Aims and design: The finding that smokers who used more pieces of nicotine gum achieved better treatment outcomes has been interpreted to indicate that the use of more pieces of gum leads to better treatment outcomes. However, these correlational findings are subject to at least three alternate explanations: (1) reverse causation; (2) a confounding third variable; or (3) behavioral processes. We tested these alternative explanations in secondary analyses of data from a clinical trial of nicotine lozenges.
Participants and methods: Subjects (n = 1030) who quit smoking for at least 2 weeks in a placebo-controlled trial of nicotine lozenges were studied. Lozenge use was recorded daily; outcome was assessed as carbon monoxide (CO)-verified 28-day continuous abstinence at 6 weeks.
Findings: To refute the reverse causation hypothesis we analyzed data on compliance during a period when subjects were abstinent: high use of active lozenge was associated with greater success; for each additional lozenge used, the odds of success increased by 10%. The third variable and behavioral processes hypotheses both predict that high lozenge use will be associated with improved outcomes even in the placebo condition. However, our analyses showed that this was not the case. Further, greater use of lozenges increased the active-placebo difference, as would be expected under a pharmacological explanation of compliance effects.
Conclusion: The analyses rebut the alternative explanations, and suggest that use of more nicotine lozenges is causally associated with better quit rates.