Table 4.

Piecewise growth-curve modeling of maladaptive caregiver variables as predictors of child symptom change (CBCL) over the course of treatment and 12-month follow-up

Outcome
Independent Variable	B	SE	t	p	95% CI
Externalizing Symptoms
Change in Symptoms over Treatment (0, 3, 6 months)
Blame of Child Phase 1	−0.56	0.46	−1.36	.232	[−1.47, 0.53]
Blame of Child Phase 2	0.02	0.44	0.04	.969	[−0.84, 0.88]
Avoidance Phase 1	−0.36	0.52	−0.73	.470	[−1.39, 0.67]
Avoidance Phase 2	−0.70	0.44	−1.59	.117	[−1.73, 0.33]
Change in Symptoms over Follow-Up (6, 9, 12 months)
Blame of Child Phase 1	0.59	0.62	0.97	.338	[−0.63, 1.81]
Blame of Child Phase 2	1.34	0.54	2.46	.016	[0.28, 2.40]
Avoidance Phase 1	2.00	0.77	2.59	.012	[0.49, 3.51]
Avoidance Phase 2	1.88	0.53	3.58	<.001	[0.84, 2.92]
Internalizing Symptoms
Change in Symptoms over Treatment (0, 3, 6 months)
Blame of Child Phase 1	−0.48	0.34	−1.40	.166	[−1.15, 0.19]
Blame of Child Phase 2	0.26	0.33	0.80	.425	[−0.37, 0.89]
Avoidance Phase 1	−0.52	0.37	−1.40	.167	[−1.25, 0.21]
Avoidance Phase 2	−0.15	0.33	−0.47	.643	[−0.80, 0.50]
Change in Symptoms over Follow-Up (6, 9, 12 months)
Blame of Child Phase 1	0.38	0.35	1.10	.274	[−0.31, 1.07]
Blame of Child Phase 2	0.64	0.31	2.07	.042	[0.03, 1.25]
Avoidance Phase 1	0.22	0.45	0.50	.621	[−0.66, 1.10]
Avoidance Phase 2	0.95	0.30	3.12	.003	[0.36, 1.54]

Note. Each caregiver variable [Blame of Child and Avoidance] was entered in a separate hierarchal piecewise regression model. Phase 1 (stabilization and skills-building phase) and phase 2 (trauma narration and processing phase) means for each variable were entered simultaneously as level-2 (person-level) predictors of change in either externalizing or internalizing symptoms. Symptom change was separated into two pieces: change over treatment, which included 0, 3, and 6 month assessments and change over follow-up, which included 6, 9, and 12 month assessments. CBCL = Child Behavior Checklist