a1 University of Southern California, Los Angeles
a2 Cincinnati Children's Hospital Medical Center
a3 Pennsylvania State University
Inconsistencies exist in literature examining hypothalamic–pituitary–adrenal (HPA) axis activity in children and adults who have experienced childhood abuse. Hence, the extent and manner to which childhood abuse may disrupt HPA axis development is largely unknown. To address these inconsistencies, the developmental course of nonstress cortisol in a long-term longitudinal study was assessed at six time points from childhood through adolescence and into young adulthood to determine whether childhood abuse results in disrupted cortisol activity. Nonstress, morning cortisol was measured in 84 females with confirmed familial sexual abuse and 89 nonabused, comparison females. Although dynamically controlling for co-occurring depression and anxiety, hierarchical linear modeling (HLM) showed that relative to comparison females, the linear trend for abused females was significantly less steep when cortisol was examined across development from age 6 to age 30, t (1, 180) = −2.55, p < .01, indicating attenuation in cortisol activity starting in adolescence with significantly lower levels of cortisol by early adulthood, F (1, 162) = 4.78, p < .01. As a more direct test of the attenuation hypothesis, supplemental HLM analyses of data arrayed by time since the disclosure of abuse indicated that cortisol activity was initially significantly higher, t (1, 425) = 2.18, p < .05, and slopes were significantly less steep t (1, 205) = −2.66, p < .01, for abused females. These findings demonstrate how the experience of childhood abuse might disrupt the neurobiology of stress, providing some support for the attenuation hypothesis that victims of abuse may experience cortisol hyposecretion subsequent to a period of heightened secretion.
c1 Address correspondence and reprint requests to: Jennie G. Noll, Cincinnati Children's Hospital Medical Center, Department of Pediatrics, Division of Behavioral Medicine and Clinical Psychology, Center for Epidemiology and Biostatistics, 3333 Burnet Avenue, MLC 3015, Cincinnati, OH 45229-3039; E-mail: firstname.lastname@example.org.
This research was supported by the National Institutes of Health (R01 MH048330; R03 HD045346), Department of Health and Human Services (ACYF 90CA1686l), the W. T. Grant Foundation, and the Smith Richardson Foundation.