Poster #44 - Linking Life Course Adversity to Sleep Disturbance: A Person- and Variable-Centered Analysis

• In Event: Poster Session 18

Abstract

Background and Purpose: Research has reliably linked adverse childhood experiences (ACEs) to poor outcomes in later life, including sleep disturbance. Less is known, however, about underlying mechanisms of effect. Emerging evidence suggests that ACEs are associated with adverse adult experiences that further compromise development and functioning. Extending this line of inquiry, we employed person- and variable-centered techniques to: (a) identify patterns of life course adversity and their influence on sleep disturbance, and (b) assess whether adult adversity mediates the association between childhood adversity and sleep disturbance.

Methods: Data originated from the Families and Children Thriving Study, a longitudinal investigation of risk and resilience among low-income families in Wisconsin that received perinatal home visiting services. The analysis included a subsample of 998 women who completed an ACE assessment with their home visitor along with an assessment of adverse adult experiences at wave 1 of the study and an assessment of sleep disturbance at wave 2 approximately one year later. Dichotomous variables were created for 10 childhood adversities, including various forms of maltreatment and household dysfunction, along with 10 adult adversities such as physical and emotional partner violence, homelessness, and discrimination. Sleep disturbance was measured using the 4-item PROMIS Sleep Disturbance-Short Form.
Latent class analysis was used to explore heterogeneous patterns of lifetime adversity. Model fit was judged based on lower Bayesian information criterion (BIC), significant Lo-Mendell-Rubin adjusted LRT test, higher entropy value, and practical significance. Once an optimal class solution was identified, analyses predicting sleep disturbance were performed through an equality test of means across classes. A path analysis with a bootstrapping procedure (n=2,000) was conducted to determine whether adult adversity mediated the association between ACEs and sleep disturbance. All analyses were conducted via Mplus 8.8.

Results: A four-class model fit the data well (BIC=21649.5; LRT=425.9, p=.001; Entropy=.84), with classes representing (1) low life course adversity (34.1%); (2) moderately high childhood adversity and moderately low adult adversity (18.5%); (3) moderately low childhood adversity and moderately high adult adversity (27.9%); (4) high life course adversity (19.5%). Mean sleep disturbance T-scores were 46.9 for class 1, 52.5 for class 2, 52.3 for class 3, and 56.1 for class 4, indicating greater lifetime adversity was associated with poorer sleep. Except for a comparison of classes 2 and 3, all between-class differences in sleep scores were significant (p < .001).
The path analysis confirmed significant linkages between ACEs and sleep disturbance (ß=.149, SE=.032), adult adversity and sleep disturbance (ß=.256, SE=.032), and childhood and adult adversity (ß=.404, SE=.026). A mediation test using the Delta method showed that the direct and indirect paths were statistically significant, signifying partial mediation. About 41% of the main effect was explained by the indirect pathway.

Conclusions and Implications: Differential exposure to childhood and adult adversity appears to be implicated in sleep disturbance through direct and indirect pathways, pointing to potential opportunities to intervene throughout the life course. Further research is needed to explore the extent to which other domains of development and functioning are affected by serial risk processes.

Authors

• Joshua Mersky, University of Wisconsin - Milwaukee

• ChienTi Plummer Lee, University of Wisconsin - Milwaukee
  Presenting Author