Anti-inflammatory diet and mental health outcomes in an aging population: evidence from a preventive population-based target trial emulation

Population-based preventive trials are rare in psychiatry, especially in an aging population, and the effects of potentially modifiable lifestyle risk factors such as diet on mental health outcomes are largely unexplored. This study aimed to estimate the effects of an anti-inflammatory diet on mental health outcomes in older adults and the potential effect of lifestyle modification. A preventive population-based target trial was emulated using data from the ASPirin in Reducing Events in the Elderly (ASPREE) study among community-dwelling older adults (≥ 70 years) in Australia. An anti-inflammatory diet was considered as an intervention, with a pro-inflammatory diet as a control. Additional analyses explored an anti-inflammatory lifestyle as an effect modifier and the combination of diet and lifestyle. Depressive symptoms and mental health quality of life were the study outcomes. A target trial protocol was prospectively designed, and a range of potential confounders was balanced using the inverse probability treatment weighting. A total of 10,955 participants (54.3% in the intervention group) were included in the target trial emulation. The risk of depressive symptoms was significantly lower in the intervention group compared to the control (RR, 0.92; 95% CI, 0.87-0.97). The between-group mean difference of mental component score in the intervention group was larger by 0.30 units compared to the control group (β, 0.30 (0.08-0.52)). When diet and lifestyle were combined, the anti-inflammatory group was associated with a lower risk of depressive symptoms and better mental health quality of life than the pro-inflammatory group (RR, 0.79 (0.70-0.89); β, 0.63 (0.23-1.03)), respectively. An anti-inflammatory diet intervention, especially when combined with an anti-inflammatory lifestyle, is associated with a reduction in depressive symptoms and better mental health quality of life.