Table 2.
Participants’ percentage of abstinence during maintenance dose period (Missing = Positive)
| Participants’ percentage of abstinence | Maintenance dose groups | BUP-XR 300/300 mg | BUP-XR 300/100 mg | Difference (95% CI)a 300–100 mg | P valueb |
|---|---|---|---|---|---|
| Opioid-injecting participants | N | 63 | 66 | ||
| Weeks 10–25 | |||||
| Unadjusted | Mean (SE) | 58.2% (4.62%) | 44.0% (4.75%) | 14.20% (1.22, 27.18%) | 0.0320 |
| Risk adjusted | Mean (SE) | 59.6% (5.19%) | 46.6% (5.34%) | 13.04% (− 1.56, 27.64%) | 0.0800 |
| Weeks 14–25 | |||||
| Unadjusted | Mean (SE) | 59.5% (4.79%) | 42.6% (4.78%) | 16.97% (3.71, 30.24%) | 0.0122 |
| Risk adjusted | Mean (SE) | 61.5% (5.35%) | 45.0% (5.40%) | 16.48% (1.58, 31.37%) | 0.0074 |
| Weeks 18–25 | |||||
| Unadjusted | Mean (SE) | 60.7% (4.85%) | 42.0% (4.83%) | 18.67% (5.25, 32.09%) | 0.0064 |
| Risk adjusted | Mean (SE) | 62.9% (5.26%) | 44.2% (5.39%) | 18.65% (3.90, 33.41%) | 0.0132 |
| Non-injecting opioid participants | N | 92 | 90 | ||
| Weeks 10–25 | |||||
| Unadjusted | Mean (SE) | 46.6% (4.22%) | 56.3% (3.99%) | − 9.65% (− 21.03, 1.73) | 0.0966 |
| Risk adjusted | Mean (SE) | 44.2% (4.56%) | 54.3% (4.19%) | − 10.06% (− 22.20, 2.08) | 0.1044 |
| Weeks 14–25 | |||||
| Unadjusted | Mean (SE) | 46.0% (4.41%) | 54.9% (4.19%) | − 8.89% (− 20.81, 3.02) | 0.1435 |
| Risk adjusted | Mean (SE) | 43.3% (4.78%) | 53.0% (4.33%) | − 9.66% (− 22.31, 2.99) | 0.1343 |
| Weeks 18–25 | |||||
| Unadjusted | Mean (SE) | 45.2% (4.55%) | 53.8% (4.28%) | − 8.51% (− 20.75, 3.74) | 0.1735 |
| Risk adjusted | Mean (SE) | 42.4% (4.91%) | 51.8% (4.40%) | − 9.41% (− 22.34, 3.52) | 0.1536 |
N No. of participants in the analysis set
Each participant’s percentage abstinence was calculated after the first (weeks 10–25), second (weeks 14–25), and third (weeks 18–25) maintenance doses as the proportion (%) of negative opioid use results among the corresponding 16, 12, and 8 weekly assessments, respectively. Missing = Positive: the missing UDS or missing self-report for illicit opioid use at a specific scheduled visit was imputed as positive for that visit. The risk-adjusted analysis compared the 100-mg versus 300-mg maintenance dose groups via inverse propensity weighting using propensity scores
CI confidence interval, SE standard error, UDS urine drug screen
aGroup mean (SE) and 95% CI of difference in group mean was based on the general linear model Robust Sandwich variance estimator
bP value was based on the Wald test