Table 3:

Sensitivity analyses for the outcome of serious adverse cardiovascular events^*

Sensitivity analysis	Statistical model	No. of RCTs	Group; no. of events, n/N		OR (95% CI)
			Varenicline	Control
Placebo comparator
Reciprocal of the treatment arm size
Continuity correction	Fixed (MH)	149,16–28	52/4908	27/3308	1.67 (1.06–2.64)
No continuity correction	Fixed (MH)	149,16–28	52/4908	27/3308	1.77 (1.09–2.88)
Use of unadjudicated cardiovascular event data from one trial	Peto OR	149,16–28	61/4908	29/3308	1.91 (1.25–2.94)
Exclusion of most influential study	Peto OR	1316–28	27/4553	7/2949	2.54 (1.26–5.12)
Placebo or active†comparator	Peto OR	159,16–29	52/5286	30/4486	1.67 (1.07–2.62)

Note: CI = confidence interval, OR = odds ratio, MH = Mantel–Haenszel test, RCT = randomized controlled trial.

^*Statistical heterogeneity was I² = 0% for all sensitivity analyses.

^†Bupropion or nicotine replacement therapy.