Category: Treatment - Other

PS3- #C72 - Predicting Progress of Disruptive Youth in Community-Based Residential Settings

Friday, Nov 17
11:00 AM – 12:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Evidence-Based Practice | Aggression / Disruptive Behaviors / Conduct Problems | Psychotherapy Outcome

It is long-established that more research is needed on the study of psychotherapy practices in usual care settings if we wish to improve mental health outcomes. A growing number of researchers in youth mental health have answered this call by examining usual care at the level of discrete techniques, with a specific focus on the use of “practices derived from the evidence-base” (PDEBs) or techniques common across evidence-based protocols for a given type of mental health problem (e.g., exposure for anxiety). Previous work in this area has investigated PDEBs’ association with youth improvement rates in intensive in-home settings, leaving a gap in research for youth in higher levels of care, like community-based residential (CBR) settings. Given this scarcity of CBR treatment as usual research, the current study examined the extent to which specific techniques (i.e., both PDEBs and non-PDEBs) predicted improvement for disruptive youth in CBR settings.


Participants included 341 youth served across all CBR agencies in Hawai‘i’s youth mental health system between 2006 and 2014. Most were males (79.5%) and were on average 15.8 (SD=1.2) years old. The majority of youth were multiracial (60.7%). Clinical data (i.e., discrete techniques used, progress ratings) were collected from a monthly therapist-report form built into the mental health system’s reporting infrastructure. Only youth with disruptive behavior concerns were included in this study. A technique was considered a PDEB for this study if it was found in at least 30% of studies examining evidence-based protocols for treating disruptive youth aged 13-18 (PracticeWise, 2015).


Multilevel modeling was used to investigate which discrete techniques significantly ( < 0.05) predicted improvements in progress ratings at the end of a youth’s treatment episode. The results from these multilevel models suggested that there was a limited number of PDEBs that significantly predicted the final progress rating for these youth: Family Therapy (model estimate β = 1.14), Modeling (β = 0.92), Parent or Teacher Praise (β =0.28), Response Cost (β = 0.53), and Self-Monitoring (β = 0.74). In addition, there were 33 non-PDEBs that significantly predicted end of treatment progress rating for these youth, with the top practices including Counseling (β = 2.16), Discrete Trial Training (β = 1.61), Eye Movement or Tapping (β = 1.38), Hypnosis (β = 1.45), and Supportive Listening or Client Centered (β = 1.35). A few of these non-PDEBs demonstrated a negative relationship between use and progress rating (i.e., lack of improvement), including Anger Management (β = -0.60) and Medication Management (β = -0.46). These overall findings suggest that a diversity of discrete techniques seem to predict improvement level at discharge for disruptive youth in CBR settings, with a majority of them being non-PDEBs and child-targeted. These findings for disruptive youth might be related to the comorbid diagnostic profile of the majority of the youth in the CBR setting. These and other findings will be discussed as they relate to continued practice quality improvement initiatives for these high-risk youth in CBR settings.

Sonia C. Izmirian

Graduate Student
University of Hawaii at Manoa
Honolulu, Hawaii

Kaitlin A. Hill

Graduate Student
University of Hawaii at Manoa
Honolulu, Hawaii

Brad J. Nakamura

Associate Professor
University of Hawaii at Manoa
Honolulu, Hawaii