The Impact of Placement Change on Sleep in Child Welfare

Background/Objectives: Sleep disturbance is common among youth in the child welfare system, yet the role of placement instability and placement setting in shaping sleep outcomes remains understudied. This study examined the association between placement instability, time spent in different care settings, and sleep disturbance among children in foster care. Methods: We conducted a retrospective cohort study using longitudinal administrative child welfare data from a Midwestern U.S. state, including 20,888 youth aged 5-18 years who entered foster care between 2010 and 2020. Sleep disturbance was assessed using the Child and Adolescent Needs and Strengths (CANS) sleep item. Baseline was defined as the first CANS assessment within one month of entry into care, and follow-up as the assessment closest to discharge or the end of a three-year observation window, whichever occurred first. We estimated association using a time-lagged linear mixed-effects model predicting sleep disturbance after each placement episode, including placement instability: 1 (reference), 2, 3, or ≥4 placement(s), time since placement, time spent in care settings (kinship, foster home, treatment foster home, congregate care, institutional care), and baseline trait factor scores derived from non-sleep CANS items, while controlling for sleep at the time of placement and demographics. Results: At baseline, 2016 children had actionable sleep disturbance (CANS sleep = 2 or 3; 1701 moderate and 315 severe). By the end of follow-up, this increased to 2884 children (2372 moderate and 512 severe). In linear mixed-effects models, placement instability demonstrated a dose-response association with higher subsequent sleep disturbance relative to one placement (2 placements: β = 0.025; 3 placements: β = 0.045; ≥4 placements: β = 0.067; all p ≤ 0.02). Time spent in kinship care was associated with lower sleep disturbance (β = -0.049; p < 0.001), whereas time spent in treatment foster homes was associated with higher sleep disturbance (β = 0.035; p < 0.001). Trauma in the family, medical/developmental needs, and internalizing/sexual issues were positively associated with sleep disturbance. Time and instability interactions showed modest attenuations of instability-associated sleep disturbance over time for higher placement counts. Conclusions: Placement instability is associated with progressively worse sleep disturbance over time among youth in foster care, even after controlling for sleep status at placement and baseline functioning. Sleep disturbance may represent an actionable indicator for the child welfare system, highlighting opportunities for targeted screening and support during placement transitions.