Table 1.

Study characteristics.

Study author and year	Type of study	Country	Sample	Age	Age group	Exposure	Exposure description	Sleep outcomes	Findings	Covariates included in analysis
Ahn et al., 2016 [43]	Cross-sectional	Korea	N = 1033	Age range: 0–36 mo	Infants and toddlers	Evening screen time	Parent reported television or video	Sleep duration, bedtime; night awakenings	TV at sleep initiation was associated with a later bedtime (β = 0.30). TV at sleep initiation was not significantly associated with any of the other sleep outcomes (direction of association not reported).	child demographic variables (age, sex, birth order), parental demographic variables (age, educational level, employment status), parental behaviors at bedtime, and other aspects of the sleep ecology (sleep arrangement, location, position)
Cespedes et al., 2014 [17]	Cross-sectional	USA	N = 6 mo: 1673; 1 y: 1227; 2 y: 1360; 3 y: 1242; 4 y: 1202	Age range: 6 mo - 4 y	Infants, toddlers and preschoolers	Total daily screen time	Parent reported television	Sleep duration	Higher TV time was associated with shorter sleep duration at ages 6 mo (β = −3.0; 95% CI, −8.0 to 2.0); 1 (β = −6.0; 95% CI, −9.0 to −2.0), 2 (β = −6.0; 95% CI, −10.0 to −2.0); 3 (β = −2.0; 95% CI, −6.0 to −2.0); and 4 (β = −4.0; 95% CI, −8.0 to 0.0) y.	child age in years at time of assessment, race/ethnicity, gender, maternal education, and household income; age 4 analysis additionally adjusted for TV in bedroom.
Chonchaiya et al., 2017 [30]	Longitudinal and cross-sectional	USA	N = 208	Mean age: 6.2 mo (time 1); 12.3 mo (time 2)	Infants and toddlers	Total daily screen time, evening screen time	Parent reported use of all electronic media	Sleep duration; sleep latency; night awakenings; naptime duration	Higher levels of total daily screen time at age 12 mo was associated with longer sleep latency at age 12 mo (β = 0.16 for weekday; β = 0.17 for weekend day). Total daily screen time at 6 mo was associated with longer sleep latency at 6 mo (during weekends). Total and evening screen time for 6 and 12 mo of age was not significantly associated with any other sleep outcomes at age 12 mo (direction of association not reported). Bedroom media use at 12 mo was not significantly associated with sleep latency at 12 mo (direction of association not reported).	age, gender, co-sleeping status, evening media use at age 12 mo, maternal education, and household income were included in the final regression models as covariates.
De Bock et al., 2013 [36]	RCT	Germany	N = 809	Mean age (SD): 5.05 y (0.2) Age range: 4–6 y	Preschoolers	Total PA, SB, MVPA	Accelerometry	Sleep quality	A trend toward improved subjective sleep quality in the intervention group was noted (β = −0.113; 95% CI, −0.003 to 0.23).	Intention-to-treat basis. The core model assumed a linear change of the outcomes with time and included two normally distributed random effects (one at the preschool level and one at the child level) to adjust for clustering in the data due to the hierarchic sampling scheme. Further, all models included the variables age, gender, rural versus urban community of preschools, and season as covariates to adjust for a potential confounding effect of these variables.
