Table 3.

Observed values at each assessment by safety counts or VCT

	Observed mean or percent		Results from mixed-effects models
	Baseline (N = 1,116)	Follow-up (N = 875)	Estimated effect size‡	Model t-statistic	Significance of effect
Counts of risky behaviors
Number of sex acts
VCT	10.21	9.29	0.78	7.76	<0.001
Safety counts	9.61	7.28
Number of unprotected sex acts
VCT	7.26	7.23	0.58	14.22	<0.001
Safety counts	7.15	4.50
Times injected drugs
VCT	46.5	34.5	0.93	4.36	<0.001
Safety counts	46.1	31.7
Times injected drugs with dirty works
VCT	4.66	3.87	0.51	12.56	<0.001
Safety Counts	5.08	1.85
Times used crack
VCT	38.6	21.6	0.90	6.01	<0.001
Safety counts	46.7	23.4
Times used alcohol or marijuana
VCT	35.5	23.0	1.03	1.49	0.14
Safety counts	32.8	22.8
Times used other drugs
VCT	95.5	62.9	0.93	6.25	<0.001
Safety counts	99.4	60.2
Binomial outcomes
Had a negative urine test for cocaine
VCT	16%	26%	1.51	1.73	0.08
Safety counts	13%	29%
Had a negative urine test for opiates
VCT	53%	60%	1.09	0.42	0.68
Safety counts	52%	61%
Practiced safe sex
VCT	43%	54%	1.19	0.92	0.36
Safety counts	42%	57%

Estimated effect sizes and significance

^‡Estimated effect size = (SC_FU/SC_BL)/(VCT_FU/VCT_BL) where SC and VCT indicate Safety Counts and VCT, respectively

BL and FU indicate values at baseline and follow-up, respectively. For outcomes measured as counts, the values represented by SC and VCT are means. For binomial outcomes, the values represented by SC and VCT are odds