Mika Sugiyama, BA1; Kenji J. Tsuchiya, MD, PhD1,2; Yusuke Okubo, MD, PhD3; et alMohammad Shafiur Rahman, PhD1,2; Satoshi Uchiyama, PhD4; Taeko Harada, PhD1,2; Toshiki Iwabuchi, PhD1,2; Akemi Okumura, PhD1,2; Chikako Nakayasu, BA2; Yuko Amma, BA2; Haruka Suzuki, BA2; Nagahide Takahashi, MD, PhD2,5; Barbara Kinsella-Kammerer, MA2,6; Yoko Nomura, MPH, PhD2,6,7; Hiroaki Itoh, MD, PhD8; Tomoko Nishimura, PhD1,2
Author Affiliations Article Information
JAMA Pediatr. Published online January 23, 2023. doi:10.1001/jamapediatrics.2022.5356
Question Is higher screen time in infancy associated with suboptimal neurodevelopment (communication, daily living skills, and socialization) at age 4 years, and are the associations mediated by frequency of outdoor play?
Findings In this cohort study, higher screen time (>1 hour a day) at age 2 years was associated with both lower communication and daily living skills at age 4 years. For daily living skills, 18% of the association was mediated and alleviated by the frequency of outdoor play at age 2 years 8 months.
Meaning Frequent outdoor play may mitigate the connection between higher screen time and later suboptimal neurodevelopment, implying potential for intervention.
Importance Whether the association between higher screen time in infancy and later suboptimal neurodevelopment can be mitigated by frequency of outdoor play is unknown.
Objective To investigate whether higher screen time at age 2 years is associated with neurodevelopmental outcomes at age 4 years and whether this association is mediated by frequency of outdoor play at age 2 years 8 months.
Design, Setting, and Participants Participants were a subsample of the Hamamatsu Birth Cohort Study for Mothers and Children (HBC Study, N = 1258). Children were born between December 2007 and March 2012 and followed up from 1 year 6 months to 4 years. The analysis was conducted from April 2021 to June 2022.
Exposures Screen time longer than 1 hour a day at age 2 years was coded as higher screen time.
Main Outcomes and Measures Standardized scores for communication, daily living skills, and socialization domains of the Vineland Adaptive Behavior Scale, second edition, at age 4 years were used (mean [SD], 100 ). The mediating factor was frequency of outdoor play at age 2 years 8 months, with 6 or 7 days per week coded as frequent outdoor play.
Results Of 885 participants, 445 children (50%) were female; mean (SD) screen time per day was 2.6 (2.0) hours. Causal mediation analyses revealed that higher screen time at age 2 years was associated with lower scores in communication at age 4 years (nonstandardized coefficient b = −2.32; 95% CI, −4.03 to −0.60), but the association was not mediated by frequency of outdoor play. Higher screen time was also associated with lower scores in daily living skills (b = −1.76; 95% CI, −3.21 to −0.31); 18% of this association was mediated by frequency of outdoor play. Frequency of outdoor play was associated with socialization (b = 2.73; 95% CI, 1.06 to 4.39), whereas higher screen time was not (b = −1.34; 95% CI, −3.05 to 0.36).
Conclusions and Relevance Higher screen time at age 2 years was directly associated with poorer communication at age 4 years. It was also associated with daily living skills, but frequency of outdoor play at age 2 years 8 months alleviated it, suggesting outdoor play mitigated the association between higher screen time and suboptimal neurodevelopment. Future research should specify the nature of the associations and intervention measures, enabling targeted interventions that reduce the potential risk in screen time.
Screen time refers to the amount of time spent watching or using screen devices, such as televisions, video game systems, tablets, and smartphones. The inverse association of screen time with children’s neurodevelopmental well-being has been investigated, with a recent meta-analysis revealing that 75% of children younger than 2 years use them.1 These young children are particularly at increased risk of delayed language development,2,3 inattention problems,4 emotional problems, defiant behaviors,5 and poorer reading and academic performance through preschool ages and childhood.6,7 This is alarming because the age of first use is becoming lower.8,9 Guidelines have been issued by health bodies for caregivers on how to avoid the use of screen devices for children younger than 18 months or to limit screen time to 1 hour daily for children younger than 2 years.10,11 However, adherence to these guidelines is low.1
Findings on the association of screen time in young children with neurodevelopmental outcomes are still inconclusive. Pagani et al7 reported that screen time at ages 6 months and 1 year was moderately associated with cognitive dysfunction in school at age 10 years. However, another study found no association between screen time at ages 6 months, 1 year, or 2 years and cognitive skills at age 3 years.12 One issue is that neurodevelopmental domains are different between studies. Some focused on developmental milestones,13 language skills,2,3,14 or socioemotional problems,5,15 while recent studies have suggested that screen time is associated with developmental disorders, such as autism spectrum disorder (ASD).16,17 Further, the directionality of the association between high screen time and suboptimal neurodevelopment has been questioned,18 although recent studies support directionality of the association.6,13,15
Further investigations on the association between screen time and neurodevelopment have examined how it may be related to third factors, including outdoor play. A recent study showed associations of outdoor play with both screen time and preschoolers’ social behavior,19 suggesting that the suboptimal neurodevelopmental outcomes of higher screen time in younger children may be mediated by the amount of outdoor play. Outdoor activities have been inversely associated with sedentary time20 and screen time21 and associated with a child’s cognitive, social, and emotional skills later in life.22 The recent COVID-19 pandemic led to children having higher screen time, less outdoor play, and lower physical activity levels,23–25 putting them at a potentially increased risk for neurodevelopmental problems. What is concerning is that data show screen time has not decreased after seclusion measures were lifted.26 Given the data on the pandemic, its aftermath, and the existing cohort, it should be possible to consider how to minimize the effect of higher screen time in a future study.
