Skip to main content
. Author manuscript; available in PMC: 2017 Mar 1.
Published in final edited form as: J Subst Abuse Treat. 2015 Nov 25;62:55–61. doi: 10.1016/j.jsat.2015.11.002

Table 2.

Results of Hierarchical Logistic Regression

Model 1 Model 2

OR (95% CI) OR (95% CI)
Clinical characteristics
 Outpatient (Yes) 0.80 (0.45, 1.43) 0.80 (0.43, 1.48)
 Previous treatment history 0.40 (0.22, 0.71) 0.43 (0.23, 0.82)
 Dual disorders 2.84 (1.44, 5.62) 2.64 (1.28, 5.42)
 Abstinence self-efficacy 0.98 (0.96, 0.99) 0.98 (0.96, 0.99)
 Trauma symptoms 1.01 (0.996, 1.02) 1.01 (0.99, 1.02)

Personal network characteristics at T3
 # of treatment related alters 0.95 (0.88,1.02)
 # of substance using alters 1.08 (1.02, 1.14)
 # of alters providing concrete support 1.04 (0.99, 1.09)
 # of alters providing sobriety support 0.95 (0.88, 1.03)
 # of critical alters 1.17 (0.89, 1.54)
 # of reciprocal relationships 0.98 (0.92, 1.04)
 # of very close alters 0.99 (0.93, 1.05)
 Density 0.98 (0.97, 0.998)
 Isolates 0.92 (0.86, 0.99)
 Components 1.26 (0.84, 1.88)

Note. Model 1, χ2(5) = 38.14, p < .001. Pseudo R2 = .126; Model 2, χ2(15) = 71.06, p < .001. Pseudo R2 = .221, Δχ2(10) = 32.92, p <.001