This monitoring study was conducted in 2 districts of Shanghai, Xuhui District (urban), and Pudong New Area (suburban), from March 2023 to July 2023. A random sampling method was used to select a kindergarten from each of the 2 districts. For each kindergarten, a cluster sampling method was used to select 5 classes, in which the average number of children in each class was 30, from the junior, middle, and senior grades. A total of 535 preschool children were included in this study, and they were required to use the Huawei iPad for 1 week. During the recruitment process, we repeatedly emphasized that children should use tablets according to their daily habits at home and that parents should neither indulge in children’s use which may lead to an overestimation of screen exposure duration nor overly regulate it which may result in an underestimation of screen exposure duration. All the children were provided with intelligent monitoring technology and completed a developmental screening questionnaire. The minimum sample size was calculated to be 462 by using the following formula: , where the reported prevalence of abnormal development among nationwide Chinese preschool children was approximately 10.8% to 26.2% [24]. In this study, the prevalence (p) was 10.8%, the type I error (α) was .05, the z_1-/2 was 1.96, the precision (d) was 0.04, and the design effect (deff) was 2 [25]. The inclusion criteria for participant enrollment were as follows: (1) children aged between 2 and 6 years; (2) children whose parents or other caregivers allowed them to use screen devices, such as televisions, smartphones, iPads, and computers; and (3) children whose parents or other caregivers understood the questionnaire.

Participant characteristics are presented as frequencies with percentages and as the means with SDs. First, we used descriptive analysis to show the demographic characteristics of the participants in the whole sample with questionnaire data and in the subsample with HSVC device records. Second, a 2-tailed Student t test and ANOVA were performed to explore the distribution of the total ASQ-3 score for the whole sample and the scores on the 5 domains according to demographic characteristics, lifestyle, screen time category of the 4 media types, screen use styles, and type of screen exposure content, as shown in Tables S2 and S3 in Multimedia Appendix 1. Then, 6 multiple linear regression analyses were performed to examine the associations between the total ASQ score and the scores on the 5 domains (dependent variable), screen time category of the 4 media types, screen use styles, and type of screen exposure content (independent variables) adjusted for age, gender, parental educational attainment, and monthly income (control variable; Table S4 in Multimedia Appendix 1). Furthermore, Spearman correlation analysis was performed to examine the univariate association between the total ASQ score and the scores on the 5 domains and between screen viewing time and exposure to different types of content according to the HSVC application, as shown in Table S5 in Multimedia Appendix 1. Finally, median regressions were used to examine the nonlinear correlation trend between the median score on the ASQ and median time of exposure to content according to the HSVC sample application, as shown in Table S6 in Multimedia Appendix 1. The estimates of the independent variables for the total ASQ score and the domain scores were summarized using regression coefficients (β) and 95% CIs. Statistical analyses were performed using R software (version 4.3.2; R Foundation for Statistical Computing).

Before the participants underwent screen exposure monitoring and questionnaire surveys, we informed their parents (legal guardians) of the whole process of participation in the surveys as well as the benefits and risks involved, and the legal guardians signed the informed consent. This study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Ethics Committee for Medical Research at the School of Public Health, Fudan University (IRB#2023-11-1088). When extracting data from the Huawei Research Platform and the web-based questionnaire library for analysis, all the data were deidentified, and an anonymous study ID was used as an identifier for each participant. In this study, we provided a gift of a paintbrush set, worth approximately US $15, to each child participating in the research.

As shown in Table 1, among a total of 535 children and 365 preschool children with tablet use, 298 (55.7%) and 205 (56.2%) were boys, and more than half of the participants’ caregivers were their mothers (n=297, 55.5% and n=205, 55.6%, respectively). The majority of participants’ parents had a bachelor degree (mothers: n=290, 54.2% and n=196, 53.7%; fathers: n=286, 53.5% and n=194, 53.2%). Most of the participants’ monthly incomes were over US$ 1750 (n=309, 57.8% and n=202, 55.4%). Nearly half of the participants were only children (n=272, 50.8% and n=185, 50.7%). Only 50 (9.3%) and 33 (9.1%) of participants had insufficient sleep during the day, while 412 (77%) and 280 (76.7%) of participants had insufficient sleep during the night (less than 10 hours). Less than half of the participants reported insufficient indoor physical activity (209, 39.1% and 138, 37.8%) or insufficient outdoor physical activity (183, 34.2% and 120, 32.9%).

