Table 7.

Parameter Estimates and Odds Ratios for Alcohol or Drug Treatment as a Covariate Predicting Transitions in Drug Use Latent Class Membership

β

	Time 2 Latent Class Membership
Time 1 Latent Class Membership	Non-users	Smokers	Moderate Drinkers/ Smokers	CCH Users

Non-users	1.00^a	1.00^a	1.00^a	1.00^a
Smokers	0.00^b	—	0.00^b	0.81
Moderate Drinkers/Smokers	0.00^b	1.76	—	0.00^b
CCH Users	0.00^b	0.00^b	−0.77	—

Odds Ratio

	Time 2 Latent Class Membership
Time 1 Latent Class Membership	Non-users	Smokers	Moderate Drinkers/ Smokers	CCH Users

Non-users	1.00^a	1.00^a	1.00^a	1.00^a
Smokers	1.00^b	—	1.00^b	2.25
Moderate Drinkers/Smokers	1.00^b	5.84	—	1.00^b
CCH Users	1.00^b	1.00^b	0.46	—

Note: CCH is crack, cocaine or heroin. Dashes indicate the reference latent class.

The multinomial logistic regression was skipped for this row of the transition maxtrix.

This transition probability was constrained to be equal to zero.