Table 2.
Network meta-analysis (NMA) and quality of evidence for abstinence
| Intervention (reference: placebo) | No of arms (placebo, n=42) | No of participants (placebo, n=4044) | Odd ratios (95% CI) | Corresponding absolute probability of abstinence (95% CI) (assumed 25% for placebo) (%) | Quality of evidence* | ||
|---|---|---|---|---|---|---|---|
| Direct estimate | Indirect estimate | NMA estimate | |||||
| Psychosocial interventions | |||||||
| Treatment as usual | 9 | 800 | - | 0.52 (0.29 to 0.94) | 0.52 (0.29 to 0.94) | 15 (9 to 24) | Low†‡§¶ |
| A-CHESS | 1 | 170 | - | 0.88 (0.35 to 2.21) | 0.88 (0.35 to 2.21) | 23 (10 to 42) | Very low†‡§¶ |
| CBT | 2 | 306 | - | 0.53 (0.23 to 1.22) | 0.53 (0.23 to 1.22) | 15 (7 to 29) | Low†§¶ |
| Short form CBT | 1 | 43 | - | 0.05 (0.00 to 1.16) | 0.05 (0.00 to 1.16) | 2 (0 to 28) | Very low†§¶ |
| Contingency management | 1 | 79 | - | 0.78 (0.17 to 3.61) | 0.78 (0.17 to 3.61) | 21 (5 to 55) | Low†‡§ |
| Coping skill training | 1 | 40 | - | 0.35 (0.10 to 1.19) | 0.35 (0.10 to 1.19) | 10 (3 to 28) | Very low†§¶ |
| Home visit | 2 | 142 | - | 0.95 (0.32 to 2.85) | 0.95 (0.32 to 2.85) | 24 (10 to 49) | Low†§¶ |
| MET | 2 | 308 | - | 0.45 (0.19 to 1.11) | 0.45 (0.19 to 1.11) | 13 (6 to 27) | Very low†§¶ |
| Drug interventions | |||||||
| Acamprosate | 18 | 2297 | 1.92 (1.52 to 2.42) | 0.74 (0.21 to 2.53) | 1.86 (1.49 to 2.33) | 38 (33 to 44) | Moderate¶** |
| Amisulpride | 1 | 37 | 0.39 (0.09 to 1.64) | - | 0.39 (0.09 to 1.64) | 12 (3 to 35) | Low†§¶ |
| Aripiprazole | 1 | 29 | - | 1.49 (0.43 to 5.18) | 1.49 (0.43 to 5.18) | 33 (12 to 63) | Low†§¶ |
| Atenolol | 1 | 50 | 0.85 (0.25 to 2.95) | - | 0.85 (0.25 to 2.95) | 22 (8 to 50) | Very low†‡§¶ |
| Baclofen | 1 | 28 | 4.63 (1.00 to 21.48) | - | 4.63 (1.00 to 21.48) | 61 (25 to 88) | Low†§¶ |
| Carbamazepine | 1 | 13 | 0.55 (0.08 to 3.90) | - | 0.55 (0.08 to 3.90) | 15 (2 to 57) | Very low†‡§¶ |
| Citalopram/escitalopram | 2 | 45 | - | 1.03 (0.33 to 3.16) | 1.03 (0.33 to 3.16) | 25 (10 to 51) | Low†§¶ |
| Disulfiram | 2 | 221 | 0.97 (0.46 to 2.01) | 0.72 (0.13 to 4.05) | 0.93 (0.48 to 1.79) | 24 (14 to 37) | Low†‡§¶ |
| Fluoxetine | 2 | 50 | 2.14 (0.48 to 9.52) | 4.51 (0.83 to 24.39) | 2.97 (0.97 to 9.05) | 50 (24 to 75) | Very low†‡§¶ |
| Flupenthixol | 1 | 142 | 0.44 (0.20 to 0.95) | - | 0.44 (0.20 to 0.95) | 13 (6 to 24) | Very low†§¶ |
| Fluvoxamine | 3 | 293 | 0.99 (0.49 to 2.01) | 1.14 (0.34 to 3.89) | 1.03 (0.57 to 1.88) | 26 (16 to 38) | Low†§¶ |
| Galantamine | 1 | 74 | 0.31 (0.11 to 0.87) | - | 0.31 (0.11 to 0.87) | 9 (4 to 23) | Low†§¶ |
| GHB | 4 | 201 | 1.65 (0.85 to 3.24) | 7.48 (2.05 to 27.28) | 2.31 (1.22 to 4.36) | 43 (29 to 59) | Very low†‡¶** |