The daily median screen time on all media was 98.6 (60-65.5) minutes, and the screen time on tablets was 41.1 (0-107.2) minutes. The mean total score and score for each domain of the ASQ for the total sample and tablet sample were 269.3 (SD 33.33) and 269.89 (SD 30.75); 55.05 (SD 7.47) and 54.86 (SD 5.51) for communication); 51.54 (SD 10.39) and 51.61 (SD 10.12) for gross motor; 50.38 (SD 11.26) and 50.45 (SD 10.86) for fine motor; 56.61 (SD 6.92) and 56.82 (SD 6.01) for problem-solving; and 55.73 (SD 7.11) and 56.14 (SD 6.45) for personal-social, respectively. The participant characteristics and total ASQ and domain scores did not significantly differ between the 2 sample populations (P>.05).

The univariate associations between demographic characteristics and lifestyle variables and ASQ-3 scores are shown in Table S2 in Multimedia Appendix 1. Age, gender, maternal and paternal educational attainment, and monthly income were associated with the total ASQ score and domain scores (P<.05). The results of the univariate association between screen exposure types and ASQ-3 scores are shown in Table S3 in Multimedia Appendix 1. Smartphone screen time, iPad screen time, education content category, animated or realistic content, co-viewing programs, and co-playing games were associated with the total ASQ score and the domain scores (P<.05).

Furthermore, as shown in Table S4 in Multimedia Appendix 1, the results of the fully adjusted linear regression model suggested that older age, compared with an age of approximately 34 months 16 days to 50 months 30 days, was more strongly associated with the personal-social score (β=2.29; 95% CI 0.02-4.56; P=.048 and β=2.79; 95% CI 0.75-4.83; P=.008). Male gender was negatively associated with the total ASQ score (β=–8.92; 95% CI –14.41 to –3.43; P=.002), fine motor score (β=–4.16, 95% CI –6.04 to –2.28; P<.001), and personal-social score (β=–2.52; 95% CI –3.71 to –1.33; P<.001). Comparing with no exposure to smartphones, the total ASQ-3 score, gross motor scores were negatively associated with the use of smartphones for less than 30 minutes (β=–8.62; 95% CI –15.58 to –1.67; P=.01 and β=–3.16, 95% CI –5.4 to –0.92; P=.01), from 30-60 minutes (β=–13.28; 95% CI –21.98 to –4.58, P=.003 and β=–4.52; 95% CI –7.32 to –1.71; P=.002) and for more than 60 minutes (β=–12.89; 95% CI –24.75 to –1.02; P=.03 and β=–4.71; 95% CI –8.54 to –0.88; P=.02). The duration of iPad screen time exceeding 1 hour (β=2.45; 95% CI 0.39-4.51; P=.02) had a more negative association with the fine motor score than did never using iPads. Engagement in co-playing games with caregivers was positively associated with the total ASQ-3 score (β=11.64; 95% CI 2.66-20.62; P=.01), communication (β=2.19; 95% CI 0.1-4.27; P=.04), and fine motor scores (β=4.76; 95% CI 1.68-7.84; P=.003), respectively.

The Spearman correlation coefficients between the screen content and ASQ-3 scores based on the HSVC data are shown in Table S5 in Multimedia Appendix 1. The results showed that education content, pace, screen interaction, animated and realistic content, date, and screen time duration were associated with the total ASQ-3 score, fine motor score, problem-solving score, and personal-social score (P<.05).

