By William Haas, MD
Carolinas Medical Center,
Department of Family Medicine,
Charlotte, NC

Dr. Haas reports no financial relationships relevant to this field of study.

SYNOPSIS: The use of light-emitting tablet devices prior to bedtime negatively impacts sleep patterns through disturbances in circadian rhythms.

SOURCE: Chang AM, et al. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. Proc Natl Acad Sci U S A 2015;112:1232-1237.

Summary Points

  • Tablet reading prior to bed increased time-to-sleep onset by 10 minutes compared to reading from a print book.
  • Mornings after tablet reading were associated with increased levels of perceived sleepiness and decreased alertness.

With the advancement of technology, the printing press is slowly giving way to digital media, and consumers are spending increasing amounts of time assimilating knowledge from electronic devices. The tablet reader, a popular device for reading and entertainment, is frequently used in the evening hours prior to bed. Some researchers are concerned that such light-emitting devices may disrupt natural circadian rhythms and perpetuate sleep deficiencies. Through a randomized, crossover study, researchers at Brigham and Women’s Hospital set out to determine the effect of light-emitting tablets on sleep patterns as compared to printed books when used prior to bed.

Twelve young adults (six females and six males) were recruited to live and sleep in separate light-controlled rooms for 14 days. Recruits were carefully screened with the following exclusion criteria: pre-existing chronic medical or psychological condition(s), pre-existing sleep disorder, use of prescription medication(s), history of night/shift work in the prior 3 years, travel across more than one time zone in the previous 3 months, or any vision abnormality preventing reading in a dimly light room. Three weeks prior to the study, the selected participants were instructed to refrain from use of medications, drugs, alcohol, nicotine, or caffeinated products, which was verified by toxicological testing. Additionally, participants were required to maintain a fixed 8-hour sleep schedule (10 p.m. to 6 a.m.) during the 3-week pre-study period, as verified by sleep/wake logs and wrist actigraphy.

After the initial 3-week acclimation period, participants were randomized to either: 1) reading from a light-emitting tablet (iPad; Apple Inc.) in a dimly light room for approximately 4 hours before bedtime or 2) reading from a printed book under similar dimly lit conditions for the same duration of time prior to bed. Each reading session took place for five consecutive evenings, after which participants crossed over to the alternative reading condition; the initial order was randomly assigned. The night prior to each 5-night reading block, as well as the final night of the study, participants completed a constant posture procedure, which consisted of sitting in a semirecumbent position for 4 hours before bed, with minimal activity and no reading. At the end of each reading session, sleeping periods occurred according to the same 8-hour block prescribed in the acclimation period (10 p.m. to 6 a.m.).

Details regarding the structure of reading sessions and lighting conditions were highly standardized and extensively outlined in the methods section. In short, tablets were set to the maximal brightness setting and placed stationary at a distance 30-45 cm from the participants’ eyes. Books, on the other hand, were held at any distance desired and were restricted to printed text only (i.e., no pictures or illustrations) and were limited to topics of pleasure or leisure (i.e., no educational material). Participants were instructed to read while seated in a fixed location in their rooms. Approximately 3 hours into the reading sessions, participants were given a 15-minute break and completed the final hour of reading while seated in bed. Adherence to the treatment protocol was overseen by a trained technician.

Primary outcomes measured included sleep latency, timing and amount of melatonin secreted, and measures of sleepiness. Sleep latency was assessed using polysomnography (PSG) recordings on the fourth and fifth night of each reading block. Three other sleep measures, including total sleep time, sleep efficiency, and time spent in each sleep stage, were also obtained using PSG recordings. Melatonin was measured from blood samples collected hourly throughout the night via an indwelling IV catheter in the forearm. Both degree of melatonin suppression and time to dim light melatonin onset were calculated from samples drawn. Of note, due to lost blood samples, melatonin suppression and onset calculations were determined from 11 participants. Finally, subjective sleepiness measures were recorded using the Karolinska sleepiness scale (KSS) every evening and morning. Objective sleepiness was measured using electroencephalogram (EEG) measurements recorded on the final two evenings and mornings of each reading block.