In this article, we examined whether (1) screen time at age 2 years is associated with neurodevelopmental outcomes at age 4 years; (2) frequency of outdoor play at age 2 years 8 months is associated with neurodevelopmental outcomes at age 4 years; and (3) frequency of outdoor play at age 2 years 8 months mediates the associations between screen time at age 2 years and neurodevelopmental outcomes. Using the second edition of the Vineland Adaptive Behavior Scale (VABS-II),27 we measured neurodevelopmental outcomes that could capture communication, daily living skills, and socialization domains.
This study was conducted as a part of an ongoing birth cohort study, the Hamamatsu Birth Cohort Study for Mothers and Children (HBC Study, N = 1258). We invited all pregnant women who visited the 2 research sites and gave birth between December 2007 and March 2012 to join the study. Further details of the procedures are described elsewhere.28,29 This study was conducted under the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
We extracted data for a subsample of children participating in the study who completed observations at ages 1 year 6 months, 2 years, 2 years 8 months, and 4 years; children missing 1 or more observations were excluded. We compared the participants (n = 885) and the excluded children (n = 373) (eTable 1 in Supplement 1). This study was approved by the institutional review board of Hamamatsu University School of Medicine. Written informed consent was obtained from all caregivers for their own and their child’s participation.
Screen Time (Exposure of Interest)
We interviewed the parents of the participating children about their children’s lifestyle at age 2 years using 1 item from the ISAAC Phase 3 Environmental Questionnaire, developed for the International Study of Asthma and Allergies in Childhood (ISAAC)30 to measure screen time. Screen time was defined as mean hours of screen time per day watching or using television, DVD, video, internet and mobile phone, or video games, both actively and passively.10 ISAAC Phase 3 started in the early 2000s, and the original item asked, “During a normal week, how many hours a day does your child watch TV?” We amended this to describe further, “TV here includes DVD, video, internet, mobile phones and video games, regardless of active or passive viewing.” In our analyses, we dichotomized the variable, with 1 indicating longer than 1 hour a day (higher screen time) and 0 representing 1 hour or less. The cutoff was consistent with the published guideline by the American Academy of Pediatrics,10 which recommended 1 hour a day of noneducational viewing for children aged 2 to 5 years.
Outdoor Play (Mediator of Interest)
Parents were asked 1 question from the ISAAC questionnaire when children were aged 2 years 8 months to ascertain the frequency and duration of outdoor play. The original question, “How many days during a normal week does your child go and stay outside for 30 minutes or longer to make him/her breathe hard?” was modified to specify the frequency of physical activity when outdoors. In the analyses, we dichotomized this variable, with 1 indicating fewer than 6 days with 30 minutes or longer of any type of outdoor play a week (ie, infrequent outdoor play) and 0 for 6 or 7 days. We had to dichotomize this variable because the original value was heavily skewed, with a median of 6 days a week, and also because commands for causal mediation analyses available in the software, Stata version 17.0 (StataCorp), only allowed mediators regressed on exposure variables in linear or logistic functions.
We adopted the Japanese version31 of VABS-II27 to measure the 3 domains of children’s neurodevelopmental outcomes we analyzed: communication (receptive, expressive, and written language skills), daily living skills (skills entailing personal, domestic, and community life), and socialization (interpersonal relationships, play, and coping skills). Assessment was made through a semistructured parental interview, which allowed us to obtain age-adjusted V scores, with a mean (SD) of 100 (15).
To select appropriate covariates, we drew directed acyclic graphics using DAGitty software.32 Variables that could cause any 2 of 3 (screen time, outdoor play, or neurodevelopmental outcomes) were treated as potential confounders and entered into the analyses. We treated child’s sex, maternal and paternal education, and ASD symptoms at age 1 year 6 months as covariates in the adjusted model (Figure). Symptoms of ASD were measured using the Modified Checklist for Autism in Toddlers33 (Japanese version34) at age 1 year 6 months. The cutoff for a possible early diagnosis of ASD was set at either 3 or more points for a total score or 2 or more points for 10 critical items as previously developed in Japan.35 We assigned a value of 1 if the child scored on or above the cutoff and 0 otherwise.
First we independently tested the associations of screen time at age 2 years and outdoor play at age 2 years 8 months with communication, daily living skills, and socialization at age 4 years using linear regression in the crude and adjusted models.