Furthermore, the results of the median regression between screen content and the ASQ-3 score after adjusting for age, gender, monthly income, co-viewing programs, and co-playing games are shown in Table S6 in Multimedia Appendix 1. This study showed that the median total score on the ASQ-3 was negatively associated with screen time for noneducational content (β=–.055; 95% CI –0.148 to –0.006; P=.03), educational and noneducational content (β=–.042, 95% CI –0.081 to –0.007; P=.001), and fast-paced content (β=–.034; 95% CI –0.062 to –0.011; P=.049). The median gross motor score was negatively associated with screen time for PG-13 rated content (β=–.015; 95%CI –0.022 to 0.009; P=.03), educational and noneducational content (β=–.018, 95% CI –0.038 to –0.001; P=.02), static content (β=–.022, 95% CI –0.050 to 0.007; P=.02). This study also found that the median fine motor score was negatively associated with the time spent viewing G rated content (β=–.032; 95% CI –0.057 to –0.003; P=.006), PG rated content (β=–.020; 95% CI –0.036 to –0.007; P=.004), noneducational content (β=–.026; 95% CI –0.067 to –0.003; P=.01), both educational and noneducational content (β=–.020; 95% CI –0.034 to –0.0010; P<.001), fast-paced content (β=–.022; 95% CI –0.033 to –0.014; P<.001), static content (β=–.034; 95% CI –0.050 to 0.018; P<.001), animated content (β=–.038; 95% CI –0.069 to –0.001; P=.004), and screen use during the daytime (β=–.026; 95% CI –0.043 to 0.005; P=.005). This study suggested that screen time during the daytime and the time spent viewing touch-based content were negatively associated with the median problem-solving (β=–.017; 95% CI –0.050 to –0.007; P=.03) and personal-social scores (β=–.016; 95% CI –0.033 to 0.003; P=.006), respectively.

Based on 7 days of monitoring data collected by the HSVC application from 2 representative samples of kindergarten children aged 34.5 to 66 months, we analyzed the association of different screen content types and the ASQ-3 score. We found that greater exposure to noneducational programs, fast-paced content, static, animated content, and screens during the daytime was associated with a lower median gross motor score and median fine motor score after controlling for covariates; moreover, greater exposure to noneducational programs and fast speed programs was also associated with lower total ASQ-3 scores.

The results showed that the mean (SD) total screen time and tablet screen time was 111.6 (140.9) minutes and 31.1 (42.5) minutes per day for children aged 34 months 16 days to 50 months 30 days, 172.7 (286.5) minutes and 41.2 (45.5) minutes per day for children aged 51 months 0 days to 56 months 30 days, and 159.3 (191.5) minutes and 42.4 (52.3) minutes per day for children aged 57 months 0 days to 66 months 0 days; these findings are similar to previous studies [32-34]. Whether educational programs are beneficial for early development in toddlers and preschool children is still controversial. This study also did not observe a positive or negative association; however, a significant finding was revealed that exposure to noneducational content was negatively associated with early development (total score on the ASQ-3) or fine motor in preschool children included in this monitoring study. Although there is little direct evidence of a relationship between the ASQ score and its domain scores, several longitudinal studies have shown that exposure to noneducational content before 3 years of age is negatively related to executive functioning and cognitive development at older ages, which are fundamental to the development of fine motor [35-37]. Another longitudinal study confirmed that screen use time was negatively correlated with fine motor skills 1 year after birth among children aged 3-6 years by cross-lagged path analysis [38]. The earlier cross-lagged conclusion may be explained by the results of this study, in which screen time in relation to categories containing both educational and noneducational content was still negatively associated with fine motor skills because it is inevitable to avoid noneducational content for preschoolers.

