Our Findings

Tracking Early Development: Insights and Innovations from PediaTrac V3.0

PediaTrac is a web-based measure designed to engage families in the continuous gathering of multidomain data on infant/toddler development from birth to 18 months of age at several time periods. PediaTrac V.3.0 utilized 571 caregiver/infant dyads (331 full-term and 240 preterm infants) across 3 study sites for analysis. There were several overarching goals that this version of PediaTrac aimed to study:

Aim 2 of PediaTrac V3.0 was to:

• characterize typical/atypical developmental trajectories using PediaTrac data from term infants and preterm infants (gestational age <37 weeks)

• examine term and preterm differences at each developmental time point

Results of Aim #1

Item and Test Information (i.e., “Reliability”) – Do the items and domains of PediaTrac produce precise and consistent results across different time periods and among infants with a range of abilities?

• The items and domains of PediaTrac were analyzed using Item-Response Theory, an advanced statistical method used to evaluate the consistency and precision of responses.

• The Motor (MOT) domain of PediaTrac measures an infant’s head, trunk, limb movement, and functional sensorimotor skills across development. Our results demonstrate that the MOT domain provides significant information (estimates of 0.93-0.94) from newborn to 18-month time periods. This indicates that the MOT domain has a very high ability to consistently and accurately assess motor development in infants across this age range, providing valuable insights into their motor skills and potential developmental delays.

• The Social/Communication/Cognition (SCG) domain of PediaTrac measures an infant’s response to stimulation, expressing and responding to emotions, communication, and acquiring knowledge. Our results demonstrate that the SCG provides significant information at both the item and domain level, (estimates of 0.97-0.98) from newborn to 18-month time periods. This indicates that the SCG domain has a very high ability to consistently and accurately assess the development of cognition, communication, and social expression in infants across this age range.

• The Early Relational Health (ERH) domain of PediaTrac (previously referred to as Attachment) measures the caregiver’s perceptions of parenting, the infant/toddler’s social-emotional status, and attachment behaviors. Our results demonstrate that the ERH domain provides significant information (estimates of 0.96-0.97) from newborn to 18-month time periods. This indicates that the ERH domain has a very high ability to consistently and accurately assess the development of attachment behaviors and emotional status of infants across this age range, as well as the caregiver’s parenting perceptions.

• The high degree of Item and Test Information may also suggest that there is redundancy in item content for each domain, and further studies will aim to modify the PediaTrac domains to improve efficiency.

Construct Validity – Do the domains of PediaTrac measure what they were intended to measure?

• The construct validity of the different domains (i.e., latent traits) of PediaTrac were modeled using Item Response Theory. Additionally, the multidimensionality of the domains was further characterized using exploratory factor analysis to identify if there were distinct subgroups of items (i.e., factors) in each domain.

• The Motor (MOT) domain of PediaTrac contained 4 to 6 separate factors after exploratory factor analysis. This indicates that within this domain, there is the potential for 4 to 6 distinct aspects of motor ability being assessed (i.e., trunk control, bilateral motor function, manual motor movement. This multidimensionality of the motor domain allows for more precise tracking of motor development in infants across time using PediaTrac.

• The Social/Communication/Cognition (SCG) domain of PediaTrac contained 2 to 6 separate factors after exploratory factor analysis. This indicates that within this domain, there is the potential for 2 to 6 distinct aspects of cognition, communication, and social behavior being assessed (i.e., affect/emotional expression, social responsiveness). This multidimensionality of the SCG domain allows for precise tracking of social-emotional, communication, and cognitive behaviors in infants across time using PediaTrac.

• The Early Relational Health (ERH) domain of PediaTrac contained 4 to 5 separate factors after exploratory factor analysis. This indicates that within this domain, there is the potential for 4 to 5 distinct aspects of infant attachment and emotional status being assessed (i.e., Parent-Child relationship, Parenting Stress, Parenting Distress, Parenting Efficacy, Perceptions of Child, & Sensitivity).

• The multidimensionality of PediaTrac domains allows for a more robust picture of an infant’s growth and progress across time. The various aspects in each dimension will allow providers and caregivers more information to detect, address, and intervene in areas of developmental concern.

Convergent & Discriminant Validity – Do the domains of PediaTrac measure what they are supposed to measure by comparison with other established measures of development?

• In order to demonstrate that the domains of PediaTrac are accurately measuring the developmental areas we aim to assess, our research team compared them to existing assessments (i.e., legacy measures) through correlational analyses.  

• Development scores on the MOT domain of PediaTrac were significantly and positively correlated across the 2-, 4-, 6-, and 9-month time periods with the caregiver-reported Ages and Stages Questionnaires-3 scales of Fine Motor and Gross Motor.

• Development scores on the SCG domain of PediaTrac were significantly and positively correlated across the 2-, 4-, 6-, and 9-month time periods with caregiver-reported Ages and Stages Questionnaires-3 scales of Communication, Personal-Social, and Problem-Solving.

• Development scores on the ERH domain of PediaTrac were found to be significantly correlated with the Parent Bonding Questionnaire, Maternal Self Esteem Inventory, and the Edinburgh Postnatal Depression Scale at the newborn, 2-, and 4-month time periods. Significant correlations were also found between caregiver responses on the ERH domain of PediaTrac and the Parenting Stress Index as well as the Brief Infant Toddler Social Emotional Assessment Competence at the 6-, 12-, and 15-month time periods.

• These results provide strong evidence that the domains of PediaTrac accurately measure the targeted developmental areas, validating its utility as a reliable assessment tool.

