Latent Profile and Transition Analysis of Parasympathetic, Sympathetic, and Adrenocortical Activity within Kindergarten Children

• In Event: 3-123 - New Methodological Approaches and Applications in the Study of Children’s Multisystem Physiology

Integrative Statement

The conceptualization of stress-responsive physiological systems as operating in integrated, coordinated manner is evident in several theoretical models of cross-system functioning. The doctrine of autonomic space, for example, describes patterns of increases and decreases in responsivity across the parasympathetic (PNS) and sympathetic (SNS) branches of the autonomic nervous system (Berntson et al., 1994). Bauer, Quas, & Boyce (2002) proposed models of joint activity between the sympathetic-adrenal-medullary (SAM) and hypothalamic-pituitary-adrenal (HPA) systems. Empirically, limited research has modeled the complexity of multisystem activity, often relying on biological categories based on median splits (which may mask inter-individual differences) or interaction terms (that are generally limited to two biological systems). Moreover, few studies have explored developmentally-regulated changes in multisystem activity during early childhood when plasticity is particularly pronounced.

The current study addressed limitations in prior methodological approaches by using latent profile analysis (LPA) to evaluate multisystem activity across three biological systems (PNS, SNS, and HPA axis). In fall and spring of the kindergarten year, children (n = 338) completed a standardized protocol with four, two-minute challenges (social interview, number recall, concentrated lemon juice on the tongue, and watching an emotion-evoking video) during which physiology was collected (Bush et al., 2011). Eight indicators (baseline and reactivity measures of heart rate (HR), respiratory sinus arrhythmia (RSA), preejection period (PEP), and cortisol) were entered as continuous measures into LPA, which maximizes the ability to detect reliable, heterogeneous subgroups within a sample if they exist (Jung & Wickrama, 2008). LPA models that specified different numbers of latent profiles (ranging from two to six) were conducted. A determination of the optimal model was made on the basis of fit indices and the substantive content of the profiles.

In both fall and spring of the kindergarten year, a 3-profile solution provided the best fit to the data. Profile 1 (fall n = 42, spring n = 54) included children whose physiological response was dominated by HPA-axis reactivity, perhaps reflecting those who are particularly challenged by social stressors. Children in profile 2 (fall n = 136, spring n = 133 in the spring) exhibited the highest baseline RSA and mounted the largest PEP response, which may capture reactions to a perceived controllable challenge among children with greater regulatory capacities. Profile 3 (fall n = 159, spring n = 150) was characterized by anticipatory arousal and lower reactivity across multiple systems, possibly reflecting those children who were highly aroused prior to the protocol and reacted minimally across the tasks (see Figure 1). Latent transition analysis of fall-to-spring profile classifications indicated higher probabilities that children remained in the same profile over time compared to probabilities of profile changes (see Table 1). Associations between the profiles and children’s behavioral outcomes will also be examined.

Findings highlight the utility of a latent profile modeling approach to detect meaningful patterns of complex multisystem activity across three biological systems in children. There was consistency in the structure/patterning of multisystem activity and in children’s profile classification across the school year, suggesting stability in cross-system responsivity during early childhood.

Authors

• Danielle Roubinov, University of California, San Francisco
  Presenting Author

• Thomas Boyce, University of California, San Francisco
  Non-Presenting Author

• Matthew Lee, Rutgers, the State University of New Jersey
  Non-Presenting Author

• Nicole R. Bush, University of California, San Francisco (UCSF), Departments of Psychiatry and Pediatrics
  Non-Presenting Author