Category: Dissemination / Implementation

PS7- #C83 - Barriers to Evidence-Based Practice Delivery in Community Mental Health: Multiple Predictor Models

Friday, Nov 17
4:00 PM – 5:00 PM
Location: Indigo Ballroom CDGH, Level 2, Indigo Level

Keywords: Implementation | Evidence-Based Practice | Community-Based Assessment / Intervention

Recent efforts to implement EBPs in community mental health settings have yielded mixed results (Southam-Gerow et al., 2012), which might be related to challenges with client behaviors in these settings. Further research is needed to better understand how therapist, client, and practice characteristics are related to barriers to EBP delivery. By better understanding predictors of barriers, we hope to inform future EBP implementation efforts on how to address numerous types of client barriers. The current study uses survey data collected as part of a five-year NIMH-funded study in order to identify potential predictors of therapist-reported barriers to EBP delivery.

A survey was completed by 668 therapists who reported on barriers experienced over the past two months for one client/caregiver with whom they delivered one of six EBPs under investigation. Therapists responded to seven (7) items inquiring about two types of barriers to EBP delivery: expressed client barriers (e.g., verbalize a lack of familiarity of concepts presented in therapy) and disengaged client barriers (e.g., avoid participating in therapy activities). The therapists and clients were ethnically diverse (42.8% and 69.6% Latino respectively).

Separate Poisson regression models were used to test the predictive validity of therapist, service, and client characteristics on expressed and disengaged client barriers to EBP delivery.

For the disengaged client barrier model, Latino therapist ethnicity, β = -.35, SE = .12, p = .003, and favorable attitudes towards the EBP, β = -.19, SE = .06, p = .001, were related to fewer perceived barriers. Higher levels of therapist emotional exhaustion significantly predicted a greater number of disengaged client barriers, β = .09, SE = .03, p = .013. Externalizing child presenting problem was the only client characteristic related to more disengaged client barriers, β = .24, SE = .11, p = .023.

For expressed client barriers, therapist use of Triple P with the identified client, β = .416, SE = .14, p = .003, higher levels of openness towards EBPs in general, β = .213, SE = .07, p = .004, and larger caseload β = .012, SE = .01, p = .019, were related to a greater number of client barriers.  Additionally, higher therapist divergence from EBPs in general, β = -.176, SE = .06, p = .005, and higher number of direct service hours, β = -.017, SE = .01, p = .014, were related to a lesser number of expressed client barriers.

Our results suggest that there are factors at the client, therapist, and service levels which each influence the number and types of barriers to EBP delivery. Understanding how to promote engagement by addressing challenges at each level could improve EBP implementation in the community. 

Juan C. Gonzalez

Project Coordinator
University of California, Los Angeles
Los Angeles, California

Anna S. Lau

University of California, Los Angeles

Dana Saifan

Graduate Student
University of California, Los Angeles

Miya Barnett

Assistant Professor
University of California, Santa Barbara
Santa Barbara, California

Lauren Brookman-Frazee

Associate Professor
University of California, San Diego