higher grades [86]. Cross-sectional data from Chinese adolescents highlighted a positive correlation between problematic smartphone use and physical symptoms such as back/neck pain and daytime fatigue [90].

One of the primary pathways through which screen-based digital media may affect adolescent development is through impaired sleep health. Indeed, studies have identified the mediating role of sleep problems in the association between screen media use and a range of negative adolescent outcomes [27, 28, 90,91,92,93]. For example, data from a diverse sample of US urban birth cohorts found that sleep duration and insomnia symptoms mediated the association between screen activities and adolescent depressive symptoms, even after adjustment for childhood depressive symptoms [29••]. Similar results have also been noted in teens from Switzerland [28] and China [93]. Two longitudinal studies of Australian teens also confirmed that nighttime screen media use was associated with lower psychosocial well-being, primarily through poorer perceived sleep quality and increased sleep problems [91, 94••]. In contrast, another study found no significant mediating role of sleep duration or disturbance in the relationship between late media use and increased depression and suicidality [95].

Future Directions in Research, Interventions, and Policies

Current empirical research demonstrates that digital media is adversely associated with a range of sleep health outcomes among teens. Limited data, however, indicate a direct causal mechanism of this robust association. Further, very few studies demonstrate easy-to-implement and effective interventions to modify digital media use, especially in the evening, to improve sleep among teens. One novel but ineffective policy on adolescent screen use and sleep was implemented by the South Korean government when they started banning juvenile online gaming from 12 a.m. to 6 a.m. in 2011 [96]. Interestingly, results from this national shutdown initiative indicated no practical effects in altering problematic Internet use [97] among middle school students (age 13–18 years) and only increased youth’s sleep duration by about 2 min [98].

Conclusions

Based upon this review of the literature, it is clear that more experimental research is needed to better understand the mechanisms underlying the widely observed association between screen time and a range of sleep health outcomes. Understanding the causal mechanisms will aid in the necessary development, implementation, and evaluation of sustainable interventions that minimize the potential adverse effects of evening screen use on sleep. Furthermore, more basic, translational, and clinical studies are necessary to investigate the effects of screen-based media on sleep and related health consequences among adolescents. Such efforts are essential in order to educate and motivate parents, clinicians, teachers, and teens to develop healthier sleep habits.