Impact of an Implementation Facilitation Strategy to Improve Task-Shifted CBT Across Education and Health Sectors in Kenya: A Mixed Methods Study

Abstract

Background. Evidence suggests mental health interventions can be effectively delivered via task-shifting in low- and middle-income countries (LMIC) with high need for mental health services. However, implementation challenges have impeded adoption and sustainment in LMIC, and methods to scale and sustainably deliver interventions with minimal governmental funding are lacking. Implementation strategies for scale-up are needed. The purpose of the present study is to examine the impact of facilitation or coaching, a multifaceted implementation strategy, on improving task-shifted delivery of a mental health intervention for children in Kenya. Facilitation has helped improve the uptake and quality of medical and mental health services in the U.S.; however, no empirical studies have examined its impact on mental health program implementation in LMIC. This study adds to the sparse implementation literature on methods to reduce the substantial mental health treatment gap in LMIC. Methods. The present study employed a mixed methods quasi-experimental design to determine the impact and perceived value of coaching in supporting trauma-focused cognitive behavioral therapy (TF-CBT) implementation in two governmental sectors: education (delivery by teachers) and health (delivery by community health volunteers [CHVs]). Participants for this study’s quantitative arm included 150 lay counselors (75 teachers, 75 CHVs) from Sequences 1-4 of an NIMH-funded stepped-wedge clustered randomized controlled trial, to compare outcomes between counselors who did not received coaching (Sequence 1) to those that did (Sequences 2-4). Participants completed surveys at the end of receiving TF-CBT training (post-training), and after delivering two sequential TF-CBT groups at their sites (post-implementation). Generalized estimating equations were used to examine the relation between coaching condition and early implementation outcomes (acceptability, feasibility, appropriateness), provider-level determinants (self-efficacy, behavioral intentions), and organization-level determinants (organizational readiness, implementation climate, implementation leadership) of TF-CBT implementation. Semi-structured interviews were conducted with 32 lay counselors (16 teachers, 16 CHVs) from Sequences 2-3 to explore perceptions of acceptability, feasibility, and utility of the coaching strategy itself. Results. Quantitative results revealed that coaching condition predicted higher acceptability and feasibility of TF-CBT among teacher counselors at post-training, and lower feasibility, appropriateness and self-efficacy to implement TF-CBT among CHV counselors at post-implementation. Coaching condition did not predict differences in counselor-reported behavioral intentions, organizational readiness, implementation climate or implementation leadership. Qualitative results indicated that coaching was perceived as highly acceptable, feasible and useful by both teacher and CHV counselors that delivered TF-CBT. Counselor perspectives indicated that coaching was most helpful in increasing counselor readiness to implement TF-CBT and developing tailored implementation workplans to target barriers throughout implementation. Conclusion. To our knowledge, this is the first global study to assess the impact of an implementation coaching strategy to successfully deliver a child mental health intervention in an LMIC. Findings from this study have significant implications on replicable methods that can be used to provide tailored implementation support for task-shifted interventions in LMIC.