As mentioned above, a total of six males and six females completed the study with an average age of 24.92 ± 2.87 years — no other demographic information was provided. With regard to sleep patterns, reading from a light-emitting tablet increased time-to-sleep onset by 10 minutes compared to reading from a print book (25.65 ± 18.78 min vs 15.75 ± 13.09 min; P = 0.009). Tablet reading also resulted in less rapid eye movement (REM) sleep compared to book reading (109.04 ± 26.25 min vs 120 ± 25.32 min, P = 0.03), but no difference was noted between total sleep time, sleep efficiency, or duration of non-REM sleep. Overall levels of melatonin were suppressed more while reading a tablet compared to reading a book (55.12 ± 20.12% vs -18.77 ± 39.57%, P < 0.001), with melatonin onset also occurring > 1.5 hours later on the day following the tablet sessions (P < 0.001). Both subjective KSS scores and EEG measurements revealed participants to be less sleepy within the hour before bedtime compared to book reading (P < 0.01). Similarly, participants felt sleepier in the morning following an evening of tablet reading compared to book reading (P < 0.001).


Through a well-designed and rigorous study, researchers at Brigham and Women’s Hospital clearly demonstrate that the use of light-emitting tablets during the hours prior to bedtime negatively impacts sleep patterns. More specifically, reading from a light-emitting tablet decreases perceived sleepiness before bed and increases time-to-sleep onset, likely as a result of an acute alerting effect from light exposure and a delay in the circadian timing system. Additionally, tablet reading increases sleepiness in the mornings after use, which may be attributed to a decrease in REM sleep and a higher likelihood of spontaneous awakenings from deeper phases of sleep.

The results of the present study, while intriguing unto themselves, should be considered in the context of their general health implications. Despite the fact that melatonin secretion occurred 1.5 hours later with tablet reading, time-to-sleep onset only increased by 10 minutes compared to book reading. The researchers suggest that this may increase the risk of sleep-onset insomnia; however, if the time-to-sleep onset does not continue to lengthen with prolonged tablet use, the additional 10 minutes required to fall asleep is unlikely to result in a higher demand for hypnotics. On the other hand, the delayed secretion of melatonin does seem concerning for the development of a delayed sleep-phase disorder, thereby leading to a chronic sleep deficiency. Most concerning, perhaps, is the overall reduction in melatonin secretion while using a light-emitting tablet before bed. Chronic suppression of melatonin has been associated with an increase in several cancers, including breast, colorectal, and prostate cancer.1 The important, yet unanswered question remains whether or not the degree of melatonin suppression with tablet use is large enough to confer any long-term health consequences. In the short-term, however, tablet use prior to bed may pose some safety concerns, especially with increased sleepiness and decreased alertness on the mornings after use. Decreased levels of morning alertness would certainly be hazardous in many professions (public transportation, medicine, construction, etc.).

Before settling on any final conclusions, the overall design of the study should be reviewed. The study was thoughtfully constructed with appropriate randomization and crossover between groups. A pre-intervention acclimation period attempted to minimize differences between the groups at baseline. The final sample size was arguably small, but was likely a result of the highly controlled environment required to ensure proper standardization of light exposure. The rigorous design of the study does raise some questions regarding generalizability. The exclusion criteria were so strict that the participants in the study are unlikely to represent the average tablet user, especially among those presenting in the clinic with complaints of a sleep disorder. Moreover, the average American’s sleep schedule is not standardized to a set 8-hour period every night. Similarly, not every person remains fixed to their tablet for nearly 4 uninterrupted hours prior to bedtime. In a non-laboratory environment, tablet users also have greater control over settings, including brightness levels and the distance the device is held from a user’s eyes.


  1. Costa G, et al. Shift work and cancer — considerations on rationale, mechanisms, and epidemiology. Scand J Work EnvironHealth 2010;36:163-179.