Duraccio et al., 2017 [31]	Cross-sectional	USA	N = 131	Mean age (SD): 4.9 y (0.5)	Preschoolers	SB, MVPA, VPA	Accelerometry	Sleep duration	For each added day of high sedentary behavior (i.e., being in top tertile of sedentary behavior), the probability of obtaining sufficient sleep decreased (1 d = 0.56; 95% CI, 0.26–0.75; 2 d = 0.51; 95% CI, 0.37–0.65; 3 d = 0.22; 95% CI, 0.11–0.33). MVPA and VPA were not associated with sleep duration.	Interaction with sex tested, ns
Garrison et al., 2011 [32]	Cross-sectional	USA	N = 617	Mean age (SD): 51 mo (8)	Preschoolers	Total daily screen time, evening screen time	Parent reported use of all electronic media	Sleep quality	Total screen time was associated with higher sleep problem scores (β = 0.244; 95%CI, 0.113 to 0.375). Each additional hour of evening screen time was associated with increases in sleep problem scores (β = 0.743; 95% CI, 0.373 to 1.114). Each additional hour of day screen time was associated with increases in sleep problem scores (β = 0.107; 95% CI, −0.047 to 0.260).	child gender, low-income status, single-adult household, and SCBE (Social Competence and Behavior Evaluation) internalizing and externalizing scores, as well as which parent completed the survey (mother versus other), each additional hour of nonviolent daytime media time, and each additional hour of violent daytime media time
Genuneit et al., 2018 [37]	Cross-sectional	Germany	N = 530	Approximate age: 3 y	Preschoolers	Total daily screen time	Parent reported use of all electronic media, TV/DVD, computer/internet use, computer gaming	Sleep habits	Total daily screen time, TV/DVD time, computer/internet use and computer gaming were associated with inconsistent sleep habits.	NA
Hager et al., 2016 [33]	Cross-sectional	USA	N = 240	Mean age: 20.2 mo	Toddlers	MVPA	Accelerometry	Sleep duration, Sleep quality	MVPA was associated with longer sleep duration (β = 0.332; SE, 0.138). Those with high sleep behavior scores (5–6) spent significantly more time in MVPA (65.3 min) compared to those with mid-range sleep behavior scores (3–4; 45.3 min) but not those with low scores (0–2; 58.3 min). Those with high sleep behavior scores (5–6) had significantly higher counts per minute (433.1 cpm) compared to those with mid-range sleep behavior scores (3–4; 348.8 cpm) but not those with low scores (0–2; 409.2 cpm).	NA
Hauck et al., 2018 [27]	Cross-sectional	USA	N = 22	Approximate age: 6mo ± 1 wk	Infants	SB	Sedentary behavior and screen time	Sleep duration; night awakenings; daytime napping	More time in SB was significantly associated with less total sleep (r = −0.524) and non-significantly associated with less night time sleep (r = −0.417), more time awake at night (r = 0.308), reduced nap duration (r = −0.104), reduced nap frequency (r = −0.068)	NA
Ikeda et al., 2012 [29]	Cross-sectional	Japan	N = 39,813	Approximate age: 4.5 y	Preschoolers	Total daily screen time	Parent reported television; computer games	Sleep duration, daytime napping	Those watching more hours of television were more likely to have shorter (<10hr) sleep durations (OR not playing = 1.0; <1hr = 1.01; 95% CI, 0.67 to 1.52; 1–2hr = 1.06; 95% CI, 0.71 to 1.58; 2–3hr = 1.37; 95% CI, 0.92 to 2.04; 3–4hr = 1.55; 95% CI, 1.04 to 2.33; ≥4hr = 1.91; 95% CI, 1.26, 2.90). Playing computer games was unfavorably associated with sleep duration (not playing = 1; <1hr = 1.11; 95% CI, 1.02 to 1.21; 1–2hr = 1.14; 95% CI, 0.98 to 1.32; ≥2hr = 1.62; 95% CI, 1.18 to 2.23. Playing computer games was associated with less daytime napping (not playing OR = 1; <1hr = 0.85; 95% CI, 0.78 to 0.92; 1–2hr = 0.80; 95% CI, 0.69 to 0.92; ≥2hr = 0.57; 95% CI, 0.40 to 0.82 Television time was associated with more daytime napping (not watching = 1; <1hr = 1.07; 95% CI 0.75 to 1.53; 1–2hr = 1.15; 95% CI, 0.81 to 1.64; 2–3hr = 1.16; 95% CI 0.82 to 1.65; 3–4hr = 1.21; 95% CI, 0.85–1.72; ≥4hr = 1.22; 95% CI, 0.85 to 1.77	regional population, gender, existence of older siblings, years of maternal and paternal education, hours spent watching, television, hours spent playing computer games, paternal and maternal work hours, and whether or not the child attended preschool, or a childcare center
Iwata et al., 2011 [44]	Cross-sectional	Japan	N = 48	Approximate age: 5 y	Preschoolers	Sports participation	Parent reported sports participation	Sleep onset; sleep end time; sleep latency; sleep efficiency	Sport lesson attendance was associated with earlier sleep onset on weekdays (B = −0.258; 95% CI, −0.728 to 0.043) and later onset on weekends (B = 0.096; 95% CI, −0.391 to 0.760). Sport lesson attendance was associated with earlier sleep end on weekdays (B = −0.342; 95% CI, −0.641 to −0.062) but not weekends (B = 0.086; 95% CI, −0.331 to 0.598). Sport less attendance was associated with longer sleep latency (B = 0.318; 95% CI, 0.393, 7.149 for weekdays; B = 0.307; 95% CI, 0.245 to 6.921 for weekends). Sport less attendance was associated with higher sleep efficiency (B = 0.318; 95% CI, 0.393 to 7.149)	NA
Ji et al., 2018 [45]	Cross-sectional	China	N = 112	Age range: 3–6 y	Preschoolers	MVPA; screen time	Accelerometry; parent reported electronic media use	Sleep duration	Those engaging in more MVPA (OR = 0.735; 95%CI, 0.189 to 2.855) or daily screen time (OR = 0.380; 95%CI, 0.107 to 1.348) were less likely to get sufficient sleep (8–13hr).	Age, gender, father's BMI, mother's BMI, total physical activity time, daily steps and daily sedentary time.
