A systematic review and meta-analysis of the relationship between heavy smoking and probability discounting.

Abstract

Probability discounting (PD), which refers to the process of adjusting the value of future probabilities when making decisions, is a method of measuring impulsive decision-making; however, the relationship between PD and nicotine remains unclear. The current study aimed at investigating the significance of PD in individuals who smoke. According to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, we searched the PubMed, Embase, PsycINFO, and Web of Science databases for articles comparing individuals who smoke and their tobacco-naïve controls using PD task as outcome measure from inception to May 2023. The main outcome was an overall difference in PD function, while subgroup analysis and meta-regression were conducted to examine the analysis methods and the moderators of PD. Fourteen studies in total involving 384 individuals who smoke and 493 controls (mean age = 24.32 years, range = 15.1-38.05 years) were analyzed. The effect of smoking on PD was significant (g = 0.51, p = .02). The discounting parameter from the equation, compared to the area under the curve, was more sensitive to estimating PD function (p = .01). Regression analysis showed positive correlations of PD with female percentage, age, and the number of probability options (all p < .04), but not with the number of choices at each probability and maximum reward magnitude (all p > .07). There was no significant publication bias across the eligible studies (p = .09). Our findings, which are the first to demonstrate a smaller PD (i.e., prone to risk-taking) in individuals who smoke, shed light on the appropriate analysis method, gender effect, age, and probability options on the PD function.