Differential effects of threat vs. deprivation dimensions on the HPA axis

• In Event: 2-147 - Is there more to childhood adversity than cumulative stress? Distinguishing threat/ deprivation in large-scale studies

Integrative Statement

Recent theory suggests that effects of childhood adversity on neurobiology vary according to the magnitude of deprivation and/or threat (Sheridan & McLaughlin, 2014). Deprivation, including neglect, relatively robustly predicts attenuated HPA axis activity, but the patterns following threatening experiences are equivocal (Koss & Gunnar, 2018). Moreover, the role of these potentially altered secretory patterns for psychopathology remains little understood. Here, we drew on multi-source assessments of maltreatment and psychopathology to examine effects of deprivation and threat in a large-scale sample using hair cortisol concentrations (HCC) – an index of long-term cumulative cortisol secretion (Study 1) – and, in a subsample, using reactivity of cortisol and neuroactive steroids to social laboratory stress (Study 2). We hypothesized that deprivation would (1) give rise to attenuated HCC and (2) blunted cortisol reactivity to social stress, both of which would mediate effects of deprivation on externalizing problems. We additionally explored effects on other neuroactive steroids, including dehydroepiandrosterone (DHEA) which antagonizes effects of cortisol.
For Study 1, we recruited 465 youth (47.6 % girls) aged 9-16 years (M = 12.59, SD = 1.77), of whom 177 experienced maltreatment. Parent-interviews and child protection services records, if available, were coded using the Maltreatment Classification System. Chronicity, severity, and number of subtypes of neglect (failure to provide, lack of supervision and moral-legal and educational neglect) or abuse (physical, sexual, and emotional abuse) served as indices of deprivation and threat, respectively. Mothers, fathers, teachers, and youth reported on children’s internalizing and externalizing problems using Strengths and Difficulties Questionnaire. For Study 2, oversampling for internalizing problems, 92 youth (M=12.44; SD = 1.55) with maltreatment (n=40) and age and gender-matched controls (n=52) were administered the Trier Social Stress Test for Children (TSST-C), providing eight salivary and two serum samples. Area Under the Curve Increase (AUCi) and pre-post TSST-C change in neuroactive steroids were analyzed, respectively. Caregivers and children completed Center of Epidemiological Studies – Depression Scales.
For Study 1, structural equation modelling indicated good model fit (RMSEA = .049; CFI = .951; SRMR = .053). Controlling for SES, age, and gender, HCC was predicted by deprivation (β = -.231, p = .004), but not threat (β = .024, p = .760; Figure 1). The effect of deprivation on externalizing problems was mediated by HCC (95% CI: .052-.088, p = .020). For Study 2, regression analyses showed that AUCi was predicted by deprivation (β = -.230, p = .040), but not threat (β = -.152, p = .198; Figure 2), controlling for SES. AUCi did not mediate effects of deprivation on externalizing problems, but threat coincided with an increased cortisol/DHEA ratio (β = .258, p = .018) which mediated the effect of threat on depressive symptoms in an OLS path-analysis (95% CI: .013-1.211).
Our data yield consistent evidence that deprivation, but not threat, coincides with attenuated long-term cumulative cortisol secretion over several months and cortisol reactivity to social stress. This attenuation – at least for cumulative cortisol – may predispose to externalizing problems. Conversely, threat may affect the interplay of DHEA and cortisol, thereby predisposing to depression.

Authors

• Lars O. White, University of Leipzig
  Presenting Author

• Charlotte Catharina Schulz, University of Leipzig
  Non-Presenting Author

• Marcus Ising, Max Planck Institute of Psychiatry, Munich
  Non-Presenting Author

• Susan Sierau, University of Leipzig
  Non-Presenting Author

• Andrea Michel, University of Leipzig
  Non-Presenting Author

• Annette Klein, University of Leipzig
  Non-Presenting Author

• Kai von Klitzing, University of Leipzig
  Non-Presenting Author

• Bertram Müller-Myhsok, Max Planck Institute of Psychiatry, Munich
  Non-Presenting Author

• Clemens Kirschbaum, Technische Universität Dresden
  Non-Presenting Author

• Manfred Uhr, Max Planck Institute of Psychiatry, Munich
  Non-Presenting Author

• Tobias Stalder, University of Siegen
  Non-Presenting Author