Krejci et al., 2011 [46]	Cross-sectional	Czech Republic and Japan	N = 1096; Czech Republic: 497; Japan: 599	Mean age: Czech Republic: 4.6 y (1.1); Japan: 3.8 y (1.2)	Preschoolers	Total daily screen time	Parented reported use of computer games	Sleep duration; bedtime	Frequency of playing computer games was not associated with sleep duration but was associated with later bedtime. Duration of playing computer games was not associated with sleep duration but was associated with later bedtime in Czech children but not Japanese. Time of the day of playing computer games was associated with shorter sleep duration bedtime in Czech children but not Japanese.	NA
Magee et al., 2014 [50]	Longitudinal and cross-sectional	Australia	N = 3427	Age range: 4–5 y (time 1); 6–7 y (time 2)	Preschoolers	Total daily screen time	Parent reported television and video use; computer use; total screen time	Sleep duration	Total screen time at age 4 was associated with shorter sleep duration at age 6 (B = −0.06; 95% CI, −0.10 to −0.02) TV/video viewing at age 4 was associated with shorter sleep duration at age 6 (B = −0.05; 95% CI, −0.09 to −0.01). Computer use at age 4 was unfavorably associated with sleep duration at age 6 (B = −0.10; 95% CI, −0.21 to 0.01). Total screen time at age 4 was associated with shorter sleep duration at age 4 (B = −0.10; 95% CI not reported).	child's sex, baseline obesity status, sleep problems, household income, and maternal education
Marinelli et al., 2014 [38]	Cross-sectional and longitudinal	Spain	N = 1202 (time 1); 1090 (time 2)	Approximate age: Time 1: 2 y, Time 2: 4 y	Toddlers	Total daily screen time	Parent reported television time	Sleep duration	Children with longer periods of television viewing at age 2 (≥1.5 h per day) had shorter sleep duration and each additional hour of television viewing decreased sleep duration (β = −0.13; 95% CI, −0.19 to −0.08). Children with longer periods of television viewing at age 2 (≥1.5 h per day) had shorter sleep duration at age 4 and each additional hour of television viewing decreased sleep duration (β = −0.11; 95% CI, −0.18 to −0.05).	BMI at baseline, BMI change, parental educational level, sex
McDonald et al., 2014 [39]	Cross-sectional	United Kingdom	N = 1702	Mean age: 15.8 mo	Toddlers	Evening screen time	Parent reported television time	Sleep duration	Children with more >1hr morning television viewing had an increased risk of short sleep duration (<11hr; OR = 1.13; 95% CI, 0.80 to 1.58). Children with more >1hr evening television viewing had an increased risk of short sleep duration (<11hr; OR = 1.89; 95% CI, 1.26 to 2.84).	Maternal education, ethnicity, sex, birth weight, older children around, evening TV, age, daytime sleep, regular night waking.
