Table 4.
Analysis of secondary outcomes (full-analysis set).
| Outcome | SoC (n = 156) | BUP-XR (n = 158) | Adjusted estimate (95% CI) | p-value |
|---|---|---|---|---|
| Clinical assessments | ||||
| Craving for opioidsc | ||||
| Zero score on CEQ-F—probability | 0.54 (0.04) | 0.92 (0.03) | OR 3.22 (1.65–6.36) | 0.001 |
| Non-zero score on CEQ-F—mean | 20.87 (2.66) | 5.80 (1.52) | IRR 0.52 (0.34–0.81) | 0.004 |
| Zero score on VAS-N—probability | 0.67 (0.04) | 0.92 (0.02) | OR 6.17 (2.61–14.45) | 0.001 |
| Non-zero score on VAS-N—mean | 17.24 (2.27) | 3.75 (1.21) | IRR 0.83 (0.59–1.15) | 0.248 |
| Zero score on VAS-W—probability | 0.60 (0.05) | 0.79 (0.04) | OR 3.00 (1.20–7.39) | 0.018 |
| Non-zero score on VAS-W—mean | 21.81 (3.43) | 8.08 (2.24) | IRR 0.70 (0.49–1.00) | 0.052 |
| Craving for cocainec | ||||
| Zero score on CEQ-F—probability | 0.52 (0.05) | 0.51 (0.04) | OR 0.94 (0.41–2.16) | 0.885 |
| Non-zero score on CEQ-F—mean | 21.90 (3.29) | 15.84 (2.53) | IRR 0.71 (0.46–1.10) | 0.124 |
| Zero score on VAS-N—probability | 0.74 (0.05) | 0.81 (0.04) | OR 1.59 (0.53–4.81) | 0.400 |
| Non-zero score on VAS-N—mean | 13.26 (2.37) | 9.57 (2.30) | IRR 0.97 (0.77–1.23) | 0.814 |
| Zero score on VAS-W—probability | 0.58 (0.05) | 0.55 (0.04) | OR 0.85 (0.38–1.97) | 0.707 |
| Non-zero score on VAS-W—mean | 24.11 (2.83) | 21.79 (2.28) | IRR 0.85 (0.77–1.02) | 0.077 |
| Opioid use | ||||
| Maximum days continuously abstinent—mean | 77.44 (12.34) | 95.08 (8.50) | IRR 1.23 (1.06–1.42) | 0.005 |
| Cocaine use | ||||
| Days abstinent—mean | 102.89 (11.90) | 112.16 (5.91) | IRR 1.09 (0.95–1.25) | 0.230 |
| Maximum days continuously abstinent—mean | 70.45 (12.33) | 71.34 (9.21) | IRR 1.01 (0.86–1.19) | 0.877 |
| Benzodiazepine use | ||||
| Days abstinent—mean | 115.06 (10.39) | 121.16 (5.64) | IRR 1.05 (0.95–1.16) | 0.312 |
| Maximum days continuously abstinent—mean | 92.65 (12.04) | 104.04 (7.85) | IRR 1.12 (0.10–0.94) | 0.198 |
| Alcohol use | ||||
| Drinking days—meanb | 1.36 (0.25) | 1.25 (0.24) | IRR 0.88 (0.51–1.52) | 0.640 |
| Standard units of alcohol—log transformed meana | 1.79 (0.14) | 1.89 (0.12) | Difference 0.09 (0.22–0.40) | 0.545 |
| Maximum units on drinking day—log transformed meana | 2.01 (0.21) | 2.11 (0.15) | Difference 0.11 (−0.24 to 0.45) | 0.545 |
| Outcome | SoC (n = 156) | BUP-XR (n = 158) | Adjusted estimate (95% CI) | p-value |
|---|---|---|---|---|
| Clinical assessments—continued | ||||
| DERS-SF—meanf | 45.13 (1.68) | 42.52 (0.96) | Difference −2.61 (−6.12 to 0.91) | 0.145 |
| QUIDS-SR—meanf | 7.49 (0.56) | 6.75 (0.50) | Difference −0.74 (−1.86 to 0.38) | 0.194 |
| WSAS—meang | −4.45 (1.38) | −8.25 (1.02) | Difference −3.80 (−6.76 to −0.84) | 0.012 |
| Clinician-reported | ||||
| ADAPT | ||||
| OUD severity—meang | −0.49 (0.36) | −0.77 (0.30) | Difference −0.28 (−0.70 to 0.136) | 0.185 |
| Concurrent problem complexity—meang | −0.64 (0.47) | −1.09 (0.37) | Difference −0.45 (−1.34 to 0.44) | 0.322 |
| Recovery strengths—meang | −0.27 (0.32) | 0.29 (0.27) | Difference 0.55 (0.07–1.05) | 0.025 |
| CGI-I—proportion not changed or worsenedd | 0.46 (0.01) | 0.10 (0.001) | OR 0.12 (0.05–0.27) | 0.001 |
| Patient-reported | ||||
| PRO-I—proportion not changed or worsenedd | 0.32 (0.01) | 0.10 (0.002) | OR 0.21 (0.09–0.47) | 0.001 |
| SURE—meane | 51.30 (1.40) | 56.03 (1.13) | Difference 6.28 (3.58–8.98) | 0.001 |
Data in parentheses are standard error, percentage or 95% confidence interval; all means are adjusted.
ADAPT, Addiction Dimensions for Assessment and Personalised Treatment; ALC-QFM, alcohol consumption scale: frequency, quantity, and maximum quantity of alcohol consumed on any one day; BUP-XR, extended-release injectable buprenorphine; CEQ-F, Craving Experiences Questionnaire-frequency version; CGI-I, Clinical Global rated evaluation of improvement (minimally improved–very much improved vs. no change–very much worse); DERS-SF, Difficulties in Emotion Regulation Scale-Short Form; HR, hazard ratio; IRR, incidence rate ratio; VAS-N/W, visual analogue scale of perceived need and want for drug; OR, odds ratio; PRO-I, Patient Reporting Outcome—Improvement (minimally improved–very much improved vs. no change–very much worse); less than moderate versus moderate and above); SoC, standard-of-care; QIDS-SR, Quick Inventory of Depressive Symptomatology-Self-Report; WSAS, Work and Social Adjustment Scale.
Mixed-effects negative binomial regression, with stratification factors and baseline score (fixed effects) and clinic (random intercept) with log transformation of average and maximum number of units.
Mixed-effects negative binomial regression, with stratification factor and baseline score (fixed effects) and clinic (random intercept).
Mixed-effects zero-inflated Poisson regression, with stratification factor and baseline score and treatment clinic (fixed effects).
Mixed-effects logistic regression, with stratification factor and baseline score (fixed effect) and treatment clinic (random intercept).
Mixed-effects tobit regression, with stratification factor (fixed effect) and treatment clinic (random intercept).
Mixed effects general linear model, with stratification factor, baseline score (fixed effects) and treatment clinic (random intercept).
Mixed-effects general linear model on change from baseline, with stratification factor (fixed effect) and treatment clinic (random intercept).