Due to their cholesterol content, limiting egg intake has been widely recommended for prevention of cardiovascular disease (CVD). However, recent reports by the 2015 Dietary Guidelines Committee and AHA/ACC suggest that there is insufficient evidence that dietary cholesterol is appreciably associated with blood cholesterol. In addition, the literature on the association with type 2 diabetes (T2D) is inconsistent. These analyses aim to determine the association between egg intake and the risk of CVD and T2D in the Nurses’ Health Study (NHS), the Nurses’ Health II Study (NHSII), Health Professionals Follow-up Study (HPFS) and the NIH-AARP Diet and Health Study (AARP), and conduct a meta-analyses. Egg intake was assessed via validated semi-quantitative food frequency questionnaires. Cox proportional hazard models, adjusted for age, lifestyle and dietary factors, were used to estimate relative risks (HR) and 95% Confidence Intervals (CI). We observed 12,832 and 16,570 cases of incident CVD and T2D in NHS, NHSII and HPFS, and 11,268 CVD mortality cases in AARP. An increase of one egg per day was not associated with risk of CVD in NHS, NHSII and HPFS (HR: 1.04, 95% CI: (0.96, 1.13)). In the AARP study an increase of one egg per day was associated with an increased risk of CVD mortality (HR: 1.13, 95% CI: (1.05, 1.20)) and, an increased risk of CVD mortality among diabetics (HR: 1.25, 95%CI: (1.11, 1.41)). One egg per day was associated with an increased risk of CVD (HR: 1.05, 95% CI: (1.01, 1.09)) in a meta-analysis of the current results and previously published studies. We also saw an increased risk of CVD among diabetics (HR: 1.24, 95% CI: (1.12, 1.37)). We observed an increased risk of T2D with an increase of one egg per day (HR: 1.09, 95% CI: (1.01, 1.18)) in NHS, NHSII and HPFS, and in the meta-analysis (HR: 1.13, 95% CI: (1.07, 1.19)). Although there does not seem to be a significant association between egg intake and risk of CVD or T2D in healthy individuals, people at risk for CVD or T2D and those who currently have T2D may want to limit egg intake.

Age-related macular degeneration (AMD) is the most common cause of irreversible blindness in older Americans. There has been a long standing interest in the role of diet in the development of AMD. As early as the first National Health and Nutrition Examination Survey in the 1970s, higher intakes of fruits and vegetables were inversely correlated with the prevalence of AMD. Carotenoids and omega3 fatty acids are the most studied dietary factors due to strong biological plausibility. However, evidence from epidemiologic studies and clinical trials on the relations has been inconsistent. Chapter I prospectively examined the intakes of lutein/zeaxanthin and other common carotenoids in relation to the risk of AMD over more than two decades of follow-up among two large US cohorts, the Nurses’ Health Study and Health Professionals Follow-Up Study. We assessed nutrient intakes by repeated food frequency questionnaires. We also computed bioavailable plasma carotenoid scores directly from food intake using validated regression models. Cox proportional hazards models were used to compute the associations. Higher intakes of bioavailable carotenoids (except lycopene) were inversely associated with advanced AMD but not intermediate AMD. Analyses based on bioavailable intakes resulted in stronger associations than conventional nutrient intakes. Chapter II prospectively evaluated the marine long-chain omega3 fatty acids. We found that long-chain omega3 fatty acids were inversely associated with visually significant intermediate AMD. There was no association with advanced AMD; however, the totality of current evidence for advanced AMD is also discordant. Chapter III further investigated the plant-derived omega3 fatty acids, α-linolenic acid (ALA). We found that higher intake of ALA was associated with intermediate AMD before 2002 but not after. This coincides with the same time period when trans ALA was found in our participants’ blood and in mayonnaise, a primary food source of ALA. Whether trans ALA mediates this positive association warrants further studies. Although randomized trials are usually believed as the “gold standard”, dietary factors are hard to be adequately studied by randomized trials due to the complexities of diet and disease relations. Thus, findings in this thesis from large long-term prospective cohort studies provide the next best form of evidence.

