Health service costs and clinical gains of psychotherapy for personality disorders: a randomized controlled trial of day-hospital-based step-down treatment versus outpatient treatment at a specialist practice

Table 6 Estimated differences in clinical gains for patients in step-down and outpatient treatment

		∆ GAF at 0 months		∆ GAF-change rate per month during trial		Model fit
		∆ mean estimate (SE)	P	∆ Slope (SE)	P	AIC
(A) Global functioning (GAF)	∆ (SDC-OPC)	1.3 (2.1)	ns	−0.4 (0.1)	.00	2586
	Moderator effects
	∆ (SDC_APD - OPC_APD)	1.6 (3)	ns	−0.5 (0.2)	.00	2583
	∆ (SDC_{BPD -} OPC_BPD)	0.6 (3)	ns	−0.3 (0.2)	ns	2594
(B) 36-month GAF change		∆ GAF (0-36-months) mean estimates (SE)				Incremental effects (SE)
		SDC	OPC
	All patients	10 (0.3)	18 (0.1)		.00	−8 (0.3)
	Moderator effects
	Predicted change: APD (not BPD)	6 (0.5)	19 (0.01)		.00	−13 (1)
	Predicted change: BPD (not APD)	14 (0.6)	17 (0.2)		ns

Table 6 shows mixed-model estimates of the Global Assessment of Functioning (GAF) in models with treatment included as a predictor (SDC: the step-down condition, OPC: the outpatient condition). (A): Mean estimated GAF (SE: standard error) at baseline (intercept) and change per month (linear slope) during the trial presented as (1) ∆ (SDC-OPC): Difference (∆) between SDC and OPC, and (2) Moderator effects: (SDC-OPC) difference for subgroups of patients with (a) avoidant PD, APD: ∆ (SDC_APD - OPC_APD) and (b) borderline PD, BPD: ∆ (SDC_BPD - OPC_BPD). Akaikes Information Criterion (AIC) indicates goodness of fit for each model. (B): 36-month GAF change: Estimations of 36-month GAF change and incremental effects (when a significant difference existed between SDC and OPC) based on mixed-model predicted values. P-values are presented if p < 0.05, ns indicates p > 0.05.

ISSN: 1471-244X