Effects of episodic future thinking in health behaviors for weight loss: A systematic review and meta-analysis

Obesity and related diseases have become one of the leading causes of death worldwide, which has been linked to biopsychosocial effects such as type 2 diabetes, cardiovascular disease, various cancers, depression, and weight stigma. Episodic future thinking (EFT) has been found to support the development of changes in health behaviors. However, the effectiveness of EFT in enhancing weight loss behavior and health outcomes is not well supported. To establish implementation options for the EFT intervention, and critically synthesize the data that assesses the impact of EFT on weight loss behavior and health outcomes. Searches were performed across 5 Chinese and 9 English databases systematically from inception to March 2023. Randomized controlled trials, written in English or Chinese were included. Two independent reviewers evaluated all relevant studies, who also assessed the risk of bias, and extracted the data. Meta-analyses were conducted using Review Manager 5.4.1. The quantity of evidence's certainty was assessed using the Risk bias assessment tool RoB2 (revised version 2019). This study was registered in PROSPERO. A total of 1740 participants were included, and 18 studies were eligible for inclusion. Meta-analysis reported a statistically significant effect size favoring EFT on delay discounting (AUC) (MD = 0.1, 95 % CI: [0.02, 0.17], P = 0.01; I2 = 73 %), delay discounting (K) (MD = − 0.85, 95 % CI: [− 1.44, − 0.26], P = 0.005; I2 = 77 %), energy intake (MD = − 107.59, 95 % CI: [− 192.21, − 22.97], P = 0.01; I2 = 57 %), grocery purchased (SMD: − 0.91, 95 % CI:[− 1.48, − 0.34], P = 0.002; I2 = 63 %), and BMI (MD = − 2.73, 95 % CI: [− 5.13, − 0.32], P = 0.03; I2 = 0 %, two studies). EFT was found to have favorable effects on delay discounting, energy intake, grocery purchased, and BMI of individuals. The presence of high heterogeneity is evident in most of the outcomes. The modalities of EFT intervention are still in the exploratory phase, there is no consensus on the valence, context type, longest delay time, and practice strategy, and it needs to be further explored for different populations. It is anticipated that additional well-designed studies will continue developing high-quality evidence in this field.