In the causal mediation analyses, we investigated whether outdoor play mediated the association between screen time and neurodevelopmental outcomes (the 3 domains) using the paramed command36,37 in Stata. Causal mediation analyses, originally proposed by Robins and Greenland,38 allow statistical models to have an interaction term for an exposure and mediator and decompose total effects into natural direct effect (NDE) and natural indirect effect (NIE) under a counterfactual framework. To achieve parsimony and ease model interpretation, we excluded the interaction terms from the statistical models in the mediation analyses if the exclusion did not meaningfully change either of the effect estimates (NDE and NIE) by more than 10%. When an interaction term is removed, the 2-way decomposition of controlled direct effect and NIE originally proposed by Baron and Kenny39 is mathematically equivalent to the 2-way decomposition of NDE and NIE proposed by Robins and Greenland.38 Accordingly, NDE is the difference in counterfactual outcomes between the presence and absence of the exposure if the mediator is held at the level in the absence of exposure. NIE is the difference in counterfactual outcomes between the presence and absence of the mediator if the exposure is held present. Total effect corresponds to a sum of NDE and NIE.
We conducted sensitivity analyses to see if the finding was replicated in subsamples of children whose outdoor play was assessed during the warm months at our site (April-November, mean monthly temperature ≥10 °C) and during the cold months. If the analyzed results were inconsistent, we repeated the causal mediation analyses of the final model, adjusting for the available covariates and the season of the outdoor play measurement (warm vs cold months).
Effect estimates were reported based on both unadjusted (crude) and adjusted models. Standard errors and corresponding 95% CIs (bias-corrected) were estimated using bootstrapping procedures with 100 replications. The cutoff for 2-sided P values was set at .05.
Table 1 presents the characteristics of the 885 participants. We also compared their baseline data with those of the nonparticipating children in the original cohort (n = 373) (eTable 1 in Supplement 1). More boys (56% vs 50%) were included, and the level of maternal education (13.9 vs 13.6 years) was higher in the participants than in the nonparticipants. No other significant difference was found.
Independent Associations of Screen Time and Outdoor Play With the 3 Domains of Neurodevelopment
In the direct-only, crude models (Table 2), higher screen time (>1 hour a day) at age 2 years was significantly and inversely associated with communication (nonstandardized coefficient b = −3.05; 95% CI, −4.91 to −1.19), daily living skills (b = −2.13; 95% CI, −3.62 to −0.65), and socialization (b = −1.97; 95% CI, −3.74 to −0.21) at age 4 years. The associations remained significant after covariate adjustment except for socialization (b = −1.34; 95% CI, −3.05 to 0.36). Infrequent outdoor play (<6 days a week) at age 2 years 8 months was not associated with communication but was significantly and inversely associated with daily living skills (b = −2.16; 95% CI, −3.55 to −0.77) and socialization (b = −3.03; 95% CI, −4.71 to −1.35). These associations remained significant in the adjusted models (Table 2).
Association Between Screen Time and Outdoor Play
Higher screen time at age 2 years was significantly associated with infrequent outdoor play at age 2 years 8 months after covariate adjustment (odds ratio, 2.03; 95% CI, 1.48-2.76).
Causal Mediation Analyses
We first checked whether exposure-mediator interactions should be included by comparing the NDE and NIE estimates with and without the interaction term. Because the estimates did not show a change greater than 10%, we omitted the interaction terms and decomposed the associations.38
For the outcome communication, a large total effect (b = −2.32; 95% CI, −3.61 to −0.49) and NDE (b = −2.17; 95% CI, −3.48 to −0.23) and a nonsignificant, smaller NIE (b = −0.15; 95% CI, −0.48 to 0.16) were found (Table 3). For daily living skills, a significant total effect (b = −1.76; 95% CI, −3.57 to −0.47), a nonsignificant NDE (b = −1.44; 95% CI, −2.71 to 0.11), and a significant NIE (b = −0.32; 95% CI, −0.77 to −0.06) were found, with 18.1% mediated. For socialization, no significant total effect was found.
In the sensitivity analyses, the direction and significance of the associations were supported in children whose outdoor play was assessed during warm months but not in those with outdoor play assessed during cold months (eTable 2 in Supplement 1). To remove potential bias associated with seasonal temperature changes,40 we iterated the final analysis, adjusting for the available covariates and the season (1 = warm, 0 = cold), where the direction and significance of the associations remained the same as the results in Table 3.
This study showed that a higher screen time (>1 hour a day) at age 2 years was associated with poorer neurodevelopmental outcomes at age 4 years. Major findings were as follows: (1) Screen time at age 2 years was associated with communication and daily living skills; (2) the frequency of outdoor play did not mediate the association for communication, but it did mediate the association for daily living skills as indicated by a nonsignificant NDE but significant NIE and total effect, with the proportion of mediation 18%; (3) screen time at age 2 years was not significantly associated with socialization, but frequent outdoor play was. These associations were upheld after adjusting for potential confounders, including child’s sex, parental education, and child’s ASD symptoms at age 1 year 6 months.
Tomopoulos et al2 demonstrated that screen time in children as young as 6 months was associated with an increased risk for delay in language development. This observation is supported by recent studies.5,41,42 The direct link between early screen time and later language and communication difficulties in these studies is in line with the results of our causal mediation analysis for communication, where total effect and NDE were significant while NIE was nonsignificant, with the coefficient of NIE much smaller than that of NDE, indicating no mediation (Table 3).
