Poster #57 - A Person-oriented Examination of the Relationship Between Early Life Stress and Symptoms of Bipolar Disorder

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Developmental Psychopathology

Integrative Statement

The etiology of childhood-onset bipolar disorder is not well-researched. Early life stress has been linked to earlier onset of bipolar disorder, more mood symptoms, and lowered responsiveness to treatment (Kim et al., 2007). Although these findings indicate that early life stress plays an important role in the etiology and outcome of bipolar disorder, little is known about the biological underpinnings of this relationship. Few studies have examined cortisol responsivity (a common measure of stress responsivity) and symptoms of bipolar disorder early in life, and those that have use variable-oriented statistical methods, finding little or no significant results (Ellenbogen et al., 2006). Biological sensitivity to context theory (Boyce & Ellis, 2005) would suggest there may be distinct subgroups of individuals for which patterns of cortisol responsivity can be risk-augmenting or risk-protective depending on their developmental environment. These subgroups would not be detected by variable-oriented statistical methods. This research uses biological sensitivity to context theory and a person-oriented statistical approach to examine the relationship between early life stress, cortisol responsivity, and symptoms of bipolar disorder early in life. A variety of behavioral and emotional measures were completed by 10-year-old children and their mothers (n=119, 42.7% white, 22.8% African American, 21.8% multiracial, 12.7% other racial backgrounds, 51% female, monthly household income M=$3,196 SD=$2,805) who have experienced various amounts of early life stress. The children also completed the Trier Social Stress Task for Children to elicit a cortisol response (Buske-Kirschbaum et al., 1997). Latent profile analysis was used to determine whether there were distinct subgroups of individuals based on two cortisol variables (AUCg--total cortisol output following challenge, and AUCi--change in cortisol over time) and two variables associated with early life stress (general life events and harsh parenting experienced in the last year; Repetti et al., 2002). Preliminary analyses show evidence of four unique profiles with varying levels of life stress and cortisol responsivity. Individuals in LP1 were characterized by low levels of both cortisol responsivity and life stress (n=49). Individuals in LP2 were characterized by high cortisol responsivity and low life stress (n=31). Individuals in LP3 were characterized by low cortisol responsivity and high levels of life stress (n=28). Individuals in LP4 were characterized by high levels of both cortisol responsivity and life stress (n=11). These profiles will be used to predict symptoms of bipolar disorder, which were assessed using mania and depression subscales derived from the Child Behavior Checklist (Achenbach, 1991). Profiles will be validated. Based on biological sensitivity to context theory, we expect to find that LP4 is associated with more symptoms of bipolar disorder relative to the other profiles as their high cortisol responsivity would be risk-augmenting in a high stress environment. Additionally, we expect to find lower levels of symptoms of bipolar disorder in LP2 relative to other profiles, as their cortisol responsivity would be risk-protective in a low stress environment. These analyses will help build a clearer understanding of the biological mechanisms through which early life stress increases risk for bipolar disorder.

Authors

• Carolyn Andrews, Michigan State University
  Presenting Author

• Gloria Bogat, Michigan State University
  Non-Presenting Author

• Cecilia Martinez-Torteya, Universidad de Monterrey
  Non-Presenting Author

• Alytia Levendosky, Michigan State University
  Non-Presenting Author