CALIFORNIA
GRANT

IMPROVING THERAPIST FIDELITY DURING EVIDENCE-BASED PRACTICE (EBP) IMPLEMENTATION
2005-2010

Investigators:
Dr. Carolyn Webster-Stratton & Dr. Jamila Reid, University of Washington
Dr. Michael Hurburt & Dr. John Landsverk, The Child & Adolescent Services Research Center (CASRC)
Dr. Bill Carter, Lynne Marsenich, M.A., & Dr. Todd Sosna, California Institute for Mental Health (CIMH)

Aim 1: To evaluate the effectiveness of the IY parent program for treating child conduct problems in an ethnically diverse TANF population compared to usual services (CONTROL).

Aim 2: To evaluate the additional contribution to child and family outcomes of the enhanced relative to the standard model (EIM vs. SIM) for implementing the IY parenting program.

Aim 3: To evaluate the relationship between treatment fidelity and child and family outcomes, and to estimate the additional contribution to intervention fidelity of the EIM relative to the SIM condition.

Improving our understanding of factors that influence the fidelity and effective implementation of evidence-based practices (EBP) is an important national health care priority. Research suggests that fidelity is compromised when interventions are transported from the research laboratory to "real world" settings. The Incredible Years (IY) parenting program is a model EPB program that has repeatedly been shown to be effective in treating children's conduct problems in randomized control group trials. This program has been selected for study because an independent APA review board reported that it was one of only two empirically validated treatments available for young children with conduct problems. However, little is known about its transportability to community mental health systems or the fidelity with which it is delivered by therapists in these settings. The aim of the current study is to experimentally test two IY implementation models for achieving treatment fidelity and positive child and parent outcomes. Finding effective implementation models to successfully treat early onset conduct problems is a major public health concern because this disorder is highly predictive of long-term antisocial behavior, including delinquency, drug abuse, and violence.

Ultimately, effective implementation of an EBP intervention in community mental health settings depends upon therapists learning the program and delivering it with adherence to the content and with competence in the therapeutic methods and processes. Even for well developed interventions, such as the IY, which has well developed training materials and a documented training procedure, high fidelity implementation may require more intensive efforts. Key factors hypothesized to affect implementation effectiveness (i.e., fidelity) include delivery of all program content, adequate organizational supports for therapists, attention to the fit between the values of the program and those of the organization and therapists, and development of therapist skills and commitment to implement the program with fidelity.

This randomized, controlled trial compares two models for implementing the IY treatment program with a usual care control condition. Organizations delivering the program under the Standard Implementation Model (SIM) will receive detailed IY materials and will participate in one-time intensive training. The Enhanced Implementation Model (EIM) includes all elements of SIM plus organizational support, promotion of the fit between therapist and organizational values and the IY program, and ongoing training, technical support, and consultation.

This study will test these two IY implementation models in a sample of 720 children with conduct problems whose families are receiving Temporary Assistance for Needy Families (TANF) in California. Many children in this population will be at very high-risk for current or future conduct problems because they are living in poverty. Two-thirds of families selected for the study will be randomly assigned to participate in the IY parenting program, paid for by Medicaid, while 1/3 will be followed naturalistically with normal TANF services (CONTROL). Among families participating in the IY conditions, half will receive these services from therapists in an agency randomized to SIM. The other half will receive services from therapists in an agency randomized to EIM. Both implementation models proposed in this study fit within the constraints faced by "real world" agencies, reducing the intensity of control and supervision utilized in prior efficacy studies but taking advantage of a range of technology to mimic those controls in a more cost-effective manner.