Early Screen Exposure Predicts Autism Symptoms at 2 Years: Data from The National Children’s Study

Document Type

Poster Presentation

Publication Date



Background. Fifty percent of autism spectrum disorder (ASD) risk is non-genetically determined (Hallmayer et al., 2011; Sandin et al., 2014), yet experiential factors related to ASD risk are largely unexamined (Heffler & Oestreicher, 2016). Greater screen viewing and an earlier onset of viewing have been retrospectively reported among toddlers with ASD (Chonchaiya, Nuntnarumit, & Pruksananonda, 2011), but the effects of screen exposure as early as 12-months on future ASD symptoms have not been examined. We hypothesized that screen exposure and indices reflecting low social interaction at 12 months would predict greater autism symptoms at age 2 years.

Methods. Participants were 2181 children (51% male) enrolled at birth in The National Children’s Study. Parents reported children’s screen exposure at 12- and 18-months; their reading and play frequency with their child at 12-months; and autism symptoms on the Modified Checklist for Autism in Toddlers (M-CHAT) at 26.7 months (+2.2 months). Screen exposure was assessed at 12-months from the item: “Does your child watch TV/DVDs?” (0=no, 1=yes) and at 18-months from the item: “Over the past 30 days, on average, how many hours per day did child watch TV and/or DVDs?” (0=none, 1=1 hr or less, 2=2 hrs, 3=3 hrs, 4=4 hrs, and 5=5 or more hrs). Because prior research has found over 4 hrs of early screen exposure to be retrospectively associated with ASD (Chonchaiya et al., 2011), we recoded the 18-month data such that 0=3 hrs or less exposure and 1=4 hrs or more exposure per day. Reading frequency was assessed by the item: “How often does participant read or look at books with child?” (1=once a week or less, 2=2-4 days a week, 3=5-6 days a week and 4=every day) and play frequency by the item “How often does participant play with toys with child?” (0=less than daily, 1=daily). Given that prematurity, maternal age, household income, and parental English language status (as a proxy for parental immigration) have each been associated with increased ASD risk, each was included as a covariate in a simultaneous regression model predicting ASD symptoms.

Results. Greater 12-month (p=.023) and 18-month (p=.042) screen exposure predicted greater ASD symptoms at 2-years (see Table 1). Less frequent parental reading to their child (p=0.008) and playing with their child (p=.023) also predicted greater ASD symptoms. In addition, pre-term birth (p= .004), low household income (p
Conclusion. This is the first study to find early screen exposure to prospectively predict autism symptoms. In addition, social interaction indices of parental time spent reading to and playing with one’s child also predicted autism symptoms. Future research is needed to determine whether early screen exposure and a lack of social interaction cause increased risk for autism symptoms as opposed to being consequences or concurrent covariates of such increased risk. If found to be causal, educating parents to limit early screen exposure and to increase social interaction with their child could present opportunities to decrease autism risk, especially among high-risk populations.

Publication Title

SRCD 2019 Biennial Meeting

Open Access