Poster #102 - Can Measurement of Infants’ Sleep and Motor Development be Streamlined? Yes and No.

• In Event: Poster Session 09

Abstract

Traditional methods of collecting sleep and motor development data are intrusive, costly, and effortful, but large-scale data collection is necessary to identify periods of and deviations from stability. Capturing typical night-to-night variability of infants’ sleep requires parents to keep prolonged detailed diaries or infants to wear potentially intrusive technology. Accurate depictions of infants’ experience or proficiency with a motor skill require detailed interviews with caregivers and/or labs to be outfitted with expensive motion-analysis equipment. Our aim was to determine whether tools for assessing motor development and sleep in infancy could be streamlined. We attempted to replicate robust findings showing that the onset of motor milestones, like crawling and walking, disrupts infants’ sleep (e.g., Berger & Moore, 2021; Scher & Cohen, 2015) using two data collection techniques not previously used in this context. To collect infants’ sleep metrics, we used a commercial auto-videosomnography system. To collect motor milestone data we used an existing screening instrument.
531 families with infants 4 to 17 months old (277 male, 248 female; 6 families did not report) who were users of Nanit, a home video baby monitoring system, participated. A machine learning algorithm uses Nanit video footage of infants in their cribs to calculate and report sleep characteristics including nightly wake episodes. Parents completed the Survey of Well-being of Young Children (SWYC) online. The SWYC is a 40-question screening instrument for children under 5 years old that includes cognitive, language, motor and social-emotional subscales. For the purposes of this abstract, data from the motor subscale are reported. Parents reported infants’ skill experience using a 3-point scale (not yet, somewhat, very much)
ANOVAs with age and motor experience as predictors of number of night wakings were conducted for rolling, pulling-to-stand, and walking. There was no main effect of experience for any milestone. However, night wakings significantly decreased with age for the 4- to 8-month-olds in the rolling analysis (Figure 1). A significant interaction between age and experience in the pulling-to-stand analysis revealed that the youngest infants (6-8 mos) who pulled-to-stand very much had more night wakings than older infants (9-11 & 12-14 mos) who pulled-to-stand very much (p’s = .001). There was no main effect of age for the 9- to 17-month-olds in the walking analysis. Chi-square analyses revealed consistently uneven distributions of parent reports of experience (Table 1).
These findings only partially replicated previous results on the impact of motor milestone onset on night wakings. Like Atun-Einy, et al. (2016), we found that infants who achieved pulling-to-stand earlier than average were more likely to have sleep disturbances than infants whose onset age was typical. Given the success at replicating previously demonstrated sleep trajectories in infancy using Nanit, the inability to broadly demonstrate the relation between experience and sleep may stems from parents’ reluctance to report moderate levels of motor experience. Nanit has the potential to automatize large-scale sleep measurement in infancy, but the SWYC should be limited to “pediatric surveillance” and diagnostics (Sheldrick, et al., 2013); its lack of nuance makes it inappropriate for research.

Authors

• Sarah E. Berger, City University of New York College of Staten Island

• Aaron DeMasi, The Graduate Center, The City University of New York
  Presenting Author

• Mali Allison Waugh, The Graduate Center, CUNY

• Wei Wang