3.1. Baseline demographic and clinical characteristics of sample
provides baseline demographic and clinical data on the sample. Forty-one patients (31 male; 10 females) agreed to participate in the study. In terms of ethnicity, 4 patients identified themselves as white, 35 as African American, and 2 as Other. The average age of patients was 42.4 years with an average of 12.7 years of education. Overall 18 patients (44%) were employed, 31 patients (76%) lived alone, 29 patients (71%) were crack users, and the average days of cocaine use in the month prior to treatment was 12.7 days.
Demographic and Clinical Characteristics of Sample (N=41)
3.2. Acceptability of Treatment
Acceptability was examined by the Acceptability of Treatment questionnaire. Responses from the questionnaire indicated that patients felt the treatments/counselors were not confusing (M = 1.4, SD = 0.80), did not send mixed message about whether one should take control of one’s addiction or give up control to a “higher power” (M = 1.2.; SD = 0.59), were clear (M = 4.3; SD = 1.17), presented a coherent model of addiction and treatment (M = 4.2; SD = 1.20), were sensible (M = 4.4; SD = 1.08), and did not have elements that were non-understandable (M = 1.6; SD = 1.00) (all of these means were in the favorable direction on the 1 to 5 scale). One item (“My counselor uses a variety of unrelated techniques in sessions”) was rated in the middle of the scale (M = 2.3, SD=1.45), suggesting that, to some extent, that unrelated techniques were used during sessions, but this mixing of techniques did not confuse patients.
Attrition rates after randomization and by study month are given in . Overall, attrition was high, with half of patients in the IDC + GDC group and 62% of those in the GDC group dropping out of treatment in the first month. All patients had dropped out of treatment by 3 months.
Number (%) of Patients who Dropped Out of Treatment
3.4. Counselor adherence/competence
Adherence/competence ratings made by the clinical supervisor were available for 14 of the 20 patients who received IDC (the other 6 cases were treated by the supervisor/expert). Selecting session 2 and one additional session between 3 and 10 (chosen randomly), we compared ratings of adherence and competence (average of 34 items each) for the 14 patients in the current study. Each item on these scales is rated on a 1 to 7 scale. The average adherence in the current study (M = 3.8, SD = 0.36) and the average competence (M = 3.9; SD =0.29) were high. The average level of adherence by the GDC counselor to GDC techniques was also very high (M = 6.2; SD = 0.37), with over 44 group sessions rated spanning the duration of the project.
3.5. ASI Outcomes
The primary efficacy measure was the ASI Drug Use Composite. The ASI Drug Use Composite scores are given in . Overall, both groups improved during the 3 month treatment period. ASI Drug Use scores decreased from 0.25 (SD = 0.07) (IDC + GDC) and 0.23 (SD = 0.07) (GDC) to 0.13 (SD= 0.08) and 0.16 (SD = 0.10), respectively.
ASI drug use composite scores over 3 months of treatment.
Abstinence was evaluated using a composite abstinence measure integrating data from urines, ASI, and cocaine inventory. The measure was scored as “abstinent” or “not abstinent” on a monthly basis based on any indication of cocaine use across the three measures.
In the IDC + GDC group, 17 patients (85%) never achieved a full month of abstinence from cocaine. One patient (5%) achieved one month of abstinence and 2 patients (10%) achieved two consecutive months of abstinence from cocaine in the IDC + GDC group. For the GDC alone group, 13 patients (62%) failed to achieve a single month of abstinence, 5 (24%) achieved one month of abstinence, 2 (10%) achieved two consecutive months of abstinence, and 1 (5%) achieved 3 consecutive months of abstinence from cocaine.
3.7. Other Secondary Outcomes
Results for secondary outcome measures are given in . On all measures, there was improvement over time. With the limited sample size, no comparisons of the treatment groups reached statistical significance. However, there was a tendency for the -IDC + GDC group to fare better on measures of psychiatric severity. At 3 months, the effect size (Cohen’s d) for IDC + GDC vs. GDC was 2.5, a very large effect.
