Early childhood (2–5 years old) is an essential period during physical activity (PA) behavior development [1
], and PA is an important correlate of health in the early years [2
]. Research shows that parents play an important role in early childhood PA development and the years beyond [1
], with a growing number of studies on child and parent PA confirming positive associations between child-parent habitual daily PA volumes throughout childhood and well into adulthood [4
]. While a large body of evidence appears to support an interdependent child-parent PA relationship, research also shows that these dyadic PA interdependencies may not be ubiquitous [6
]. As such, further research is needed to better understand the interdependent child-parent PA relationship and contributing factors.
Studies of dyadic PA in young children and their parents have reported differential associations between child and parent PA after stratifying analyses by child and parent sex [6
]. However, little is known about what contributes to these differences in the dyadic PA relationship between boys, girls, and their parents. A recent review of methods for measuring child and parent co-participation in PA has suggested the use of accelerometry in tandem with an objective measure of child-parent proximity to provide more robust descriptions of child-parent PA [9
]. Such a multi-sensor approach captures dyadic PA in terms of both dyadic activity intensities and spatial dynamics [9
], which in their combination may help to reveal interpersonal factors associated with the child-parent PA relationship at a key developmental stage.
Few studies, however, have measured child and parent PA and dyadic proximity in early childhood using wearable technology [7
]. In a sample of 1–5-year-olds and their mothers, no differences in joint child-parent PA volumes were found between mothers and their children with respect to child sex [7
]. By contrast, a report of objectively measured child-parent PA and proximity in older children (8–14-year-olds) showed that child sex was associated with differences in joint child-parent PA volumes [10
]. Given the potential influences of interpersonal proximity, child sex, and parent sex on the child-parent PA relationship, the need for further research on dyadic PA and proximity during early childhood is evident. This need is further underscored by the overall paucity of objective dyadic PA-proximity reports that are currently available to inform the developing definition of child and parent co-participation in PA [9
Therefore, the aims of this descriptive study were to measure dyadic physical activity and interpersonal spatial proximity in 2 year-old boys and girls and their parents using wearable technology in order to characterize: (1) habitual daily child-parent PA interdependence, with and without a measure of dyadic proximity, (2) hour-to-hour interactive child-parent PA interdependence over a 3 day period, and (3) joint physical activity behaviors in child-parent dyads.
This study aimed to characterize the dyadic activity–proximity relationship in a sample of 2-year-old boys and girls and their parents. Using a dyadic analysis statistical methodology, results showed that child and parent mean daily SED and LPA, but not MVPA behaviors, were interdependent. These results, however, were conditional on child sex and child-parent proximity. Hour-to-hour dyadic PA behaviors were also interdependent after controlling for dyadic proximity, which was inversely associated with maternal hourly PA. Results also showed that children who engaged in higher daily MVPA volumes participated in greater amounts of joint (i.e., mutual and proximal) TPA time with their mothers than children with lower daily MVPA volumes. Moreover, with respect to joint TPA time, engaging in higher volumes of daily MVPA for children was associated with participating in dyadic PA across greater distances when compared to less active children. However, joint PA results were also conditional on child sex. Overall, the findings from our study suggest that interpersonal spatial dynamics between dyadic counterparts may help to explain some of the variability in early childhood PA, specifically with respect to boys’ and girls’ daily activity volumes and intensities.
At the dyadic level, child-mother daily PA was inversely associated with the proportion of time that counterparts were proximal, and child sex was found to moderate the inverse relationship between dyadic PA and proximity. However, further analyses of hour-to-hour activity–proximity data revealed that the inverse relationship between dyadic proximity and PA was only significantly associated with maternal hourly PA, and that it had no association with children’s hour-to-hour PA behaviors. It is important to note, however, that our model made no constraints on mutuality with respect to child and mother activity intensities at any given hour. As such, our data suggest that mothers were less active as they and their children spent more time in proximity, irrespective of their counterpart’s level of activity, which included proximal sedentary time. In another study, of objectively measured activity and proximity in young children and their mothers, joint TPA time was inversely associated with mothers’ PA at times when they were not proximal to their children [7
]. Though direct comparisons between studies may be limited due to differences in proximity data processing approaches [7
], both point toward inverse associations between dyadic proximity and maternal PA behaviors. Thus, further research is needed to better explain influential factors in the observed inverse relationship among mothers, as well as to confirm its presence across diverse dyadic cohorts. Additionally, child and parent hour-to-hour PA variances in our study remained correlated between counterparts after adjusting for within-subject hourly PA tracking, dyadic cross-partner interactions, and dyadic proximity. The remaining shared dyadic variance may point toward genetic, cultural, and environmental factors that were unexplained by the model [1
]. As such, more research is needed to respectively characterize the relative contributions of learned, inherited, and environmental factors on the development of PA behavior in early childhood.