In Chapters 1 and 2, we examined two key components of heathy dietary patterns, specific types of dietary fat and consumption of fruit and vegetables, in relation to total and cause-specific mortality in two prospective cohort studies, the Nurses’ Health Study and the Health Professionals Follow-up Study. Dietary intake was assessed using semi-quantitative food frequency questionnaires at baseline and updated every 2 to 4 years. In Chapter 1, we found that higher saturated and trans fat intakes were associated with higher mortality, whereas polyunsaturated and monounsaturated fat intakes were inversely associated with mortality. Replacing 5% of energy from saturated fats with equivalent energy from polyunsaturated fats and monounsaturated fats was associated with 27% and 13% estimated reductions in total mortality, respectively. Intake of n-6 polyunsaturated fat, especially linoleic acid, was inversely associated with mortality, while marine n-3 polyunsaturated fat intake was associated with a modestly lower total mortality. In Chapter 2, we observed an inverse and nonlinear association between fruit and vegetable consumption and mortality. Consumption of five servings of fruit and vegetables daily, two servings for total fruit and three servings for total vegetables, were associated with the lowest total mortality, but above that level, higher consumptions were not associated with additional risk reductions. Higher consumptions of most fruit and vegetable subgroups were associated with lower risks of total mortality, whereas higher intakes of starchy vegetables such as peas and corn were associated with slightly higher risk of total mortality. In Chapter 3, we investigated the interrelationships between plasma ceramide concentrations, Mediterranean dietary pattern and cardiovascular disease (CVD) in the PREDIMED trial, a randomized controlled trial on the Mediterranean diet for primary prevention of CVD, using a case-cohort design. We observed strong positive associations between plasma ceramide concentrations and CVD risk. The association between ceramide concentration and incident CVD significantly varied by intervention groups. A Mediterranean diet may mitigate the deleterious effects of elevated plasma ceramide concentration. In summary, our findings from the three studies support current dietary recommendations to replace saturated and trans fat with unsaturated fats, increase fruit and vegetable consumption and adopt healthy Mediterranean-style dietary patterns.

The home nutrition environment and the consumer nutrition environment present two important settings for addressing the high prevalence of overweight and obesity among children and adults. They provide the broader context for the three papers that constitute this dissertation. Chapters 1 and 2 are situated within the home nutrition environment and further our understanding of the role that fathers play in child feeding. Semi-structured interviews were conducted with 40 fathers and qualitative methods were used to organize and analyze the data. Both papers draw from the same data set. Chapter 1 identified the strategies used by fathers to feed their children or, their food parenting practices. Thematic analysis was conducted to identify 13 responsive practices and 7 unresponsive practices. Differential use of food parenting practices was found by fathers’ education and residential status. Chapter 2 examined how fathers and mothers co-parent around responsibilities for child feeding tasks and FPPs used. Sixty two percent of the fathers used a variety of approaches to manage planning, procuring and preparing food along with the mother. Co-operative FPPs around structuring, monitoring and organizing the child’s meals were reported by 46% of fathers. Forty percent reported instances of conflicting FPPs regarding access to energy-dense, nutrient-poor snacks and introducing variety in the diet. Dissimilarities in practices were driven by differences in parental eating habits, feeding philosophies and concern for child health. They resulted in the practices of one parent being undermined and in child tantrums or refusal to eat. Chapter 3 is based within the consumer nutrition environment. The goal of this randomized, controlled, experimental study was to test the influence of different formats for displaying added sugars on consumer understanding, perceptions, and purchase intentions. Amazon’s Mechanical Turk was used to recruit 2,509 U.S adults. Participants were randomized to one of eight display formats and responded to a quiz that assessed study outcomes. Displaying added sugars in relative formats (grams accompanied by high/medium/low text, % DV, or the combination of the two) led to the most accurate understanding of added sugars content and judgments about product healthfulness. None of the eight display formats impacted purchase intentions.

Background: Food marketing has been implicated as a driver of obesity. However, few studies have examined point-of-purchase marketing in supermarkets and restaurants, or marketing in lower-income countries. Furthermore, policy solutions to counteract marketing and provide consumers with objective nutritional information require evidence of efficacy. Paper 1. We documented child-oriented marketing practices, product claims, and health-evoking images on 106 cereals sold in Guatemala City, Guatemala. Linear regression was used to evaluate the association between various marketing strategies and nutritional quality. We found that child-oriented cereals had worse overall nutritional quality and higher sugar content compared to non-child oriented cereals. Cereals with health claims were not significantly healthier than those without claims. Paper 2. Menu items marketed on general and kids’ menu boards and signage were recorded at McDonald’s, Burger King, Wendy’s, and Taco Bell throughout the United States in 2010 and 2013. Linear regression was used to model changes over time in nutritional quality of marketed foods and beverages. On general menu boards, marketed items became healthier, improving in overall nutritional quality and decreasing in calories. This pattern was evident in all chains except Taco Bell, where marketed items increased in calories. On general menu boards and signage, while marketed foods improved in nutritional quality, marketed beverages remained the same or became worse. However, on kids’ menu boards, both foods and beverages became healthier. Paper 3. 1,958 adults completed an online survey. Participants were randomized to one of six labeling groups: No label; Calories per bottle; Traffic light; Warning label; Calories+Warning; Traffic Light+Warning. Differences in purchase intentions and perceptions for SSBs across groups were assessed with ANCOVAs. Single health warning labels on SSBs appeared as effective as single traffic lights or combined calories plus warning labels. Warning labels were most effective when combined with traffic light labels in conveying health-related risks associated with SSB overconsumption. Conclusions: Although marketed products in fast food restaurants in the United States appear to be showing modest improvements in nutritional quality, the increase in child-oriented marketing on nutritionally-poor items in other countries is concerning. Continued research on nutrition-minded policy solutions is essential.

