Summary of findings 3. Reduction and pharmacotherapy versus reduction alone for smoking cessation.
| Reduction and pharmacotherapy compared to reduction alone for smoking cessation | ||||||
| Patient or population: people who smoke Setting: community; primary care (Australia; Canada; Czech Republic; Denmark; Egypt, Germany; Japan, Mexico, New Zealand; Switzerland; Taiwan; UK; USA) Intervention: reduction to quit aided by pharmacotherapy Comparison: reduction to quit alone (placebo or no pharmacotherapy) | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect (95% CI) | № of participants (studies) | Certainty of the evidence (GRADE) | Comments | |
| Risk with placebo/no pharmacotherapy | Risk with pharmacotherapy | |||||
| Main analysis (all pharmacotherapy): smoking cessation (≥ 6 month follow‐up) | Study population | RR 1.68 (1.09 to 2.58) | 8636 (11 RCTs) | ⊕⊕⊝⊝ LOWa,b,c | The rows below are subgroups of this main analysis. | |
| 5 per 100 | 8 per 100 (5 to 13) | |||||
| Subgroup: combination NRT; smoking cessation (≥ 6 month follow‐up) | Study population | RR 1.02 (0.61 to 1.69) | 1124 (3 RCTs) | ⊕⊕⊝⊝ LOWd,e,f | ‐ | |
| 15 per 100 | 15 per 100 (9 to 25) | |||||
| Subgroup: nicotine patch; smoking cessation (≥ 6 month follow‐up) | Study population | RR 0.34 (0.02 to 5.31) | 85 (1 RCT) | ⊕⊝⊝⊝ VERY LOWg,h | ‐ | |
| 15 per 100 | 5 per 100 (0 to 80) | |||||
| Subgroup: fast‐acting NRT only; smoking cessation (≥ 6 month follow‐up) | Study population | RR 2.56 (1.93 to 3.39) | 5323 (7 RCTs) | ⊕⊕⊕⊝ MODERATEi | ‐ | |
| 2 per 100 | 6 per 100 (5 to 8) | |||||
| Subgroup: varenicline only; smoking cessation (≥ 6 month follow‐up) | Study population | RR 3.99 (2.93 to 5.44) | 1510 (1 RCT) | ⊕⊕⊕⊝ MODERATEi | ‐ | |
| 6 per 100 | 24 per 100 (18 to 33) | |||||
| Subgroup: bupropion only; smoking cessation (≥ 6 month follow‐up) | Study population | RR 1.27 (0.67 to 2.40) | 594 (1 RCT) | ⊕⊕⊝⊝ LOW f,,j | ‐ | |
| 5 per 100 | 7 per 100 (4 to 13) | |||||
| *The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; RCT: randomised controlled trial; RR: Risk ratio | ||||||
| GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect | ||||||
aWe did not downgrade due to risk of bias: A sensitivity analysis removing studies judged to be at high risk of bias did not change our interpretation of the effect. bDowngraded by one level due to inconsistency: substantial heterogeneity was detected (I2 = 78%). A subgroup analysis grouping by type of pharmacotherapy used explained a small amount of this, but moderate heterogeneity remained unexplained. cDowngraded by one level due to imprecision: the CIs of the absolute effect indicate potentially no benefit of pharmacotherapy, whereas the upper limit indicates the potential for a clinical benefit. dDowngraded by one level due to risk of bias: we rated two of the three studies at high risk of bias, due to the use of an unmatched placebo. However, removal of these studies did not change the interpretation of the effect. eWe did not downgrade, due to inconsistency despite an I2 of 44%, as there was a high degree of CI overlap. fDowngraded by one level due to imprecision: the event rate is low and the CIs of the effect estimate incorporate clinically relevant potential benefit and harm of the intervention. gDowngraded by one level due to risk of bias: the only study in this comparison was at high risk of bias. hDowngraded by two levels due to imprecision: the event rate is very low (n = 6) and the CI of the effect estimate indicates considerable harm as well as benefit. iDowngraded by one level due to imprecision: the overall number of events was low (< 300). jDowngraded by one level due to risk of bias: we rated the only included study at high risk of attrition bias, due to high rates of dropout.