| Levetiracetam | 1 | 95 | 1.03 (0.46 to 2.34) | - | 1.03 (0.46 to 2.34) | 26 (13 to 44) | Low†§¶ |
| Lisuride | 1 | 57 | 0.38 (0.13 to 1.12) | - | 0.38 (0.13 to 1.12) | 11 (4 to 27) | Very low†§¶ |
| Lithium | 1 | 28 | 1.43 (0.39 to 5.23) | - | 1.43 (0.39 to 5.23) | 32 (12 to 64) | Low†§¶ |
| Modafinil | 1 | 41 | 2.48 (0.72 to 8.53) | - | 2.48 (0.72 to 8.53) | 45 (19 to 74) | Low†§¶ |
| Naltrexone | 17 | 878 | 1.29 (0.86 to 1.92) | 1.59 (0.81 to 3.10) | 1.36 (0.97 to 1.91) | 31 (24 to 39) | Low†§¶ |
| Nefazodone | 1 | 50 | 0.57 (0.19 to 1.76) | - | 0.57 (0.19 to 1.76) | 16 (6 to 37) | Very low†‡§¶ |
| Oxcarbazepine | 2 | 72 | - | 2.46 (0.91 to 6.61) | 2.46 (0.91 to 6.61) | 45 (23 to 69) | Very low†§¶ |
| Pregabalin | 1 | 31 | - | 1.97 (0.58 to 6.74) | 1.97 (0.58 to 6.74) | 40 (16 to 69) | Low†§¶ |
| Quetiapine | 1 | 29 | 6.75 (1.20 to 38.05) | - | 6.75 (1.20 to 38.05) | 69 (29 to 93) | Low†§¶ |
| Tianeptine | 1 | 170 | 1.22 (0.58 to 2.57) | - | 1.22 (0.58 to 2.57) | 29 (16 to 46) | Low†§¶ |
| Tiapride | 2 | 187 | 0.56 (0.30 to 1.05) | - | 0.56 (0.30 to 1.05) | 16 (9 to 26) | Moderate§¶ |
| Topiramate | 3 | 194 | 2.26 (0.83 to 6.13) | 1.72 (0.84 to 3.52) | 1.88 (1.06 to 3.34) | 39 (26 to 53) | Very low†‡§¶** |
| Trazodone | 1 | 88 | 0.61 (0.20 to 1.84) | - | 0.61 (0.20 to 1.84) | 17 (6 to 38) | Very low†‡§¶ |
| Combined interventions | |||||||
| Placebo+CBT | 1 | 50 | 0.83 (0.28 to 2.42) | - | 0.83 (0.28 to 2.42) | 22 (9 to 45) | Very low†‡§¶ |
| Nefazodone+CBT | 1 | 53 | 0.77 (0.26 to 2.23) | - | 0.77 (0.26 to 2.23) | 20 (8 to 43) | Very low†‡§¶ |
| Acamprosate+nurse visit | 1 | 50 | - | 4.59 (1.47 to 14.36) | 4.59 (1.47 to 14.36) | 60 (33 to 83) | Very low a c d |
| Acamprosate+NTX | 1 | 40 | 5.57 (1.82 to 16.96) | 1.63 (0.33 to 7.95) | 3.68 (1.50 to 9.02) | 55 (33 to 75) | Low†§¶** |
| GHB+EST | 1 | 12 | - | 5.13 (0.53 to 49.92) | 5.13 (0.53 to 49.92) | 63 (15 to 94) | Low†§¶ |
| GHB+NTX | 1 | 18 | - | 12.64 (2.77 to 57.78) | 12.64 (2.77 to 57.78) | 81 (48 to 95) | Very low†‡§¶ |
| NTX+EST | 1 | 12 | - | 2.57 (0.25 to 25.85) | 2.57 (0.25 to 25.85) | 46 (8 to 90) | Low†§¶ |
| NTX+GHB+EST | 1 | 12 | - | 25.65 (2.13 to 309.46) | 25.65 (2.13 to 309.46) | 90 (41 to 99) | Low†§¶ |
A-CHESS=Addiction‐Comprehensive Health Enhancement Support System; CBT=cognitive behavioural therapy; MET=motivational enhancement therapy; GHB=sodium salt of gamma hydroxybutyric acid (sodium oxybate); EST=escitalopram; NTX=naltrexone.
*
See supplement 4 for full details of criteria for downgrading quality of evidence.
†
Within study bias.
‡
Indirectness.
§
Imprecision.
¶
Heterogeneity.
**
Incoherence.