Mindell et al., 2013 [53]	Cross-sectional	Australia, New Zealand, Canada, United Kingdom, United States, China, Hong Kong, India, Japan, South Korea, Malaysia, Philippines, Singapore, Thailand	N = 2590; Australia and New Zealand: 286; Canada: 272; United Kingdom: 298; United States: 284; China: 248; Hong Kong: 82; India: 294; Japan: 48; South Korea: 312; Malaysia: 121; Philippines: 76; Singapore: 81; Thailand: 88)	Age range: 3–6 y	Preschoolers	Total daily screen time;	Parent reported television, computer or electronic game use	Sleep duration; sleep latency; bedtime; night awakenings;	More screen time was associated with longer sleep latency (r = 0.11), later bedtime (r = 0.21), more night awakenings (r = 0.07) and longer night time sleep duration (r = 0.08). More screen time was not significantly associated with duration of night awakenings, total sleep duration, or daytime sleep (direction of association not reported).	NA
Nathanson et al., 2018 [34]	Cross-sectional	USA	N = 402	Age range: 3–5 y	Preschoolers	Total daily screen time; evening screen time	Parent reported television use; mobile electronic device use	Sleep duration	More time spent using a tablet during the evening (β = 0.12; SE, 0.12), smartphone (β = 0.03; SE, 0.20), game player (β = 0.06; SE, 0.20), iPod or watching TV (β = 0.2; SE, 0.07) were associated with lower sleep duration. More time spent using a tablet (β = 0.13; SE, 0.04), iPod (β = 0.02; SE, 0.10) or watching TV (β = 0.2; SE, 0.03) was associated with lower sleep duration. More time spent using a smartphone (β = −0.1; SE, 0.07), or laptop (β = −0.01; SE, 0.06) was associated with longer sleep duration.	Mothers education, mothers employment, household income, child's age, childcare attendance, TV viewing,
Nathanson et al., 2014 [35]	Cross-sectional	USA	N = 107	Mean age (SD): 53.4 mo (0.87)	Preschoolers	Total daily screen time; evening screen time	Parent reported television time	Sleep duration	More time spent watching TV during the evening was associated with shorter sleep duration (r = −0.3). Background TV time all day, background TV time in the daytime and background TV time in the night time was correlated with shorter sleep duration (r = −0.3, r = −0.3; r = −0.2, respectively). Total time spent watching TV or time spent watching TV during the day were correlated with sleep duration (r = −0.2; r = −0.1, respectively).	Income education and age
Nevarez et al., 2010 [18]	Cross-sectional	USA	N = 1676 (time 1); 1228 (time 2); 1365 (time 3)	Approximate age: 6 mo (time 1); 12 mo (time 2); 24 mo (time 3)	Infants, Toddlers	Total daily screen time	Parent reported television time	Sleep duration	At age 6 mo' more time spent watching TV was associated with shorter sleep duration (β = −0.1; 95%CI, −0.16 to 0.02). At age 12 mo more time spent watching TV was associated with shorter sleep duration (β = −0.1; 95%CI, −0.18 to −0.04). At age 24 mo more time spent watching TV was associated with shorter sleep duration (β = −0.1, 95%CI: −0.15 to −0.02).	Maternal age, parents education, household income, sex, race/ethnicity
Ota et al., 2007 [48]	Cross-sectional	Not reported	N = 330	Mean age (SD): 4.2 y	Preschoolers	Total daily screen time	Parent reported television time	Sleeping habits	Those in the regular sleeping habits group watched significantly less TV than those in the irregular group (1.7hr/d ± 1.1 compared to 2.0 h/d ± 1.2).	NA
Plancoulaine et al., 2015 [40]	Cross-sectional	France	N = 1028	Approximate age: 3 y	Preschoolers	Outside PA; Total daily screen time	Parent reported television time and other screens	Sleep duration	More time spent watching TV was associated with shorter sleep duration (<12hr/d) in boys (OR = 1.65; 95% CI, 1.23 to 2.21) but not girls (OR = 1.06; 0.76 to 1.47). Outside physical activity was not associated with sleep duration.	Socio-economic factors; Family income; Educational level; Childcare arrangements; Maternal isolation/depression (for girls); Maternal BMI; night waking (for girls); parent present at falling asleep; watching TV; Food score; BMI z-score
Reynaud et al., 2016 [41]	Longitudinal	France	N = 1346	Approximate age: 2 y (time 1); 3 y (time 2); 5–6 y (time 3)	Toddlers and Preschoolers	Total daily screen time	Parent reported television time	Night awakenings	Those spending more time watching TV at age 3 y were more likely to belong to the 2–5 common night awakenings trajectory at age 5–6 y (OR = 1.3; 95% CI, 1.13 to 1.58).	Childcare center, Parental education status, Household income, Maternal depression, Child gender, Child ponderal index, First child, Passive smoking at home, Collective care arrangement, Atopic profile, Ear nose throat infection, Falling asleep with parental presence, Bottle feeding at night, Activity, Shyness, Emotionality.
