National Trends in Seclusion and Restraint Utilization within Child Residential Treatment Centers

Background & Purpose: Reduction of seclusion and restraint (SR) usage with clients accessing mental healthcare is an ongoing and robust area of scholarship that qualifies as a component of trauma-informed care (TIC), as harms associated with the use of SR practices have been well-documented within the literature. A majority of scholarship on this topic has examined the use of SR practices on adults within inpatient hospital settings, which may not be generalizable to SR usage within child residential treatment centers (RTCs). To date, national trends in SR use in child RTCs have not been examined over time. In order to address this gap, the purpose of this study was to examine national trends in the use of SR within child RTCs over a decade. The study aimed to estimate the percentage of child RTCs that used SR practices and examine predictors associated with increased likelihood of a facility utilizing SR between 2010-2020.

Methods: A secondary analysis of a subset of the National Mental Health Services Survey (N-MHSS) from 2010-2020 was conducted (n-range=580-781). The N-MHSS is an annual census of all mental health treatment facilities within the United States that collects administrative data for the purpose of providing information to individuals accessing behavioral health services. The outcome variable examined was usage of SR practices within the previous year. Client-level predictive factors examined included facility percentage of male, white, and involuntarily committed clients, while facility-level predictive factors examined included programming for youth with severe emotional disturbances (SED), use of psychotropic medication, facility ownership, acceptance of Medicare, and number of mental health beds. Separate multi-level logistic regression was performed by year with a random intercept by state included to account for clustering within states.

Results: One-way ANOVA regression indicated that the percentage of child RTCs using SR significantly decreased between 2010-2020 [F(4, 58074)=75.62, p<.001]. A post-hoc Bonferroni test found that the percentage of facilities reporting the use of SR in 2010 was significantly higher than in 2014, 2016, 2018, and 2020 (all p<.001). Multi-level logistic regression analysis found that facility client demographics (percent male, white, and involuntarily committed) did not significantly predict facility use of SR in 2010, 2014, or 2016. However, the facility-level factors of SED program (all p<.01), use of psychotropic medication (all p<.001), and facility size as measured by number of mental health beds (all p<.01) consistently predicted usage of SR in 2010 [F(11)=68.38, p<.001], 2014 [F(12)=74.25, p<.001], and 2016 [F(12)=74.48, p<.001]. Facility ownership and accepting Medicare payments did not significantly predict facility usage of SR.

Conclusions & Implications: Use of SR within child RTCs consistently declined between 2010-2020. Additionally, larger facilities, facilities with an SED program, and facilities that used psychotropic medications with clients were more likely to use SR practices between 2010-2016. This finding indicates a pattern of children with more severe mental health symptomology experiencing increased likelihood of SR, making them more likely to experience retraumatization at the hands of the mental healthcare system. Further research on national trends in SR practices within child RTCs is needed.