Trajectories of Mental Health Service Utilization Among Youth in Foster Care: A Test of Sibling Co-Placement

Background and purpose: Research has demonstrated a robust relationship between foster youth mental health symptoms and mental health service use, particularly through mid-adolescence (Burns et al., 2004; Horwitz et al., 2012; Villodas et al., 2014). However, little cross-sectional or longitudinal research has examined the influence of foster youth mental health needs on service utilization for siblings living together (“sibling co-placement”) vs. separately. The current study examined the time-based influence of foster youth mental health symptoms on mental health service utilization trajectories in a sample of 328 early to mid-adolescent foster youth participating in a dyadic sibling intervention in a metropolitan Pacific Northwestern city over a period of 18 months. A specific study aim involved testing for sibling co-placement, controlling for mental health symptomatology and other sociodemographic characteristics.

 

Methods: Quantitative data were drawn from the Supporting Siblings in Foster Care study, a longitudinal RCT of a sibling relational enhancement intervention (Kothari et al., 2017). Siblings ranged from 7-15 years old at study enrollment; about 60% of participants were youth of color. At baseline, caregivers completed the Child Behavioral Checklist (CBCL) internalizing and externalizing subscales, and reported on child age, gender, race/ethnicity, placement with relative (or non-relative), and sibling co-placement status. Over 10, 60-day data collection waves, caregivers also reported on youths’ number of outpatient mental health services. This service utilization count variable was used as a dependent variable in a multilevel mixed-effects Poisson regression model to examine whether baseline CBCL total scores and sibling co-placement were associated with mental health service use over time.

Results: At baseline, the CBCL total score was 60.4 (range 25-86), and 72% of siblings were co-placed. Over time, the number of outpatient mental health services ranged from 5.4 (Time 1) to 3.2 (Time 10). The multivariate regression model showed a significant CBCL by time interaction, such that siblings with lower baseline CBCL scores showed a decline in service use over time, while those with higher baseline CBCL scores showed a stable, high level of mental health service use over time. Being female and sibling co-placement were associated with significant time-based increases in service use.

Conclusions and implications: Findings suggest that the relationship between mental health needs and mental health service receipt may vary by time, with higher needs translating to prolonged, higher levels of service use. Additionally, initial sibling co-placement may be associated with increased service utilization over time. The current study thus provides the first descriptive and multivariate findings concerning the incidence of mental health services for siblings living together vs. separately. In regards to implications for practice, because research has suggested that siblings in foster care can experience interpersonal conflicts and have substantial socioemotional needs (McBeath et al., 2014), co-placed siblings may particularly benefit from mental health supports. Future research is needed to test whether sibling relationship quality and consistency may directly or indirectly buffer the influence of sibling co-placement on mental health services.