Similarly, we found a significant total effect in the association between screen time and daily living skills (Table 3). The coefficient of NDE was directed to minus, as was expected from the literature, but was nonsignificant. Notably, a significant NIE suggests that the association between screen time and daily living skills is mediated at least partly by a change in the frequency of outdoor play. Specifically, a higher screen time at age 2 years decreased the score of daily living skills at age 4 years, although the decrease in the score may be reduced by 18% with a change in the frequency of outdoor play at 2 years 8 months, from less than 6 days to 6 and 7 days a week.
As expected, we found that the coefficient of total effect as a sign of association between screen time and socialization was negative, although nonsignificant. The literature offers limited support for an association between screen time and socialization: Some studies support a significant assciation19,43,44; others do not.12,42 In contrast, frequency of outdoor play is consistently associated with socialization. Similarly, consistent with the literature,7,24 higher screen time is associated with frequent outdoor play. Intriguingly, outdoor play is associated only with daily living skills and socialization and not communication (Table 2). Because an increase in outdoor play is correlated inversely with sedentary behaviors and positively with physical activity in young children,20,45 understanding the role of outdoor play in the link between early screen time and later neurodevelopment may offer leads for possible interventions.
Parental education and ASD symptoms are potential cofounders that need to be considered. Both are associated with changes in screen time18,46 and are also determinants of neurodevelopmental outcomes.47,48 We observed a reduction in both coefficients after adjustment (Table 2). As they were expected to confound the association between outdoor play and neurodevelopmental outcomes,49,50 we tested for exposure-outcome association and mediator-outcome association, both showing that confounding did not fully explain the associations.
Different mechanisms that might explain a direct association between screen time and neurodevelopment in communication have been suggested. One study suggested that an increase in audible time to television decreased the magnitude of attention to parental vocalization, leading to a decrease in child vocalization.51 Another study supported this, suggesting that background television viewing might decrease parent-child interaction.52 These associations have been observed in young children with ASD, who are prone to delays in language skill development.53 Specifically, the amount of passive screen viewing has been inversely associated with the receptive language score of children with ASD at ages 1 to 3 years,42 suggesting that high screen time in young children is associated with suboptimal neurodevelopment in communication, possibly through reduced attention to adults and compromised comprehension in conversation, and that such a explanation stands irrespective of levels of language skills or ASD diagnosis. However, understanding the association of screen time with daily living skills and socialization is complex and needs to take account of cofounders such as the cumulative risk of socioeconomic disadvantage and genetics, dating back to birth or earlier.
A prospective study has suggested that screen time, together with parenting attitude, can function as a mediator between early socioeconomic disadvantage and distal outcomes, including socioemotional suboptimality.54 It is also possible that the time spent in outdoor activities could mediate the relationship between parenting attitude and screen time.55–57 Taken together, it is imperative to investigate the role of outdoor play in optimal child neurodevelopment, not limited to potential remedial and mitigating factors. For example, recent studies have indicated the relevance of “green time”58 and sleep59 when considering the role of outdoor play in the triangulation of screen time, parenting attitude, and socioemotional neurodevelopment. All these factors need to be reevaluated in the face of natural disasters such as a pandemic.
Strengths and Limitations
This study has several advantages. First, it is a longitudinal study, with a relatively large sample size. Second, causal mediation analysis allowed us to assess the mediating effect of outdoor play in the association between screen time and neurodevelopment, suggesting a potential point for intervention. However, it also has limitations. The use of parental reports to measure screen time may have resulted in underestimation. Information about the types of screen programs watched and played is lacking; this should have been collected because the effect of high screen time differs depending on the type of program.5,60 Concerning the generalizability of the findings, children raised in the early 2010s in this study might not have been exposed to screen time as much as those raised in the 2020s. Ownership of digital devices, including mobile phones and personal computers, was stable, at around 95%, during the 2010s, and watching YouTube movies had been increasingly prevalent in Japan since the 2011 Great East Japan Earthquake.61
In this cohort study, higher screen time at age 2 years was associated with suboptimal neurodevelopment in communication at age 4 years, and this association was not confounded or mediated by factors examined in this study. Higher screen time at age 2 years was associated with suboptimal development in daily living skills at age 4 years and mediated by outdoor play at age 2 years 8 months. Future research should specify the nature of the associations and intervention measures to reduce the potential risk inherent in higher screen time and explore the mechanisms underlying the association between screen time and neurodevelopmental outcomes. Further, updating guidelines regarding media use is extremely important for parents, educators, researchers, and the children themselves.
Accepted for Publication: October 27, 2022.
Published Online: January 23, 2023. doi:10.1001/jamapediatrics.2022.5356
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Sugiyama M et al. JAMA Pediatrics.
Corresponding Author: Kenji J. Tsuchiya, MD, PhD, Hamamatsu University School of Medicine, Research Center for Child Mental Development, Handayama 1 Higashiku, Hamamatsu 431-3192, Japan (email@example.com).
Author Contributions: Drs Tsuchiya and Nishimura had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Ms Sugiyama and Dr Tsuchiya contributed equally to this study.
Concept and design: Sugiyama, Tsuchiya, Takahashi, Nishimura.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Sugiyama, Tsuchiya, Itoh.
