A systematic review to critically appraise methodological rigour in research on ultra-processed food and cardiovascular disease and hypertension

Abstract

Despite growing research linking ultra-processed food (UPF) consumption to risk of cardiovascular disease (CVD) and hypertension, no study has systematically evaluated the methodological rigor underlying these associations. We systematically searched major databases to identify eligible studies. Data were extracted for dietary assessment methods, UPF classification, covariate selection, confounding control, statistical modelling and effect estimates. Random-effects meta-analysis was conducted to pool effect estimates. Meta-regression and sensitivity analyses were performed to explore sources of heterogeneity. Substantial heterogeneity was observed in the application of the NOVA classification for categorising UPFs across the 46 eligible studies. Only two studies employed a directed acyclic graph to inform confounder selection; 43 used models with suboptimal adjustment, and 42 were overfitted due to adjustment for potential mediators. Pooled analyses indicated that higher consumption of UPFs was associated with a 9% higher risk of CVD and a 16% higher risk of hypertension, with stronger associations observed for coronary heart and cerebrovascular diseases. While higher UPF intake is consistently associated with increased risks of CVD and hypertension, methodological limitations may attenuate the observed associations. Strengthening methodological rigour through harmonised UPF classification and causal frameworks is essential to better elucidate the effect of UPF consumption on cardiometabolic health.