The dietary patterns derived by reduced-rank regression in association with Framingham risk score and lower DASH score in Hoveyzeh cohort study

Abstract

AbstractThe relationship between dietary patterns (DPs) and cardiovascular disease (CVD) has been the subject of much research, but given the significance of this disease, studying the factors affecting it through different methodological considerations is of utmost importance. This study aimed to investigate the association between the four dietary patterns (DPs) derived from reduced-rank regression (RRR) and the risk of CVD predicted by the Framingham Risk Score (FRS) in the Arab residence of Khuzestan, Iran. Furthermore, the predefined Dietary Approaches to Stop Hypertension (DASH) would be used as a comparative model to assess the validity of the extracted DPs. In this cross-sectional study, 5799 individuals aged 35–70 without a CVD diagnosis were selected among the participants of the Hoveyzeh cohort study (HCS). The Risk of CVD was assessed using the FRS model. A semi-quantitative food frequency questionnaire evaluated dietary intake. Four DPs were derived using RRR with 28 food groups as predictors and total protein (g/d), fiber(g/d), fat(g/d), and magnesium intake (mg/d) as response variables. Multinomial and binary logistic regression were used to assess the relationship of DPs with intermediate (10–20%) and high (> 20%) levels of FRS and lower DASH scores (< 4.5), respectively. Four primary DPs were derived, which explained 89.10 of the total explained variance in participants’ dietary intake. Multinomial regression was applied between FRS (10–20%) and (> 20%) across quartiles of four identified DPs. After adjustment for potential confounders, higher tendency to 1st and 2nd DPs in Model 1, OR = 4.67 (95% CI 3.65; 6.01), OR = 1.42 (95% CI 1.13; 1.79) were presented accordingly. The 1st DP, characterized by higher intake of refined grains and lower intake of vegetables oil, sugar, mayonnaise and artificial juices, the 2nd DP characterized by higher intake of hydrogenated fat and lower consumption of tomato sauce and soft drink was associated with greater odds of CVD with the intermediate level of FRS. However, higher adherence to the 3rd DP, characterized by higher intake of fruits, vegetables and legumes and lower intake of fish, egg, red meat, processed meat, mayonnaise, sugar and artificial juices, the 4th DP characterized by higher intake of coffee, nuts and lower intake of sugar, mayonnaise and artificial juices was associated with a lower risk of FRS. Moreover, lower DASH score considered in binary logistic regression across quartiles of four identified dietary patterns. 1st and 2nd DPs were directly related to lower DASH scores, while 3rd and 4th DPs had high comparability with the DASH diet and inversely contributed to the lower DASH score. Total DASH score was significantly correlated to four derived DPs. Our findings confirm the current knowledge regarding the beneficial effects of healthy plant-based DPs and the avoidance of high-fat and processed foods to prevent CVD.