Critical revision of the manuscript for important intellectual content: Sugiyama, Tsuchiya, Okubo, Rahman, Uchiyama, Harada, Iwabuchi, Okumura, Nakayasu, Amma, Suzuki, Takahashi, Kinsella-Kammerer, Nomura, Nishimura.
Statistical analysis: Sugiyama, Tsuchiya, Okubo, Nomura, Nishimura.
Obtained funding: Tsuchiya, Nishimura.
Administrative, technical, or material support: Uchiyama, Harada, Okumura, Nakayasu, Amma, Suzuki, Kinsella-Kammerer, Itoh, Nishimura.
Supervision: Tsuchiya, Okubo, Takahashi.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology in Japan (19H03582, 21KK0145, and 22H00492 to Dr Tsuchiya; 20K07941 to Dr Nishimura), Japan Agency for Medical Research and Development (AMED) (JP21gk0110039 to Dr Tsuchiya), and the National Institute of Mental Health (R01MH102729 to Dr Nomura). The study was conducted as part of the Collaborative Research Network for Asian Children with Developmental Disorders (CRNACDD), United Graduate School of Child Development, Osaka University, Kanazwa University, Hamamatsu University School of Medicine, Chiba University, and University of Fukui.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Data Sharing Statement: See Supplement 2.
Additional Contributions: We are grateful to all the study participants. We thank Tetsuo Kato, MD, of the Kato Maternity Clinic, and K. Sugihara, MD, PhD; M. Sugimura, MD, PhD; K. Takeuchi, MD, PhD; K. Suzuki, MD, PhD; Y. Murakami, MD, PhD; Y. Kohmura, MD, PhD; Y. Miyabe, MD, PhD; K. Hirai, MD, PhD; Y. Nakamura, MD, PhD; R. Koizumi, MD, PhD; H. Murakami, MD, PhD; Y. Kobayashi-Kohmura, MD, PhD; K. Muramatsu-Kato, MD, PhD; N. Kanayama, MD, PhD, and all attending obstetricians of Hamamatsu University School of Medicine for their full support with participant enrollment. We also thank the chief midwife, Kiyomi Hinoki, BA, and all midwives and staff at the maternity clinic of Hamamatsu University School of Medicine for their support with participant enrollment. We thank Emi Higashimoto, BA; Noriko Kodera, BA; Atsuko Nakamura, BA; Yumeno Kugizaki, BA; Yukiko Suzuki, BA; Riyo Takabayashi, BA; Maiko Honda, MA; Hiroko Muraki, MA; Makiko Narumiya, MA; Ryuji Nakahara, MA; Yuka Omori, PhD; Asako Tokizawa, MD; Damee Choi, PhD; Takanobu Horikoshi, MD, PhD; Keisuke Wakusawa, MD, PhD; Yosuke Kameno, MD, PhD; Daisuke Kurita, MD, PhD; Kiyokazu Takebayashi, MD, PhD; Masamichi Yokokura, MD, PhD; Tomoyasu Wakuda, MD, PhD; Ryosuke Asano, PhD; Thanseem Ismail, PhD; Keiko Iwata, PhD; Yasuhide Iwata, MD, PhD; Emiko Kawai, PhD; Masayoshi Kawai, MD, PhD; Yujin K. Kuroda, MD, PhD; Kaori Matsumoto, PhD; Hideo Matsuzaki, MD, PhD; Tsuruko Mori, PhD; Kyoko Nakaizumi, MD, PhD; Kazuhiko Nakamura, MD, PhD; Anitha A Pillai, PhD; Yui Seno, PhD; Chie Shimmura, PhD; Shiro Suda, MD, PhD; Genichi Sugihara, MD, PhD; Katsuaki Suzuki, MD, PhD; Kohei Yamada, PhD; Shigeyuki Yamamoto, PhD; Yujiro Yoshihara, MD, PhD; Yusaku Endoh, MD, PhD; Koichi Hirano, MD, PhD; Teruhiko Suzuki, MD, PhD; Norio Mori, MD, PhD; Nori Takei, MD, PhD; Hitoshi Kuwabara, MD, PhD; Hidenori Yamasue, MD, PhD; Masatsugu Tsujii, PhD; Atsushi Senju, PhD; and Toshirou Sugiyama, MD, PhD, for data collection and management and administrative and academic support. We are also grateful to Charles Davey, BA, for his professional editing support in English.