Mean (SD) Scores on Secondary Outcome Measures at Baseline and Month 3
3.8. Treatment Length in Relation to Reduction in Drug Use
Within the GDC group (n = 18), there was a non-significant correlation (r = −0.07, p = 0.80) between the number of group counseling sessions attended and average ASI Drug Use composite score during months 1 to 3, controlling for baseline ASI Drug Use composite scores. Within the IDC + GDC group, however, there were significant correlations between the number of group sessions attended (r = −0.50, p = 0.04) and average ASI Drug Use composite during months 1 to 3, and the number of individual sessions attended (r = −0.59, p = 0.01) and average ASI Drug use composite during months 1 to 3 (in both cases, controlling for baseline ASI Drug Use composite scores).
3.9. Comparison to the NIDA CCTS
Our current patient sample was more severe than the sample in the NIDA CCTS. Duration of cocaine use was longer (about 11 years, compared to 7 years in the NIDA CCTS; t (43) = 3.83, p = 0.0004; DF based on Satterthwaite’s correction for heterogeneity of variances); days of cocaine use in the past 30 days at screening was higher (about 12.7 vs. 10.4; t (526) = 1.77, p = 0.08). However, ASI Psychiatric Severity Composite scores (0.21 vs. 0.19; t (526) = 0.70, p = 0.47.); HAM-D scores (10.0 vs. 8.9; t (44) = 0.87, p = 0.39), and ASI Drug Use Composite scores (about 0.24; t (526) = −0.09, p = 0.93.) were not significantly different in the two samples. The current sample was older (42 years vs. 34 years; t (526) =8.18, p < 0.0001), and largely minority (African American), with about 10% of the sample white (vs. 58% in the previous study; χ2 (1) = 35.31, p < 0.0001). Fewer patients in the current study were employed (44% vs. 60%; χ2 (1) = 4.20, p = 0.04).
Overall attrition rates in this study were higher than the NIDA CCTS. In the NIDA CCTS, 36% in IDC+GDC and 32% in GDC dropped out in the first month, compared to half of the patients dropping out in the first month in the current study (χ2 (1) =2.75, p = 0.10 for comparing the IDC + GDC groups in the two studies; χ2 (1) = 7.09, p = 0.008 for comparing the GDC alone groups in the two studies).
We also compared ratings of adherence and competence (average of 34 items that were in common between the measures used in the two studies) for the 14 patients in the current study with 98 patients in the NIDA CCTS. The average adherence in the current study (M = 3.8, SD = 0.36) was actually significantly (p < .0001) higher than that in the NIDA CCTS (M = 2.4, SD = 0.39). Average competence ratings were similar in the two studies (3.9 vs. 4.0, respectively).
ASI Drug Use outcomes in the current study were not significantly different from those obtained in the NIDA CCTS. In the NIDA CCTS, the IDC + GDC group decreased from 0.25 to 0.10 at 3 months on the ASI Drug Use composite; the GDC group decreased from 0.24 to 0.13 (F [2, 97] =0.71, p = 0.40 for IDC contrast and F [2, 102]=0.93, p = 0.33 for GDC contrast). However, it is important to note that in the NIDA CCTS treatment extended to 6 months and patients continued to improve. The end of treatment ASI Drug Use composite mean score for the IDC + GDC group at 6 months was 0.11, and was 0.13 for the GDC group in the NIDA CCTS. These 6 month scores are 20% and 16%, respectively, lower than the 3 month means for IDC + GDC and GDC alone in the current study, although the differences between the 6- month outcomes of the NIDA CCTS and the 3-month outcomes of the current study were not statistically significant for either the comparison of the IDC + GDC groups in the two studies (F [1, 104] = 1.21, p = 0.27) or the GDC alone groups in the two studies (F [1, 108] = 1.65, p = 0.20).
Examination of abstinent rates, however, revealed lower efficacy of IDC + GDC and GDC alone in the current study compared to IDC + GDC and GDC alone in the NIDA CCTS. In the NIDA CCTS, 71% of patients who received IDC+GDC compared to 15% for IDC + GDC in the current study (χ2 (1) = 23.2, p < 0.0001), and 58% of those who received GDC alone compared to 38% for GDC alone in the current study (χ2 (1) = 2.79, p = 0.10) achieved 1 month of abstinence on the composite cocaine abstinence measure. Rates of achieving 3 consecutive months of abstinence for IDC + GDC (38%) and GDC alone (27%) in the NIDA CCTS were significant higher than in the current study (χ2 (1) = 11.29, p < 0.0001 and χ2 (1) = 4.84, p = 0.03, respectively).