This study also revealed that exposure to fast-paced content (more than 6 times per minute) was negatively associated with the total ASQ score and fine motor score, while these relationships were not found for low-paced (less than 2 times per minute) or medium-paced (2 to 6 times per minute) content. Furthermore, the daily tablet screen time via tablet for fast-paced content (76.4 minutes) was far greater than that for medium- (8 minutes) and slow-paced content (9.1 minutes); namely, the predominant pace of the screen content for children aged 2-6 years was fast, which is consistent with the findings of previous studies showing that preschool children often view fast-paced content, with rapid screenshots and angle changes [18]. Several controlled experimental studies have shown that watching fast-paced television significantly impairs executive function immediately in children aged 4-6 years [17,18], and these test programs could appropriately reflect fine motor, problem-solving, communication, and personal-social skills.

Touch-enabled media technological devices (tablets and smartphones) play an important role not only because of their growing popularity among children but also because of their influence on early development; the use of tablets and smartphones was associated with decreased scores on the ASQ and its domains in this study, while other media did not show this relationship. Although exposure to touch-enabled content was not associated with the ASQ score, exposure to static content was negatively associated with gross motor and fine motor scores. There is some experimental evidence that toddlers can more readily learn from touchscreen devices than from televisions [39], possibly because the use of tablets is beneficial for motor development in preschoolers [40].

Animated content is widespread in children’s media, and even realistic programs incorporate unrealistic elements. This study revealed that exposure to fantastic and both fantastic and realistic content was negatively associated with the fine motor score; these results are consistent with previous experimental studies [17]. In the working memory system, the content of screen stimuli is encoded, processed, and stored, while animated content is difficult to process because there is no stored schematic to use for interpreting animated content, especially for children’s immature brains [17,41]. Engagement in such activities could lead to a decrease in local neurotransmitters in the lateral prefrontal cortex, subsequently resulting in a reduced availability of these neurotransmitters [42].

This study also revealed that general audience and PG content were associated with lower fine motor scores, while PG-13 content was associated with lower gross motor scores. A plausible explanation regarding this result may be that regardless of what the suggested audience of the screen content is, screen use can replace the time spent on physical activity [12,43]. A previous study revealed that the viewing of PG-13 and restricted content was associated with higher hyperactivity and aggression scores and a lower social skills rating. However, this association was not found in early childhood [44].

To our knowledge, this study was the first to examine preschool children’s tablet screen time for different content types via 7-day intelligent monitoring records. By analyzing a sample of representative kindergarteners, we analyzed the associations between screen time and exposure to content types and early development. Several limitations of this study, however, should be mentioned.

First, because of the 7-day monitoring design, we could not calculate the causal relationship between screen exposure and early development. Longitudinal prospective studies and randomized clinical trials are needed to determine causal and cause-and-effect relationships regarding genetic and environmental factors. Second, this monitoring study could accurately measure tablet screen exposure, while other media, such as televisions and smartphones, were measured via questionnaires on the frequency of different types of screen use; thus, the HSVC application should be appropriately used in other devices, such as smartphones or game machines. Third, although the participants were from developed provinces in China, we recruited kindergarteners from urban and suburban areas, which might influence the generalizability of the results to lower-income districts because all areas in Shanghai have relatively high incomes compared with the national average income level. Fourth, in the screen time assessment collected by questionnaires, a limitation was that no distinction was made between weekends and weekdays in the screen time question asked to families and the real-time screen time calculation, which may result in the existence of deviations in the estimation of the screen time. Fifth, although we informed parents to let children use tablets according to their usual habits at home, this process might still increase or change children’s screen exposure behaviors to some extent. In future studies, we will explore ways to make parents accept the installation of intelligent monitoring technology on their own tablets and smartphones so as to restore children’s screen exposure behaviors to the greatest extent possible. Finally, although this survey investigated the age intervals of children according to the age versions of ASQ-3, this study failed to investigate the specific dates of birth of children. Therefore, it may affect the rigor of statistical analysis to some extent.

This monitoring study showed that screen time associated with exposure to different content types was significantly associated with early development among preschool children, especially for noneducational, static, fast-paced, and animated content, which were associated not only with the total ASQ score but with gross motor and fine motor scores. The highest priority should be placed on limiting the aggregated screen time of noneducational, static, fast-paced, and animated content programs or games, which provide insight into defining high-quality screen content in guidelines.