Table 8.
Evidentiary basis for strong recommendation favoring varenicline over nicotine patch in patients with mental disorders, with high-certainty evidence
| Certainty assessment |
Summary of findings |
Certainty | Importance | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No. of samples (studies) | Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | No (%) or % |
Relative effect (95% CI) | Absolute (per 1,000 patients) |
|||||
| Varenicline | Nicotine patch | Varenicline | Nicotine patch | ||||||||||
| Tobacco abstinence at 6 mo | 6,132 (6 RCTs) | Not serious | Not serious | Not serious | Not serious | Not serious | 1,104/6,131 (18) | 12.9 | RR 1.38 (1.27 to 1.50) | 49 more ↑35–↑65) | 129 | High | Critical |
| Serious adverse events | 6,531 (5 RCTs) | Not serious | Not serious* | Not serious | Not serious† | Not serious | 1,174/6,531 (18) | 12.1 | RR 1.07 (1.00 to 1.15) | 8 more (0–↑18) | 121 | High | Critical |
RR, risk ratio; CI, confidence interval; RCT, randomized controlled trials.
*
Heterogeneity (I2)=88%, but it was evaluated that there is no need to lower the level of evidence due to the same direction of the effects.
†
The CI contains 1.00, but between 0.75 and 1.25, so the evidence level was not lowered due to imprecision.