Undernutrition in early life increases children’s risk of mortality, morbidity, and impaired growth and development. This thesis analyzes data from two randomized controlled trials in Dar es Salaam, Tanzania. The first trial assessed the effect of daily multivitamin (vitamins B-complex, C and E) supplementation on mortality and morbidity in infants born to HIV-infected mothers. 2387 infants were randomized to multivitamins or placebo at 6 weeks and followed-up for two years. The second trial assessed the effect of zinc and/or multivitamins (vitamins B-complex, C and E) on morbidity in infants born to HIV-uninfected mothers. 2400 infants were randomized to either zinc + multivitamins, zinc only, multivitamins only, or placebo at 6 weeks and were followed for 18 months. Chapter 1 assesses the effect of zinc and/or multivitamin supplements in the second trial on longitudinal child growth – defined by change in height-for-age, weight-for-age and weight-for-height z-scores (HAZ, WAZ and WHZ) and stunting, underweight and wasting (<-2 SD 2006 WHO standard for each indicator respectively). We found small, but significant effects of supplements on change in WHZ and WAZ, but did not find a statistically significant effect of zinc and/or multivitamin supplements on stunting, wasting or underweight. Chapter 2 assesses the effect of zinc and/or multivitamin supplements in the second trial on early child development (ECD) assessed using the cognitive, motor (fine and gross) and language (receptive and expressive) scales of the Bayley Infant Scales of Development 3rd Edition (BSID-III). We did not find a significant effect of supplements on early childhood development as assessed by the BSID-III. Chapter 3 pools the two trials in order to compare mortality, morbidity and growth in HIV-infected, HIV-exposed-but-uninfected (HIV-EU) and HIV unexposed infants. HIV-infected children had the highest rates of mortality, morbidity and growth failure. HIV-EU infants had higher rates of mortality and morbidities than unexposed infants; but lower rates than their HIV-infected peers. Conclusions: Alternative approaches (beyond zinc and/or multivitamin supplementation) to improve growth and ECD in vulnerable populations should be pursued. Child health interventions should target not only HIV-infected but also HIV-EU children, given their increased susceptibility to morbidity and mortality.

Recent findings from genome-wide association studies and pharmaceutical trials of CETP inhibitors suggest that HDL cholesterol might not fully capture the anti-atherogenic functions of HDL. The protein component of HDL might be more informative in explaining HDL functions because proteins have characteristics that can modify the biological properties of HDL. In Chapter 1, I examined whether some proteins located in the HDL size or density range by proteomics studies were present in distinct apoA-I HDL subspecies that could be identified and quantified by a novel sandwich ELISA. Known HDL subspecies containing apoA-II, apoC-III and apoE comprised 70%, 6% and 10%, respectively, and found that novel HDL subspecies containing apoC-I, apoJ, alpha-1 antitrypsin, plasminogen, and apoC-II comprised 14%, 7%, 5%, 3% and 1%, respectively, of the total apoA-I in HDL. HDL that contains apoC-III has been associated with a higher prevalence of obesity and an elevated risk of CHD, opposite to what has been seen for HDL that does not contain apoC-III. To further our knowledge on apoC-III-based HDL subspecies, I investigated their associations with insulin sensitivity measured by oral glucose tolerance test (OGTT) in Chapter 2 and early stages of carotid atherosclerosis measured as carotid intima-media thickness (cIMT) in Chapter 3. Using the novel sandwich ELISA established in Chapter 1, I measured the concentrations of apoA-I in HDL with and without apoC-III in the European multi-center “Relationship between Insulin Sensitivity and Cardiovascular disease” (RISC) study. ApoA-I in HDL with and without apoC-III demonstrated significantly opposite associations with both 3-year change in insulin sensitivity and cIMT at baseline. ApoA-I with apoC-III was associated with a decrease in insulin sensitivity and higher cIMT, whereas the concentration of apoA-I without apoC-III was associated with an increase in insulin sensitivity and lower cIMT. The results for apoC-III itself in HDL was consistent with apoA-I with apoC-III. Total apoA-I was null in both cases. These findings suggest that the presence of apoC-III on HDL diminishes and impairs otherwise beneficial effects of HDL on glucose regulation and atheroprotection, and support the potential of HDL apoC-III as a promising target for diabetes and atherosclerosis prevention and treatment.