McArthur BA, Volkova V, Tomopoulos S, Madigan S. Global prevalence of meeting screen time guidelines among children 5 years and younger: a systematic review and meta-analysis. JAMA Pediatr. 2022;176(4):373-383. doi:10.1001/jamapediatrics.2021.6386
Tomopoulos S, Dreyer BP, Berkule S, Fierman AH, Brockmeyer C, Mendelsohn AL. Infant media exposure and toddler development. Arch Pediatr Adolesc Med. 2010;164(12):1105-1111. doi:10.1001/archpediatrics.2010.235
Madigan S, McArthur BA, Anhorn C, Eirich R, Christakis DA. Associations between screen use and child language skills: a systematic review and meta-analysis. JAMA Pediatr. 2020;174(7):665-675. doi:10.1001/jamapediatrics.2020.0327
Christakis DA, Zimmerman FJ, DiGiuseppe DL, McCarty CA. Early television exposure and subsequent attentional problems in children. Pediatrics. 2004;113(4):708-713. doi:10.1542/peds.113.4.708PubMedGoogle ScholarCrossref
Chonchaiya W, Sirachairat C, Vijakkhana N, Wilaisakditipakorn T, Pruksananonda C. Elevated background TV exposure over time increases behavioural scores of 18-month-old toddlers. Acta Paediatr. 2015;104(10):1039-1046. doi:10.1111/apa.13067PubMedGoogle ScholarCrossref
McArthur BA, Browne D, McDonald S, Tough S, Madigan S. Longitudinal associations between screen use and reading in preschool-aged children. Pediatrics. 2021;147(6):e2020011429. doi:10.1542/peds.2020-011429PubMedGoogle ScholarCrossref
Pagani LS, Fitzpatrick C, Barnett TA, Dubow E. Prospective associations between early childhood television exposure and academic, psychosocial, and physical well-being by middle childhood. Arch Pediatr Adolesc Med. 2010;164(5):425-431. doi:10.1001/archpediatrics.2010.50
51% of Japanese children get first phone at elementary school age, poll finds. The Japan Times. May 2, 2022. Accessed October 18, 2022. https://www.japantimes.co.jp/news/2022/05/02/national/japan-children-smartphones/
Kelly H. What age should you give a kid their first phone? The Washington Post. October 13, 2022, Accessed October 18, 2022. https://www.washingtonpost.com/technology/2022/10/13/what-age-kid-first-phone/
Council on Communications and Media. Media and young minds. Pediatrics. 2016;138(5):e20162591. doi:10.1542/peds.2016-2591PubMedGoogle ScholarCrossref
World Health Organization. Guidelines on physical activity, sedentary behaviour and sleep for children under 5 years of age. Accessed June 11, 2022 https://apps.who.int/iris/handle/10665/311664
Schmidt ME, Rich M, Rifas-Shiman SL, Oken E, Taveras EM. Television viewing in infancy and child cognition at 3 years of age in a US cohort. Pediatrics. 2009;123(3):e370-e375. doi:10.1542/peds.2008-3221PubMedGoogle ScholarCrossref
Madigan S, Browne D, Racine N, Mori C, Tough S. Association between screen time and children’s performance on a developmental screening test. JAMA Pediatr. 2019;173(3):244-250. doi:10.1001/jamapediatrics.2018.5056
Zimmerman FJ, Christakis DA, Meltzoff AN. Associations between media viewing and language development in children under age 2 years. J Pediatr. 2007;151(4):364-368. doi:10.1016/j.jpeds.2007.04.071PubMedGoogle ScholarCrossref
Neville RD, McArthur BA, Eirich R, Lakes KD, Madigan S. Bidirectional associations between screen time and children’s externalizing and internalizing behaviors. J Child Psychol Psychiatry. 2021;62(12):1475-1484. doi:10.1111/jcpp.13425PubMedGoogle ScholarCrossref
Dong HY, Feng JY, Wang B, Shan L, Jia FY. Screen time and autism: current situation and risk factors for screen time among pre-school children with ASD. Front Psychiatry. 2021;12:675902. doi:10.3389/fpsyt.2021.675902PubMedGoogle ScholarCrossref
Kushima M, Kojima R, Shinohara R, et al; Japan Environment and Children’s Study Group. Association between screen time exposure in children at 1 year of age and autism spectrum disorder at 3 years of age: the Japan Environment and Children’s Study. JAMA Pediatr. 2022;176(4):384-391. doi:10.1001/jamapediatrics.2021.5778
Slobodin O, Heffler KF, Davidovitch M. Screen media and autism spectrum disorder: a systematic literature review. J Dev Behav Pediatr. 2019;40(4):303-311. doi:10.1097/DBP.0000000000000654PubMedGoogle ScholarCrossref
Hinkley T, Brown H, Carson V, Teychenne M. Cross sectional associations of screen time and outdoor play with social skills in preschool children. PLoS One. 2018;13(4):e0193700. doi:10.1371/journal.pone.0193700PubMedGoogle ScholarCrossref
Gray C, Gibbons R, Larouche R, et al. What is the relationship between outdoor time and physical activity, sedentary behaviour, and physical fitness in children? a systematic review. Int J Environ Res Public Health. 2015;12(6):6455-6474. doi:10.3390/ijerph120606455PubMedGoogle ScholarCrossref
Chandra M, Jalaludin B, Woolfenden S, et al; Watch Me Grow Study Group. Screen time of infants in Sydney, Australia: a birth cohort study. BMJ Open. 2016;6(10):e012342. doi:10.1136/bmjopen-2016-012342PubMedGoogle ScholarCrossref
Burdette HL, Whitaker RC. Resurrecting free play in young children: looking beyond fitness and fatness to attention, affiliation, and affect. Arch Pediatr Adolesc Med. 2005;159(1):46-50. doi:10.1001/archpedi.159.1.46
Moore SA, Faulkner G, Rhodes RE, et al. Impact of the COVID-19 virus outbreak on movement and play behaviours of Canadian children and youth: a national survey. Int J Behav Nutr Phys Act. 2020;17(1):85. doi:10.1186/s12966-020-00987-8PubMedGoogle ScholarCrossref
Schmidt SCE, Anedda B, Burchartz A, et al. Physical activity and screen time of children and adolescents before and during the COVID-19 lockdown in Germany: a natural experiment. Sci Rep. 2020;10(1):21780. doi:10.1038/s41598-020-78438-4PubMedGoogle ScholarCrossref
Neville RD, Lakes KD, Hopkins WG, et al. Global changes in child and adolescent physical activity during the COVID-19 pandemic: a systematic review and meta-analysis. JAMA Pediatr. 2022;176(9):886-894. doi:10.1001/jamapediatrics.2022.2313
Jia P, Zhang L, Yu W, et al. Impact of COVID-19 lockdown on activity patterns and weight status among youths in China: the COVID-19 Impact on Lifestyle Change Survey (COINLICS). Int J Obes (Lond). 2021;45(3):695-699. doi:10.1038/s41366-020-00710-4PubMedGoogle ScholarCrossref
Sparrow SS, Cicchetti DV, Balla DA. Vineland Adaptive Behavior Scales. 2nd ed. Pearson; 2005.
