Key points are not available for this paper at this time.
BACKGROUND: Cardiovascular disease (CVD) is a leading global cause of mortality, strongly influenced by diet. Research in this field has evolved from the analysis of nutrients to the evaluation of dietary patterns (DP), with the aim of more accurately reflecting real eating behaviors. A variety of statistical methods are employed to identify DP, leading to heterogeneous results. This variability highlights the need for methodological standardization to improve comparability and evidence synthesis. OBJECTIVE: The aim of this study was to evaluate the statistical methodologies used to identify a posteriori DP and determine their relationship with CVD. METHODS & PROCEDURES: A systematic search of the PubMed, Scopus and Web of Science databases was carried out, including observational studies conducted on adults aged 18 years and over. The review protocol was registered in PROSPERO (CRD4202525642904). The analysis of 18 studies that identified a posteriori DP related to CVD was conducted. RESULTS: Across the included studies, DP were mainly derived using principal component analyses (PCA), although variability existed in component selection and factor loading thresholds. Most studies assessed sampling adequacy, applied Varimax orthogonal rotation, and selected components based on eigenvalue thresholds >1, although cut-offs varied. A small number of studies combined PCA with other multivariate approaches. Despite heterogeneity in identified patterns, Western, traditional, animal-based, and Mediterranean DP were most frequently reported. Western patterns were associated with increased CVD, whereas Mediterranean patterns showed a protective effect. CONCLUSIONS: CVD-related DP are highly variable, depending on factors such as sex, culture and socio-economic status. This underlines the importance of standardizing dietary collection and assessment methods.
Popescu et al. (Thu,) studied this question.