Targeted evaluation of the time periods during which children and mothers were engaged in joint PA revealed that toddlers’ daily MVPA volumes were associated with differences in child-mother joint PA time. Specifically, children in our study who engaged in higher daily volumes of moderate–vigorous PA (≥60 min/day) also spent an average 18% of their time (~2 h) in joint TPA with their mothers each day, as where children with lower daily MVPA volumes (<60 min/day) spent 13% of their time (~1.5 h) in joint PA with their mothers on average daily. These results are especially striking in light of the null relationship between child-mother MVPA time, and appear to suggest that children’s independent MVPA time is associated with joint child-mother total PA. Additional research is needed on familial co-participation in PA [9
], especially toward considering joint PA recommendations in the early childhood years when PA modeling is considered an important feature of PA development [1
]. Finally, given the average 30 min difference in joint PA between groups, these results invite research on the benefits of ~30 min joint PA activity interventions in young children and parents, particularly as they grow older and PA recommendations widely suggest participation in ≥60 min of daily MVPA [21
Furthermore, our novel analysis of relative distance scores between dyadic counterparts while engaged in joint PA showed that child-mother spatial dynamics were associated with differences in toddlers’ daily MVPA volumes. Specifically, girls and boys who engaged in ≥60 min MVPA/day also participated in joint PA at greater distances from their mothers; this particular activity–proximity dynamic was especially important for boys. Among girls with lower daily MVPA volumes, however, there were no differences in the relative distances at which they participated joint TPA when compared to any other group. Notably, girls in our study were also found to participate in more joint PA with their mothers than boys. A prior report of dyadic PA and proximity in 8–14-year-olds and their parents also showed that girls spent more time in joint MVPA with their parents than boys [10
], which similarly points toward sex-dependent differences in the dyadic activity relationship [6
]. Despite differences between boys and girls with respect to child-mother joint PA time, no differences were found between the overall time that boys and girls spent in PA. While our report of no difference in overall PA volumes between girls and boys is congruent with other studies of PA in 2-year-olds [27
], these results remain striking because studies also report differences in PA between girls and boys by 3 years of age [28
]. Taken together, these findings encourage especial attention toward the influence of sociocultural and interpersonal factors on the differential trajectories of PA behavior in young girls and boys. To better understand the contextual qualifiers that may explain these apparent sex-based differences, further research on PA behaviors in young boys and girls is needed with a particular focus on the concurrent influences of dyadic interpersonal spatial dynamics, parental support for PA, and parenting style [29
Regarding limitations of the study, the adult sample in our study was predominantly comprised of mothers and all the urban-dwelling families in our study reported an annual income below the federal poverty level, thus our results may be more indicative of activity–proximity relationships in mother-toddler dyads. Also, the use of dyadic analysis limits the direct comparison of these findings to other studies that have used a dyadic analysis approach to study child-parent PA behaviors [23
]. Findings from our study may, therefore, be generalizable to other under-resourced toddler–parent dyads living in areas of pronounced need within a major urban center.
To our knowledge, this study is the first to use a dyadic analysis statistical methodology to analyze dyadic physical activity [4
], as such parameter estimates were appropriately and necessarily adjusted for interdependent relationships between child and parent signals [23
]. Along with the recent recommendation to include objective measures of dyadic proximity in studies of child-parent PA [9
], we add that studies of dyadic PA and proximity should consider the use of dyadic analysis in the treatment of their dyadic data. Results from our study show its utility in analyzing correlated dyadic data, but also reveal its potential to uncover additional layers of information regarding cross-partner and within-subject PA behavior patterns. For example, an unintended, albeit notable, artifact of employing the hour-to-hour APIM was the discovery that toddlers’ hourly PA volumes were stable over time. Thus, while it is well-established that young children characteristically engage in random “short burst” PA patterns [12
], our data show that toddlers’ activity patterns also appear to exhibit a significant degree of consistency throughout the day.