Séguin et al., 2016 [28]	Cross-sectional	Canada	N = 52	Approximate age = 45 mo	Preschoolers	Total daily screen time	Parent reported television time, computer, game console or other electronics use	Sleep patterns	More time using the computer (r = −0.38), video game console use (r = −0.32) and other electronic media use (r = −0.33) was associated with shorter sleep duration.	NA
Sijtsma et al., 2015 [42]	Cross-sectional	The Netherlands	N = 759	Age range: 3–4 ys	Preschoolers	Total daily screen time	Parent reported television time	Sleep duration	Higher amounts of screen time were associated with shorter sleep duration (r = −0.16).	NA
Taylor et al., 2015 [19]	Longitudinal	New Zealand	N = 143	Mean age (SD): 3.0 y (0.0)	Preschoolers	MVPA; total PA	Accelerometry	Sleep stability	Children displaying a more stable sleep pattern had higher levels of day-time physical activity than all other groups (Mean (SD) MVPA: low average 97 (47) minutes; variable medium sleep 91 (39) minutes; high average sleep 79 (35) minutes; consistent medium sleep 111 (49) minutes). Children displaying a more stable sleep pattern had higher counts per minutes than all other groups (Mean (SD) CPM: low average 791 (266) minutes; variable medium sleep 790 (234) minutes; high average sleep 725 (208) minutes; consistent medium sleep 913 (332) minutes).	NA
Vijakkhana et al., 2015 [47]	Longitudinal and cross-sectional	Thailand	N = 208	Approximate age: 6 mo (time 1); 12 mo (time 2)	Infants and Toddlers	Total daily screen time; evening screen time	Parent reported screen media use	Sleep duration	Evening media exposure at 6mo was associated with shorter 6mo night time sleep duration (weekday r = −0.3; weekend day r = −0.2). Evening media exposure at 12mo was associated with shorter 12mo night time sleep duration (r = −0.2 for both weekday and weekend day). Evening media exposure at 6mo was associated with shorter 12mo night time sleep duration (r = −0.2 for both weekday and weekend day). Higher levels of media viewing at 6mo was associated with shorter 6mo night time sleep during weekdays (r = −0.1) but not during weekends (r = 0.0) Higher levels of media viewing at 12mo were not associated with 12mo night time sleep duration (r = 0.0 for both weekday and weekend). Higher levels of media viewing at 6mo were not associated with 12mo night time sleep duration (r = 0.0 for both weekday and weekend).	12-mo bedroom media use, chronological age at 12mo, gender, 12mo cosleeping status, maternal education in y, mothers and fathers income in Baht
Wang et al., 2019 [49]	Cross-sectional	Taiwan	N = 183	Average age (SD): 6.61 mo (0.36)	Infants	Total PA; SB; floor play	Accelerometry and parent reported floor play	Sleep duration; sleep efficiency	PA was significant associated with a lower sleep percentage (β = −0.02), and non-significantly associated with less 24-h sleep (β = −0.03) and more time napping (β = 0.03) SB was significantly associated with less total 24-h sleep (β = −5.89) and not significantly associated with higher sleep percentage (β = 0.06) and more time napping (β = 1.41). Floor play was associated with less total 24-h sleep (β = −4.18), higher sleep percentage (β = 0.14) and less time napping (β = −3.56) but none were significant.	Gender, infant BMI, breastfeeding status, maternal employment status
