Effects of experimental negative affect manipulations on ad libitum smoking: a meta-analysis

To quantify the effect of negative affect (NA), when manipulated experimentally, upon smoking as measured within laboratory paradigms. Quantitative meta-analyses tested the effects of NA versus neutral conditions on (1) latency to smoke and (2) number of puffs taken.

Twelve experimental studies tested the influence of NA induction, relative to a neutral control condition (n = 1190; range = 24-235). Those providing relevant data contributed to separate random-effects meta-analyses to examine the effects of NA on two primary smoking measures: (1) latency to smoke (nine studies) and (2) number of puffs taken during ad libitum smoking (11 studies). Hedge's g was calculated for all studies through the use of post-NA cue responses relative to post-neutral cue responses. This effect size estimate is similar to Cohen's d, but corrects for small sample size bias.

NA reliably decreased latency to smoke (g = -0.14; CI = -0.23 to -0.04; P = 0.007) and increased number of puffs taken (g = 0.14; CI = 0.02 to 0.25; P = 0.02). There was considerable variability across studies for both outcomes (I(2)  = 51 and 65% for latency and consumption, respectively). Potential publication bias was indicated for both outcomes, and adjusted effect sizes were smaller and no longer statistically significant.

In experimental laboratory studies of smokers, negative affect appears to reduce latency to smoking and increase number of puffs taken, but this could be due to publication bias.