. Author manuscript; available in PMC: 2011 Oct 1.

Published in final edited form as: J Consult Clin Psychol. 2010 Oct;78(5):681–690. doi: 10.1037/a0019735

Table 5. Summary of Model-Estimated Transition Probabilities for Conditional Latent Transition Analysis.

	IPT (n = 73)^a		BWL(n = 64)		CBTgsh (n= 66)^b

	Responder	Non-Responder	Responder	Non-Responder	Responder	Non-Responder
Class 1	.99^*	.01^*	.99^*	.01	.99^*	.01^*
Class 2	.81²	.19	.60²	.40	.59¹	.41
Class 3	.61³	.29	.64²	.36	.74²	.26
Class 4	.99	.01	.99	.10	.99^*	.01^*

Note. Logits were transformed to probability scale to aid in interpretation. The probabilities should be interpreted as the probability of diagnostic group transitioning into the responder group, conditional upon assignment to a specific treatment.

Parameter fixed to extreme in model estimation

significantly < Class 1, Class 3, Class 4 p < .05.

significantly < Class 1, Class 4, p <.05.

significantly < Class 1, Class 2, Class 4, p < .05

Class 2: IPT > BWL, GSH

Class 3: GSH > BWL, IPT