Williams et al., 2014 [20]	Cross-sectional	New Zealand	N = 216	Approximate age: 3 y (time 1); 5 y (time 2)	Preschoolers	Total PA, SB, LPA, MVPA	Accelerometry	Sleep duration; night awakenings	The most active children spent 0.92 h (55 min) less time asleep at night compared with the least active children at 3 years of age. More active children were also awake more at night, for 16–19 min. These children spent less time in sedentary activity (2.49 h at age 3) and more time in light (0.14 h) and MVPA (2.95 h).	Awake at night; Sedentary time; Light activity; MVPA
Xu H et al., 2016 [51]	Longitudinal and cross-sectional	Australia	N = 497 (time 1); 415 (time 2); 369 (time 3)	Approximate age: 2 y (time 1); 3.5 years (time 2); 5 y (time 3)	Toddlers and Preschoolers	Outdoor play; Total daily screen time	Parent reported electronic media use	Sleep duration; bedtime; sleep latency; night awakenings	Higher levels of screen time at age 2 were associated with shorter night time sleep (β = −0.1; 95% CI, −0.23 to −0.03) and longer sleep latency (β = −2.5; 95% CI, 0.63 to −4.35) at age 2. Those with higher levels of screen time at age 2 were less likely to be in the long sleep group (>10hr/d; OR = 0.8; 95% CI, 0.64 to 0.95) and more likely to wake up at night (OR = 1.4; 95% CI, 1.15 to 1.72) at age 2. Levels of screen time at age 3.5 y were not associated with night time sleep (β = 0.0; 95% CI, −0.09 to 0.05). Higher levels of screen time at age 3.5 y were associated with longer sleep latency at age 3.5 y (β = 0.4; 95% CI, −1.16 to 1.97). Those with higher levels of screen time at age 3.5 y were not more likely to be in the long sleeping group or wake up at night at age 3.5 y (OR = 1; 95% CI 0.86–1.18; OR = 1.0; 95% CI, 0.82 to 1.10, for sleep duration and night awakenings respectively). Higher levels of screen time at age 2 y were associated with shorter night time sleep (β = −0.1; 95% CI, −0.09 to −0.01) and longer sleep latency (β = 1.6; 95% CI, 0.53 to 2.63) at age 5 y. Those with higher levels of screen time at age 2 y were less likely to be in the long sleeping group at age 5 (OR = 0.87; 95% CI, 0.76 to 1.0) and were more likely to wake up at night at age 5 y (OR = 1.53; 95% CI, 1.10 to 2.14).	childcare attendance, annual household income, mother's country of birth, age, education level, employment and marital status at baseline.
Zhang et al., 2019 [52]	Cross-sectional	Australia	N = 173	Average age: 19.7 mo	Toddlers	Total PA; MVPA; total daily screen time	Accelerometry and parent reported screen time	Sleep duration; sleep quality; sleep variability	Those participating in <302.9 min/d TPA had increased chances of sleeping > 646.8 min/d (OR = 2.38; 95%CI, 1.27–4.45), being in the high variability (>59.2 min/d difference between days) group (OR = 1.27; 95%CI, 0.68–2.40) and sleep problems (OR = 1.33; 95%CI, 0.71–2.50) Those participating in <55.1 min/d MVPA had an increased chance of sleeping > 646.8 min/d (OR = 1.06; 95%CI, 0.85–1.95), of being in the high variability (>59.2 min/d difference between days) group (OR = 1.23; 95%CI, 0.66–2.31), and had less chance of experiencing sleep problems (OR = 0.96; 95%CI, 0.51–1.79) Those who did not meet the screen time guidelines had a lower chance of sleeping > 646.8 min/d (OR = 0.98; 95%CI, 0.38–2.51), a greater chance of being in the high variability (>59.2 min/d difference between days) group (OR = 2.13; 95%CI, 0.77–5.90) and a greater chance of experiencing sleep problems (OR = 1.41; 95%CI, 0.55–3.65)	Age, sex, socio-economic status, body mass index

PA, physical activity; SB, sedentary behavior; MVPA, moderate-to-vigorous physical activity; VPA, vigorous physical activity; OR, Odds ratio; β, adjusted beta coefficient; B, unadjusted beta coefficient; r, Pearsons correlation coefficient; SE, standard error; 95% CI, 95% confidence interval.