Table 3.

Change in alcohol and drug outcomes between initial interview and 3 months and initial interview and 6 months

	Alcohol abstinence		ASI alcohol scores		ASI drug scores, full sample		ASI drug scores, subsample that reports drug use at any timepoint
	With all covariates	With significant covariates only	With all covariates	With significant covariates only	With all covariates	With significant covariates only	With all covariates	With significant covariates only
Change, Initial interview to 3 months	1.065** (0.410)	0.968** (0.371)	-0.105** (0.0283)	-0.104** (0.0284)	-0.0228* (0.0114)	-0.0239* (0.0114)	-0.0266 (0.0158)	-0.0289 (0.0157)
Change, Initial interview to 6 months	1.213** (0.380)	1.125** (0.343)	-0.128** (0.0258)	-0.127** (0.0258)	-0.0150 (0.0105)	-0.0153 (0.0104)	-0.0258 (0.0139)	-0.0264 (0.0138)
Baseline characteristics HIV status: Positive	-0.292 (0.594)	N/A	0.028 (0.065)	N/A	0.0329* (0.0157)	0.0164 (0.0154)	0.0448* (0.0193)	0.0234 (0.0188)
Income	0.000 (0.000)	N/A	0.000 (0.000)	N/A	0.0000 (0.0000)	N/A	0.0000 (0.0000)	N/A
Age	0.060 (0.039)	N/A	-0.003 (0.004)	N/A	-0.0019 (0.0010)	N/A	-0.0020 (0.0012)	N/A
Gender: Female	-0.505 (0.513)	N/A	-0.044 (0.055)	N/A	-0.0079 (0.0133)	N/A	-0.0081 (0.0168)	N/A
Race: Black	-1.369 (1.273)	N/A	0.004 (0.130)	N/A	0.0227 (0.0333)	N/A	0.0407 (0.0355)	N/A
Race: White	-1.207 (1.281)	N/A	0.082 (0.131)	N/A	0.0128 (0.0335)	N/A	0.0219 (0.0361)	N/A
Observations	133	133	133	133	133	133	94	94
Participants	1.065**	0.968**	-0.105**	-0.104**	-0.0228*	-0.0239*	-0.0266	-0.0289

^**p<0.01

^*p<0.05

Note: Logistic GEE regression analysis used for Alcohol Abstinence, which is a dichotomous variable in which 1=abstinence. Multiple GEE regression analysis used for ASI Alcohol and Drug Scores, with higher scores indicating greater alcohol and drug use. For Race, “Black” is an indicator variable for Black participants versus White and other races, and “White” is an indicator variable for White participants versus Black and other races.