Takagai S, Tsuchiya KJ, Itoh H, Kanayama N, Mori N, Takei N; HBC Study Team. Cohort profile: Hamamatsu Birth Cohort for Mothers and Children (HBC Study). Int J Epidemiol. 2016;45(2):333-342. doi:10.1093/ije/dyv290PubMedGoogle ScholarCrossref
Tsuchiya KJ, Matsumoto K, Suda S, et al; HBC Study Team. Searching for very early precursors of autism spectrum disorders: the Hamamatsu Birth Cohort for Mothers and Children (HBC). J Dev Orig Health Dis. 2010;1(3):158-173. doi:10.1017/S2040174410000140PubMedGoogle ScholarCrossref
Ellwood P, Asher MI, Beasley R, Clayton TO, Stewart AW; ISAAC Steering Committee. The international study of asthma and allergies in childhood (ISAAC): phase three rationale and methods. Int J Tuberc Lung Dis. 2005;9(1):10-16.PubMedGoogle Scholar
Tsujii M, Murakami T, Kuroda M, Ito H, Hagiwara T, Someki F. The Japanese Version of Vineland Adaptive Behavior Scales. 2nd ed. Nihon Bunka Kagakusha; 2014.
Textor J, van der Zander B, Gilthorpe MS, Liskiewicz M, Ellison GTH. Robust causal inference using directed acyclic graphs: the R package ‘dagitty’. Int J Epidemiol. 2016;45(6):1887-1894.PubMedGoogle Scholar
Robins DL, Fein D, Barton ML, Green JA. The Modified Checklist for Autism in Toddlers: an initial study investigating the early detection of autism and pervasive developmental disorders. J Autism Dev Disord. 2001;31(2):131-144. doi:10.1023/A:1010738829569PubMedGoogle ScholarCrossref
Inada N, Koyama T, Inokuchi E, Kuroda M, Kamio Y. Reliability and validity of the Japanese version of the Modified Checklist for Autism in Toddlers (M-CHAT). Res Autism Spectr Disord. 2011;5(1):330-336. doi:10.1016/j.rasd.2010.04.016Google ScholarCrossref
Kamio Y, Inada N, Koyama T, Inokuchi E, Tsuchiya K, Kuroda M. Effectiveness of using the Modified Checklist for Autism in Toddlers in two-stage screening of autism spectrum disorder at the 18-month health check-up in Japan. J Autism Dev Disord. 2014;44(1):194-203. doi:10.1007/s10803-013-1864-1PubMedGoogle ScholarCrossref
Valeri L, Vanderweele TJ. Mediation analysis allowing for exposure-mediator interactions and causal interpretation: theoretical assumptions and implementation with SAS and SPSS macros. Psychol Methods. 2013;18(2):137-150. doi:10.1037/a0031034PubMedGoogle ScholarCrossref
Emsley R, Liu H, Boston College Department of Economics. PARAMED: Stata module to perform causal mediation analysis using regression models. Revised April 26, 2013. https://econpapers.repec.org/software/bocbocode/s457581.htm
Robins JM, Greenland S. Identifiability and exchangeability for direct and indirect effects. Epidemiology. 1992;3(2):143-155. doi:10.1097/00001648-199203000-00013PubMedGoogle ScholarCrossref
Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173-1182. doi:10.1037/0022-3522.214.171.1243PubMedGoogle ScholarCrossref
Richiardi L, Bellocco R, Zugna D. Mediation analysis in epidemiology: methods, interpretation and bias. Int J Epidemiol. 2013;42(5):1511-1519. doi:10.1093/ije/dyt127PubMedGoogle ScholarCrossref
Duch H, Fisher EM, Ensari I, et al. Association of screen time use and language development in Hispanic toddlers: a cross-sectional and longitudinal study. Clin Pediatr (Phila). 2013;52(9):857-865. doi:10.1177/0009922813492881PubMedGoogle ScholarCrossref
Fridberg E, Khokhlovich E, Vyshedskiy A. Watching videos and television is related to a lower development of complex language comprehension in young children with autism. Healthcare (Basel). 2021;9(4):423. doi:10.3390/healthcare9040423PubMedGoogle ScholarCrossref
Rocha HAL, Correia LL, Leite AJM, et al. Screen time and early childhood development in Ceará, Brazil: a population-based study. BMC Public Health. 2021;21(1):2072. doi:10.1186/s12889-021-12136-2PubMedGoogle ScholarCrossref
Wan MW, Fitch-Bunce C, Heron K, Lester E. Infant screen media usage and social-emotional functioning. Infant Behav Dev. 2021;62:101509. doi:10.1016/j.infbeh.2020.101509PubMedGoogle ScholarCrossref
