Dietary sources of (poly)phenols and the risk of cardiovascular disease in the United Kingdom

Higher total fruit and vegetable intake (FV) is associated with a lower CVD risk, however, the relative importance of particular fruits or vegetables (subgroups) is unknown. The association between coffee, tea intakes, and CVD risk is also inconsistent. FV, coffee, and tea are (poly)phenol-rich foods commonly consumed in the UK. This thesis studied the association between these food groups, their respective subgroups and CVD mortality, incidence, and blood pressure (BP) in the UK Women’s Cohort Study (UKWCS) and the National Diet and Nutritional Survey Rolling Programme (NDNS RP). Survival analysis was applied to calculate the risk of CVD mortality, and CVD incidence with increasing FV, coffee and tea intake in the UKWCS. The odds of being hypertensive were assessed using logistic regression in the UKWCS (self-reported) and the NDNS RP (measured). Measured BP levels were also investigated in the NDNS RP using multivariate regression. Aspects of methodology were explored between dietary assessments in the UKWCS using the Kappa statistic. Overall, CVD risk was lower with higher intakes of FV, especially for total fruit, berries, citrus and grapes, but not for total vegetable. Greater intakes of total vegetables, fruit vegetables, and pomes were associated with a lower incidence of self-reported BP. Systolic BP was lower with increasing FV intake while diastolic BP lowered with a greater pome consumption. Attenuated associations in subpopulations with CVD risk factors suggest that protective effects are more apparent in healthy subpopulations. Coffee and tea intakes were not associated with the risk of CVD in either UK populations. The FFQ in the UKWCS was able to rank the participants by low and high intakes, providing evidence for significant and null associations detected above. Novel findings from the total phenolic content within fruit juices and concentrates from the UK highlights the importance of updating Phenol Explorer, to accurately estimate (poly)phenol intake in the future. Implementation of RCTs based on hypotheses generated here would aid the determination of causal links between FV subgroups such as berries, citrus, pomes, grapes, and fruit vegetables and CVD risk.