Improving measurement-based care implementation in youth mental health through organizational leadership and climate: a mechanistic analysis within a randomized trial

Table 3 Direct and indirect effects of the Leadership and Organizational Change for Implementation (LOCI) strategy on clinic implementation climate and digital MBC fidelity

LOCI effect	Outcome: T₄ clinic implementation climate^a								Outcome: MBC fidelity index^b
	Mediator: T₂ implementation leadership				Mediator: T₂ transformational leadership				Mediator: T₄ clinic implementation climate
		95% CI				95% CI				95% CI
	Coeff	LL	UL	p	Coeff	LL	UL	p	Coeff	LL	UL	p
Indirect effect	0.51	0.15	0.95	.004	0.16	− 0.05	0.45	.135	0.14	0.01	0.37	.033
Direct effect	0.11	− 0.29	0.51	.582	0.38	0.05	0.72	.024	0.05	0.00	0.23	.482
Proportion mediated	0.82	0.25	1.00	.004	0.29	0.00	0.86	.136	0.71	0.02	1.00	.045

Effect estimates generated using Imai’s multilevel causal mediation R package [88]. Estimates in bold are statistically significant at p < .05 (two-tailed). All models include implementation condition (random assignment to LOCI vs. training and technical assistance only) as the focal antecedent plus covariates of state, clinic size, and baseline value of the mediator. CI, confidence interval; MBC, measurement-based care; LL, lower limit; T₂, 4-month post-baseline; T₄, 12-month post-baseline; UL, upper limit
^aK = 20 clinics, N = 166 clinicians. These models also include baseline clinic implementation climate as a covariate. Indirect and direct effects indicate the change in T₄ implementation climate (range = 0–4) caused by LOCI
^bK = 17 clinics, N = 231 youth. Indirect and direct effects indicate the proportion difference in MBC fidelity (range = 0 − 1) caused by LOCI

ISSN: 1748-5908