Stone MR, Faulkner GE. Outdoor play in children: associations with objectively-measured physical activity, sedentary behavior and weight status. Prev Med. 2014;65:122-127. doi:10.1016/j.ypmed.2014.05.008PubMedGoogle ScholarCrossref
Madigan S, Racine N, Tough S. Prevalence of preschoolers meeting vs exceeding screen time guidelines. JAMA Pediatr. 2020;174(1):93-95. doi:10.1001/jamapediatrics.2019.4495
Yang S, Paynter JM, Gilmore L. Vineland Adaptive Behavior Scales: II. Profile of young children with autism spectrum disorder. J Autism Dev Disord. 2016;46(1):64-73. doi:10.1007/s10803-015-2543-1PubMedGoogle ScholarCrossref
Nishimura T, Kato T, Okumura A, et al. Trajectories of adaptive behaviors during childhood in females and males in the general population. Front Psychiatry. 2022;13:817383. doi:10.3389/fpsyt.2022.817383PubMedGoogle ScholarCrossref
Lee EY, Bains A, Hunter S, et al. Systematic review of the correlates of outdoor play and time among children aged 3-12 years. Int J Behav Nutr Phys Act. 2021;18(1):41. doi:10.1186/s12966-021-01097-9PubMedGoogle ScholarCrossref
Tatsumi Y, Mohri I, Shimizu S, Tachibana M, Ohno Y, Taniike M. Daytime physical activity and sleep in pre-schoolers with developmental disorders. J Paediatr Child Health. 2015;51(4):396-402. doi:10.1111/jpc.12725PubMedGoogle ScholarCrossref
Christakis DA, Gilkerson J, Richards JA, et al. Audible television and decreased adult words, infant vocalizations, and conversational turns: a population-based study. Arch Pediatr Adolesc Med. 2009;163(6):554-558. doi:10.1001/archpediatrics.2009.61
Kirkorian HL, Pempek TA, Murphy LA, Schmidt ME, Anderson DR. The impact of background television on parent-child interaction. Child Dev. 2009;80(5):1350-1359. doi:10.1111/j.1467-8624.2009.01337.xPubMedGoogle ScholarCrossref
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. American Psychiatric Association; 2013.
McArthur BA, Browne D, Racine N, Tough S, Madigan S. Screen time as a mechanism through which cumulative risk is related to child socioemotional and developmental outcomes in early childhood. Res Child Adolesc Psychopathol. 2022;50(6):709-720. doi:10.1007/s10802-021-00895-wPubMedGoogle ScholarCrossref
Zhao J, Zhang Y, Jiang F, et al. Excessive screen time and psychosocial well-being: the mediating role of body mass index, sleep duration, and parent-child interaction. J Pediatr. 2018;202:157-162.e1. doi:10.1016/j.jpeds.2018.06.029PubMedGoogle ScholarCrossref
Xu H, Wen LM, Rissel C. Associations of parental influences with physical activity and screen time among young children: a systematic review. J Obes. 2015;2015:546925. doi:10.1155/2015/546925PubMedGoogle ScholarCrossref
Wong RS, Tung KTS, Rao N, et al. Parent technology use, parent-child interaction, child screen time, and child psychosocial problems among disadvantaged families. J Pediatr. 2020;226:258-265. doi:10.1016/j.jpeds.2020.07.006PubMedGoogle ScholarCrossref
Oswald TK, Rumbold AR, Kedzior SGE, Moore VM. Psychological impacts of “screen time” and “green time” for children and adolescents: a systematic scoping review. PLoS One. 2020;15(9):e0237725. doi:10.1371/journal.pone.0237725PubMedGoogle ScholarCrossref
Janssen X, Martin A, Hughes AR, Hill CM, Kotronoulas G, Hesketh KR. Associations of screen time, sedentary time and physical activity with sleep in under 5s: a systematic review and meta-analysis. Sleep Med Rev. 2020;49:101226. doi:10.1016/j.smrv.2019.101226PubMedGoogle ScholarCrossref
Byrne R, Terranova CO, Trost SG. Measurement of screen time among young children aged 0-6 years: a systematic review. Obes Rev. 2021;22(8):e13260. doi:10.1111/obr.13260PubMedGoogle ScholarCrossref
Ministry of Internal Affairs and Communication. Information and Communications in Japan 2020. Accessed October 7, 2022. https://www.soumu.go.jp/main_sosiki/joho_tsusin/eng/whitepaper/2020/index.html