Reduction in depressive symptoms predicts improvement in eating disorder symptoms in interpersonal psychotherapy: results from a naturalistic study

Table 3 Fixed effects from Dynamic Structural Equation Models testing the session-wise lagged relationships between symptoms of eating disorder and depression

	Model 1				Model 2
Fixed effects, within level	Estimate	sd	p	95% CI	Estimate	sd	p	95% CI
PHQ_t-1 → PHQ_t	0.58	0.05	<.001	0.48, 0.69	0.44	0.08	<.001	0.29, 0.59
REDS_t-1 → PHQ_t	0.24	0.05	<.001	0.14, 0.34	−0.01	0.08	.92	−0.16, 0.16
REDS_t-1 → REDS_t	0.73	0.05	<.001	0.63, 0.83	0.20	0.08	.02	0.04, 0.36
PHQ_t-1 → REDS_t	0.13	0.05	.008	0.04, 0.23	0.13	0.06	.04	0.00, 0.25
PHQ_t ↔ REDS_t	0.15	0.017	<.001	0.12, 0.19	0.12	0.01	<.001	0.10, 0.16
Fixed effects, between level
PHQ_intercept ↔ REDS_intercept	0.089	0.211	0.52	−0.10, 0.69	0.68	0.41	.01	0.15, 1.71
PHQ_slope ↔ REDS_slope					0.00	0.00	.04	0.00, 0.01
PHQ_intercept ↔ REDS_slope					−0.02	0.02	.26	−0.07, 0.02
PHQ_slope ↔ REDS_intercept					−0.02	0.02	.08	−0.07, 0.00
PHQ_intercept ↔ PHQ_slope					−0.03	0.02	.002	−0.09, − 0.01
REDS_intercept ↔ REDS_slope					−0.06	0.03	<.001	−0.13, − 0.02

Note. PHQ_t-1 and REDS_t-1 are the PHQ-9 and REDS scores, respectively, at session t-1, while PHQ_t and REDS_t are the PHQ-9 and REDS scores at scores at session t. Unidirectional arrows (→) represent regression paths (potentially causal), while bidirectional arrows (↔) represent covariances

ISSN: 2050-2974