Abstract/Summary

Current public dietary guidelines recommend limiting the intake of saturated fatty acids (SFA) and red and processed meat, while having optimal intake of low-fat dairy foods to meet the nutritional requirements. Prior observational evidence suggested beneficial cardiovascular diseases (CVD) associations with substitution of meat SFA with dairy SFA, but the associations of individual types of meat and dairy with CVD risk remain unclear, especially in the UK population for which evidence is lacking. Simultaneously, red meat and dairy foods are important sources of high-quality proteins in many Western diets, yet the impact of increasing protein intake from these foods and other protein sources on risk markers of CVD remains unclear. In this context, this PhD thesis aimed to (i) examine the associations of substituting total and different types of meat with total and different types of dairy and of one type of meat with another, equivalised to SFA, with the risk of CVD, and (ii) examine the impact of increasing protein intake from various foods within the diet on risk markers of CVD. Using data from 21,841 participants of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk study (56% female; mean age: 58.5 ±9.2 years), I modelled the isoenergetic substitution of SFA from total meat, processed meat, red meat (unprocessed and processed) and poultry with SFA from total dairy, milk, yogurt and cheese and assessed the associations with the incidence of CVD, CHD and stroke. Substituting 2.5% of total energy (en%) of SFA from total meat with total dairy was associated with 11% lower risk of CVD (Hazard ratio (HR): 0.89; 95% confidence intervals (CI): 0.82, 0.96) and 12% lower risk of CHD (0.88: 0.80, 0.96). Food group analyses showed that substitution of processed meat SFA with cheese SFA was associated with lower incidence of CVD (0.77: 0.68, 0.88); CHD (0.77: 0.66, 0.90), and stroke (0.81: 0.67, 0.99). Beneficial associations with CVD (0.84: 0.75, 0.94) and CHD (0.83: 0.73, 0.95) was also found with substitution of processed meat SFA with milk SFA. Substitution of red meat SFA with cheese SFA was inversely associated with the risk of CVD (0.86: 0.76, 0.97) and CHD (0.86: 0.74, 1.00). On the contrary, higher incidence of stroke was found with substitution of poultry SFA with milk SFA (2.06: 1.09, 3.89), yogurt SFA (2.55: 1.27, 5.13), and cheese SFA (1.96: 1.04, 3.70). In 120,496 participants of the UK Biobank (57% female, mean age: 55.9 ±7.8 years), I replicated the above study and extended the findings from EPIC-Norfolk by modelling substitution of SFA from more extensive categories of meat (total meat, processed meat, unprocessed meat, red meat, unprocessed red meat, processed red meat, poultry, unprocessed poultry and processed poultry) with SFA from total dairy, milk, yogurt and cheese in relation to incident CVD, CHD and stroke. Furthermore, I modelled the substitution of SFA from red meat with poultry and of SFA from processed meats with unprocessed meats in relation to CVD outcomes. The substitution of 2.5 en% of SFA from total meat with total dairy was associated with 9% (0.91: 0.86, 0.96) and 8% (0.92: 0.86, 0.98) lower incidence of CVD and CHD, respectively. Analyses of types of meat and dairy revealed lower incidence of CVD with substitution of SFA from processed meat, but not unprocessed meat, with the same quantities of SFA from total dairy (0.88: 0.81, 0.95), milk (0.91: 0.82, 1.00), yogurt (0.78: 0.67, 0.92) and cheese (0.86: 0.80, 0.94). Similarly, substituting SFA from processed red meat with total dairy (0.88: 0.82, 0.96), yogurt (0.80: 0.68, 0.94) and cheese (0.88: 0.81, 0.96) was associated with lower CVD incidence. Similar associations were observed for the substitution of SFA from unprocessed red meat with total dairy (0.91: 0.85, 0.98), yogurt (0.82: 0.70, 0.97) and cheese (0.91: 0.85, 0.98), although lower in magnitude. Moreover, substituting SFA from processed poultry, but not unprocessed poultry, with total dairy (0.64: 0.43, 0.94), milk (0.67: 0.45, 0.99), yogurt (0.58: 0.38, 0.87) and cheese (0.64: 0.43, 0.94) was strongly inversely associated with lower incidence of CVD. Within meat categories, substituting SFA from processed with unprocessed meat was associated with lower incidence of CVD (0.87: 0.76, 0.98). Specifically, substitution of SFA from processed poultry with unprocessed poultry was associated with lower CVD incidence (0.61: 0.40, 0.93). Associations were similar for CHD and stroke. The effect of increasing protein intake within a diet on measures of adiposity, blood pressure, blood lipids and markers of glucose metabolism was evaluated in a systematic review and meta-analysis of 54 RCTs (n participants=4344, 65% female, mean age: 46 ±10 years). Over a mean study duration of 18 weeks (range: 4 to 156 weeks), higher protein diets (range: 20–45 en%) showed small but beneficial effects on weight loss (standardised mean difference (SMD): -0.13, 95% CI: -0.23, − 0.03), fat mass loss (-0.14: - 0.24, − 0.04), systolic BP (-0.12: -0.21, − 0.02), total cholesterol (-0.11: -0.19, − 0.02), triacylglycerol (- 0.22: -0.30, − 0.14) and insulin (-0.12: -0.22, − 0.03), compared to lower protein diets (range: 10–23 en%). Findings from this review have been incorporated in the updated DNSG (Diabetes Nutrition Study Group) dietary recommendations for people with type 2 diabetes. Overall, evidence from large UK population-based cohorts consistently shows that substituting meat SFA with dairy SFA is associated with lower incidence of CVD. Analyses by types of meat and dairy consistently reveal lower CVD incidence with substitution of processed meat SFA with milk or cheese SFA. These findings suggest the importance of considering food sources of SFA and align with current recommendations to limit processed meat intake, with possible benefits of dairy foods. Unprocessed meat, specifically, unprocessed poultry, seems to be most beneficial for cardiovascular health, suggesting that unprocessed meat and processed meat should have distinct guidelines. Moreover, increasing dietary protein intake, in a healthy diet, does not have deleterious effects on cardiovascular health. Observational and experimental studies are needed to examine the associations of specific animal- and plant-based protein sources with CVD and other health outcomes, in diverse populations, to better inform specific food-based dietary recommendations for chronic disease prevention.