Traumatic stress symptoms predict restraint incidents in children and adolescents in psychiatric residential treatment

The early identification of youth at risk for restraint incidents is an important next step to reducing the likelihood of such incidents. Yet, the extant research has not comprehensively investigated the idiographic factors that contribute to the restraint of youth in psychiatric residential treatment facilities (PRTFs). The current study investigated client-level predictors of restraint incidents, with specific emphasis on youth client trauma history and traumatic stress symptoms as assessed at admission. Participants were children and adolescents (N = 150; 55.3% female, 66.7% White, 33.3% Black or biracial) aged 6-17 (M = 11.8 years) admitted to a PRTF in the northeastern United States. A negative binomial regression with maximum likelihood estimation was conducted to examine the relative contributions of age, gender, length of stay, number of psychiatric diagnoses, body mass index (BMI), and traumatic stress symptoms at intake to the frequency of restraint incidents. The model was significant, chi(2)(6, N = 150) = 30.326, p < .001, and both length of stay, beta = .005, p < .001, IRR = 1.005, and traumatic stress symptoms at intake, beta = .072, p = .007, IRR = 1.074, were identified as significant predictors within the model. Although length of stay is an obvious predictor of restraint incidents, the current study is the first of which we are aware to identify traumatic stress symptoms at intake as a potential indicator of restraint frequency following admission. Clinical implications of these results are discussed.