Do implementation intentions reduce alcohol consumption? A meta-analysis

Abstract

Background: Asking individuals to form implementation intentions (i.e., if-then plans) has been shown to reduce subsequent alcohol consumption. The main objective of this study was to quantify the effect-size difference in alcohol consumption between individuals forming versus not forming implementation intentions. Methods: Systematic literature searches were conducted using the following databases: Cochrane Database of Systematic Reviews; EThOS; PsychArticles; PubMed; Web of Science. Nine studies that reported alcohol consumption for control and implementation intention groups were included. Findings: Random-effects meta-analysis was performed (N = 3102, K = 9) using effect-size differences (d) in alcohol consumption between control and implementation intention participants. Overall, there was a small, significant, positive effect-size difference in alcohol consumption between implementation intention and control participants (d+ = 0.21 [0.06; 0.36]; χ2(8) = 17.71, p = 0.02). To explore heterogeneity in results, length of follow-up (e.g., within one month; beyond one month) was examined as a moderator variable. There was a small, significant, positive effect-size difference in alcohol consumption when data was collected within one month (d+ = 0.43 [0.21; 0.66]; χ2(8) = 17.71, p = 0.02). However, the effect-size difference in alcohol consumption was non-significant when data was collected beyond one month (d+ = 0.07 [-0.02; 0.17]; χ2(8) = 17.71, p = 0.02). Discussion: Meta-analysis confirms the utility of implementation intentions as an intervention to reduce alcohol consumption: Individuals who formed implementation intentions reported lower alcohol consumption at follow-up than individuals who did not form implementation intentions. These effects were stronger for shorter follow-up time points.