Plant-based diets, defined as “vegetarian” diets, are associated with reduced risk of type 2 diabetes (T2D) and coronary heart disease (CHD). To examine the health effects of gradual reductions in animal food consumption while increasing plant food intake, and to distinguish between healthy and less healthy plant foods, we created three graded plant-based diet indices using semi-quantitative food frequency questionnaires (SFFQ). In the overall plant-based diet index (PDI), all plant foods received positive scores while animal foods received reverse scores. In the healthful PDI (hPDI) healthy plant foods (e.g. whole grains, fruits) received positive scores, while less healthy plant foods (e.g. sweetened beverages, refined grains) and animal foods received reverse scores. In the unhealthful PDI (uPDI) less healthy plant foods received positive scores, while healthy plant foods and animal foods received reverse scores. This dissertation examined the reliability and validity of these diet indices, and evaluated their associations with T2D and CHD incidence. In chapter 1, we used data from The Women’s and Men’s Lifestyle Validation Studies (n=1354) to examine the reliability and validity of SFFQ-assessed plant-based diet indices. We found reasonable one-year reliability for the SFFQ-assessed indices. The indices correlated with energy-adjusted 7-day diet record nutrients and plasma biomarkers in expected directions, with hPDI associated with high dietary quality, and uPDI associated with poor diet quality. In chapters 2 and 3, we examined the associations of these indices with T2D and CHD. We included ~70,000 women from Nurses’ Health Study (NHS) (1984-2012), ~90,000 women from NHS2 (1991-2011), and ~40,000 men from Health Professionals Follow-up Study (1986-2010). Dietary data were collected every 2-4 years using SFFQs. We documented 16,162 incident T2D, and 7754 incident CHD cases during ~4,00,000 person-years of follow-up. In pooled multivariable-adjusted analysis, PDI was inversely associated with T2D and CHD. hPDI had a stronger inverse association with both endpoints, while uPDI was positively associated with both diseases. In conclusion, we found reasonable reliability and validity for three graded plant-based diet indices assessed with SFFQs. Our study suggests that plant-based diets, especially when rich in high-quality plant foods, are associated with substantially lower risk of T2D and CHD.

Reducing maternal and child mortality was established as a global priority with the signing of the Millennium Declaration in September 2000. Neonatal vitamin A supplementation and very early breastfeeding initiation are scalable interventions which may improve infant survival. Although breastfeeding has proven benefits for infant health, the potential health consequences of breastfeeding for HIV-infected women are not well studied. In paper one, “The effect of neonatal vitamin A supplementation on morbidity and mortality at 12 months: A randomized trial”, we assessed the efficacy of neonatal vitamin A supplementation (NVAS) in reducing infant morbidity and mortality. Using data from an individually randomized clinical trial of 31,999 infants in Tanzania, we found that NVAS did not affect the risk of death or the incidence of morbidities. However, we noted that postpartum maternal vitamin A supplementation modified the effect of neonatal vitamin A supplementation on infant mortality. In paper two, “Effect of delayed breastfeeding initiation on infant survival: a systematic review and meta-analysis”, our objective was to synthesize the evidence regarding the association between breastfeeding initiation time and infant morbidity and mortality. We pooled five studies, including 136,047 infants. We found a clear dose-response relationship; the risk of neonatal mortality increased with increased delay in breastfeeding initiation. We found a similar pattern when the analysis was restricted to exclusively breastfed infants or low birthweight infants. There was limited evidence regarding the association between breastfeeding initiation time and infant morbidity and growth. We concluded that health policy frameworks and models to estimate newborn and infant survival should consider the independent survival benefit associated with early initiation of breastfeeding. In paper three, “Breastfeeding and Maternal Health among HIV-infected Women in Tanzania”, our objective was to assess the relationship between infant feeding practices and the incidence of maternal mortality, morbidity, and indicators of poor nutritional status from six weeks to two years postpartum in a prospective cohort of Tanzanian women living with HIV. We concluded that breastfeeding may be associated with mixed health outcomes. Additional research should investigate whether HIV-infected women require nutritional support, in addition to antiretroviral therapy, during and after lactation.

Although testicular cancer is a relatively rare cancer in Canada, accounting for only 1.1% of all malignant neoplasms in males, it is the most common cancer among men 20 to 45. Understanding of the causes of testicular cancer risk in general, and the association with diet in particular, remains limited. Data from the National Enhanced Cancer Surveillance System were used to explore the relationship of diet and testicular cancer risk. There were 601 cases of testicular cancer and 744 controls available for study. We systematically examined 17 food groups, 15 nutrients, and 7 individual foods based on data collected through a 69-item food-frequency questionnaire. Our results suggest that higher dairy product intake, specifically cheese, is associated with a higher risk of testicular cancer in Canadian males. Risk differences were observed between histological subtypes of testicular cancer. This thesis provides a basis for future studies designed to address testicular cancer etiology.