Differences in Magnitude of Cue Reactivity Across Durations of Smoking History: A Meta-analysis.

Abstract

Cue-elicited craving may vary due to duration of smoking history, increasing as more years of smoking strengthen associations between nicotine intake and cues. However, research on this relationship is virtually absent. This project assessed the relationship between cue reactivity and years of smoking. Data from 53 studies (68 effect sizes) were analyzed. Eligible studies were those measuring self-reported craving following cue exposure in nontreatment seeking smokers and reporting mean years smoking. Preliminary subgroup analyses identified methodological factors influencing cue-reactivity effect sizes; primary meta-regression analysis assessed differences across years smoking; exploratory analyses assessed potential for ceiling effects. Effect sizes varied due to abstinence requirement and cue presentation modality, but not dependence severity. Unexpectedly, meta-regression analysis revealed a decline in effect sizes across years smoking. Exploratory analyses suggested declines may have been due to a ceiling effect in craving measurement for those with longer smoking histories-more experienced smokers reported higher levels of craving at baseline or following neutral cue exposure, but all reported similar levels of craving after smoking cue exposure. Methodological factors and duration of smoking history influenced measurement of cue reactivity. Highlighted were important relationships between years smoking and magnitude of cue reactivity, depending on use of baseline or neutral cue comparisons. Further research is needed to assess differences in cue reactivity due to duration of smoking history using participant-level data, directly testing for ceiling effects. In addition, cue-reactivity studies are needed across young adults to assess onset of associations between nicotine intake and cues. This meta-analysis project contributes to the cue-reactivity literature by reporting on the previously ignored relationship between duration of smoking history and magnitude of cue-elicited craving. Results suggest that declines in cue-reactivity effect sizes across years of smoking may have been due to study-level methodological factors, but not due to differences in sample-level dependence severity. Cue-reactivity effect sizes were stable across years of smoking in studies using a neutral cue comparison but declined sharply in studies when baseline assessment (typically coupled with an abstinence requirement) was used.