Pediatric TBI Treatments: Optimal Timing, Targets, and Patient Characteristics.

摘要

Abstract The evidence base regarding psychosocial treatment for behavior problems that arise after pediatric traumatic brain injury (TBI) has been limited by poorly controlled designs and small and heterogeneous samples. However, a series of pilot randomized clinical trials (RCT) and four larger, multi-site RCTs by our group have underscored the potential promise of family-centered treatments (Online Family Problem Solving (OFPS) and Online Parent Skills Training (OPST)) in reducing behavior problems, executive dysfunction, and caregiver distress following pediatric TBI. Innovative joint analyses of combined data from these studies would provide the statistical power to develop and test models of treatment effects that incorporate individual, injury, and social environmental characteristics together with treatment adherence. We propose to conduct secondary data analyses of eight RCT conducted between 2001 and 2015 involving 767 children between the ages of 3 and 19 years who were hospitalized overnight for complicated mild to severe TBI. Outcomes were assessed pre-treatment and 6 months later in all studies, with several of the studies assessing maintenance of treatment effects at follow-up time points 3-6 months after treatment completion. Parents completed measures of child behavior problems and executive function behaviors. Older children completed self-report measures of depression, behavior, and executive functioning. Parents also completed ratings of caregiver distress, depression, and family functioning. Data were also collected on adherence to treatment, receipt of other treatments, and school services for all participants. We will employ modern longitudinal multiple imputation models to harmonize measurements across studies prior to joint cross-study analysis to address the following aims: 1) examine the association of time since injury with treatment response; 2) examine the relationship of patient characteristics and premorbid conditions (e.g., ADHD, LD) to treatment response; 3) Identify which aspects of child behavior (internalizing, externalizing, executive function skills) are most sensitive to the effects of OFPS and OPST; and 4) Develop a comprehensive model of predictors of behavioral recovery over time that incorporates the effects of treatment along with other rehabilitative therapies. Analysis models that synthesize all eight studies will employ mixed effects to capture cross-study variability, and will include longitudinal correlations, nonlinear means, and profile summaries as appropriate to identify treatment effects and treatment effect heterogeneity. Findings will provide important new information about how to optimize and tailor family-centered treatments for post-TBI behavior problems. It will also enhance our understanding of trajectories of behavioral recovery following pediatric TBI and allow us to identify measures/subscales that are most sensitive to recovery. Given the inherent difficulty and frequent failure of RCTs for pediatric TBI, the proposed project has the potential to inform individualized management of pediatric TBI and substantively move the field forward.