Testing the Constraint Theory of Addiction: Cannabis Constraints Discriminate Users from Nonusers and Heavy from Light Users

Table 3

Final logistic regression models predicting three levels of cannabis use.

	B	S.E.	Wald	df	Sig.	Exp (B)
Users vs. nonusers
Gender	−0.419	0.536	0.611	1	0.434	0.657
Tobacco last 12 months	0.566	0.158	12.780	1	0.000	1.762
Alcohol last 12 months	0.285	0.140	4.105	1	0.043	1.329
Addiction concerns	−1.462	0.316	21.362	1	0.000	0.232
Use legal highs	1.129	0.302	13.945	1	0.000	3.092
Illegality	−0.741	0.296	6.261	1	0.012	0.476
Use when others use	−0.814	0.300	7.379	1	0.007	0.443
Constant	1.186	0.836	2.011	1	0.156	3.274
Light vs. heavy users
Gender	−0.261	0.440	0.353	1	0.553	0.770
Tobacco (last 12 m)	0.091	0.076	1.424	1	0.233	1.095
Addiction concerns	−1.345	0.321	17.619	1	0.000	0.260
Know users	1.427	0.265	29.004	1	0.000	4.168
Lack of availability	−1.071	0.246	18.960	1	0.000	0.343
Unpleasant effects	−0.756	0.266	8.063	1	0.005	0.470
Too expensive	−0.642	0.227	8.025	1	0.005	0.526
Use mainly for health reasons	0.500	0.199	6.318	1	0.012	1.648
Ethics and religion	0.769	0.278	7.641	1	0.006	2.158
Constant	−1.366	0.684	3.993	1	0.046	0.255
Addicted vs. not
Gender	0.597	0.593	1.014	1	0.314	1.817
Tobacco last 12 m	−0.182	0.097	3.556	1	0.059	0.834
Addiction concerns	1.215	0.401	9.194	1	0.002	3.371
Do not know users	−1.804	0.311	33.685	1	0.000	0.165
Illegality	1.231	0.401	9.438	1	0.002	3.423
Constant	3.211	0.925	12.059	1	0.001	24.792