Results of Aim #2

Cross-Sectional Analysis of Subgroups – comparing subgroups for each time point

• In order to examine developmental differences between term and preterm infants,  cross-sectional analyses between term and preterm (gestational age <37 weeks) infants was conducted for the MOT, SCG, and ERH domains of PediaTrac.

• In the MOT domain, infants born preterm had significantly higher motor development scores than infants born at term at the newborn time point. At the 2- and 4-month time points, group differences were not significant. At the 6-, 9-, and 12-month time points, infants born at term had higher motor development scores than infants born preterm. While preterm infants may have more advanced motor skills in the first month of life, these results reflect a failure to maintain pace in motor development compared to infants born full term.

• In the SCG domain, there were no statistically significant differences between development scores of full-term and preterm infants at the newborn, 2-, 4-, or 6-month periods. However, at the 9-month time period, term infants demonstrated significantly better parent-reported social/communication/cognition development compared to preterm infants. Interestingly, female infants were found to demonstrate higher social/communication/cognition skills across groups at the newborn time period.

• The ERH domain did not show statistically significant differences between term and preterm infants at any time period.

• These results highlight PediaTrac’s ability to detect differences in development during infancy. This information can inform care providers of developmental risk and more immediate referral.

Trajectories of Developmental Subgroups – comparing the rate of growth for subgroups

• To identify trajectories of development between term and preterm infants, IRT modeling was conducted to examine the rate of growth.

• In the MOT domain, the rate or trajectory of motor skills development was steeper in term infants when compared to preterm infants in the newborn through 12-month period. Interestingly, we also found that female full-term infants had a greater rate of motor development when compared to males.

• In the SCG domain, once again, of the rate or trajectory of social, communication, and cognition development was steeper in term infants compared to preterm infants in the newborn through 12-month period. Additionally, female infants demonstrated a greater rate of development of social, communication, and cognitive skills when compared to males.

• Our results show that developmental trajectories for the SCG and MOT domains are able to distinguish term and preterm infants over time. These results also reflect PediaTrac’s utility for early detection of developmental risk or delay in infancy.

Results of Aim #3

Comparison of PediaTrac Trajectories (newborn-12mo) and Bayley Scales of Infant and Toddler Development-4 (BSID-4) at 24 months – Can PediaTrac developmental trajectories predict future development of infants/toddlers?

• Another goal of PediaTrac v3.0 was to examine the ability of PediaTrac developmental trajectories to individually and cumulatively predict overall development at 24 months. This was done by using a sample of 103 participants of the original PediaTrac study who completed Bayler’s Scales of Infant and Toddler Development (BSID-4) at 24 months of age. Preliminary correlational analyses were conducted to determine if there was a relationship between PediaTrac developmental trajectory estimates compared to an existing developmental assessment measure.

• Results demonstrated significant positive relationships between the MOT domain developmental trajectory estimate (from newborn to 12mo) and the BSID-4 Motor, Language, Cognitive, and Social-Emotional Standard Scores at 24 months.

• The SCG domain developmental trajectory estimate (from newborn to 12-month) was significantly positively related to the BSID-4 Motor, Cognitive, and Social-Emotional Standard Scores at 24 months.

• These results demonstrate that PediaTrac developmental trajectories are related to later infant development, demonstrating their potential as reliable indicators of risk when compared to traditional clinician-administered comprehensive developmental assessments.

Other Findings

Predicting Autism Risk

• Previous research has found that early motor delays may impact language and social development, which are associated with increased risk of autism spectrum disorder (ASD). One alternative goal of PediaTrac was to identify early markers of autism risk using developmental scores in motor and social/communication domains.

• PediaTrac development scores in the MOT and SCG domains at the 9-month time period were able to predict ASD risk on the Modified Checklist for Autism in Toddlers, Revised (M-CHAT-R) at the 18-month time period.

• Of the factors in the MOT domain, lower scores in “trunk stability/manual motor skills” and “crawl/cruise” predicted significantly higher M-CHAT-R risk scores at 18 months. In the SCG domain, only the “imitation/emerging verbal communication” factor predicted significantly higher M-CHAT-R risk scores at 18 months.

• These results show the need for PediaTrac to modify and refine the measurement of the constructs within the domains, as only some of the MOT and SCG factors were significant predictors of ASD risk.

Validity Scale in PediaTrac Detects Variation in Response Styles of Caregivers

• PediaTrac contains an embedded validity scale to detect atypical, negative, and positive response styles of caregivers. This validity scale is the first of its kind to be implemented within a caregiver-reported instrument of infant development. Analyses were conducted to determine if sociodemographic factors and response styles were related to reported developmental status of infants/toddlers.

• When response styles of caregivers were compared with sociodemographic variables, results showed that atypical and positive response styles were reported more often by caregivers with lower educational attainment. This indicates that these caregivers reported more advanced development in infants, as well as more variable scores. Results also showed that negative response styles were reported more often by caregivers with higher educational attainment, indicating that these caregivers reported their infant’s development as less advanced.

• Variation of scores within the SCG domain are also associated with other sociodemographic factors of caregivers. At the 2-month time period, caregiver’s age and positive response style explained a significant amount of variation in scores. At the 6-month time period, neighborhood deprivation and caregiver’s positive response style once again explained a significant amount of variation in scores.

• Interestingly, results also showed that higher neighborhood deprivation was associated with higher SCG domain development scores at the 6-month time period.

• Our results suggest that there are complex demographic and socio-cultural factors that likely impact caregiver estimates of development that must be considered when assessing infant/toddler development and developmental risk.