Background: Hispanics account for more than half of the total United States (US) population growth between 2000 and 2010. To gain a comprehensive understanding of a predominantly Hispanic urban community in Northern Manhattan, the aims of this cross-sectional observational study were: (1) to characterize the actual and perceived neighborhood food environment in Northern Manhattan, (2) to understand the relationship between the actual and perceived neighborhood food environment, sociodemographic characteristics and the likelihood of consuming five or more servings of fruits and vegetables per day, and (3) to describe the contribution of participants’ sociodemographic characteristics and health behavior to their health outcomes. Methods: This cross-sectional observational study was undertaken as part of the larger Washington Heights/Inwood Informatics Infrastructure for Comparative Effectiveness Research (WICER) project. English or Spanish-speaking Hispanic participants (n=4,019) 18 years and older living in Northern Manhattan’s five ZIP codes were recruited and interviewed by English-Spanish bilingual community health workers. Food outlets selling fruits and vegetables were identified using the North American Industry Classification System (NAICS) definitions obtained from the ReferenceUSA’s national business database. The neighborhood food environment was characterized by integrating the geocoded addresses of WICER study participants with external geographic-level data on food outlets present in the participants’ respective 0.25-mile and 0.5-mile residential radii. Data were analyzed using bivariate and multivariate binary logistic regressions. Results: The food outlet types that sell fruit and vegetable in Northern Manhattan include small and medium/large size Supermarket/Grocery store, Meat Market, and Fruit and Vegetable Market. The majority of these food outlets (91.5%) are single location stores that have a smaller store space. The presence of Fruit and Vegetable Markets (2+ Stores in 0.25-mile: OR=1.59, p = 0.003; 1 Store in 0.5-mile: OR=2.28, p = 0.008; 2+ Stores in 0.5-mile: OR=3.10, p = 0.00) significantly increase the odds of participant’s perception that a large selection of fresh fruits and vegetables is available in their neighborhood. The presence of Fruit and Vegetable Markets (2+ Stores in 0.25-mile: OR=1.51, p = 0.003; 1 Store in 0.5-mile: OR=2.25, p = 0.004; 2+ Stores in 0.5-mile: OR=3.31, p = 0.00) as well as the presence of medium/large size Supermarket/Grocery in 0.25-mile (OR=1.05, p = 0.013) significantly increase the odds of participant’s perception that the fresh fruits and vegetables in their neighborhood are of high quality whereas the presence of Meat Market in the participant’s 0.25-mile (OR=0.74, p = 0.002) significantly lower the odds. The presence of Fruit and Vegetable Markets (1 Store in 0.25-mile: OR=1.23, p = 0.047; 2+ Stores in 0.25-mile: OR=1.37, p = 0.020; 2+ Stores in 0.5-mile: OR=1.94, p = 0.018) as well as the presence of medium/large size Supermarket/Grocery (0.25-mile: OR=1.05, p = 0.020; 0.5-mile: OR=1.05, p = 0.018) significantly increase the odds of participant’s perception that a large selection of low-fat products is available in their neighborhood whereas the presence of Meat Market in the participant’s 0.25-mile (OR=0.83, p = 0.042) significantly lowers the odds. Variables that significantly increase the participants’ odds of consuming five or more servings of fruits and vegetables per day include having more than a high school education (0.25-mile and 0.5-mile models: OR=1.62, p = 0.004) and being foreign-born (0.25-mile model: Foreign-born in Dominican Republic: OR=1.77, p = 0.032; Foreign-born outside of the United States or the Dominican Republic: OR=2.44, p = 0.007; 0.5-mile model: Foreign-born in the Dominican Republic: OR=1.73, p = 0.040; Foreign-born outside of the United States or the Dominican Republic: OR=2.48, p = 0.006). In contrast, the participants’ perception that a large selection of fresh fruits and vegetables is available in their neighborhood (0.25-mile model: OR=0.63, p = 0.011; 0.5-mile model: OR=0.64, p = 0.016) and the presence of Fruit and Vegetable Market in their 0.5-mile radius (1 Store: OR=0.32, p = 0.006; 2+ Stores: OR=0.38, p = 0.009) significantly lower the odds. Variables that significantly increase the odds of body mass index (BMI) in the overweight range were age (OR=1.02, p = 0.00), being foreign-born outside of the United States or the Dominican Republic (OR=1.76, p = 0.006), self-reported diabetes (OR=1.37, p = 0.026), and perceived weight as overweight (OR=4.46, p = 0.00) whereas being female (OR=0.67, p = 0.00) significantly lowers the odds. Variables that significantly increase the odds of BMI in the obese range were age (OR=1.02, p = 0.00), self-reported diabetes (OR=1.78, p = 0.00), and perceived weight as overweight (OR=19.39, p = 0.00) whereas having more than a high school education (OR=0.72, p = 0.021) significantly lowers the odds. Variables that significantly increase the odds of hypertension were age (OR = 1.04, p = 0.00) and self-reported diabetes (OR = 1.57, p = 0.00) whereas being female (OR = 0.72, p = 0.00) significantly lowers the odds. Education (>High School) significantly increases the odds (OR=1.43, p = 0.00) of self-report of good health. In contrast, variables that significantly lower the odds were age (OR=0.98, p = 0.00), being female (OR=0.60, p = 0.00), higher fruit and vegetable consumption (OR=0.66, p = 0.007), self-reported diabetes (OR=0.51, p = 0.00), and obesity (OR=0.64, p = 0.00). Variables that significantly increase the odds of self-report of good health include having more than a high school education (OR = 1.43, p = 0.00) whereas age (OR = 0.98, p = 0.00), female gender (OR = 0.60, p = 0.00), higher fruit and vegetable consumption (OR = 0.66, p = 0.007), self-reported diabetes (OR = 0.51, p = 0.00), and obesity (OR = 0.64, p = 0.00) significantly lower the odds. Conclusion: This study contributed to our understanding of the relationships among neighborhood food environment, health behaviors, and health outcomes in a predominantly Hispanic underserved urban community in New York City. While most findings were similar to those reported in the literature, our findings related to the relationship between participants’ perceived neighborhood food environment and actual healthy food access and fruit and vegetable consumption were in contrast to other studies in that increased perceived availability and actual availability lowered the odds of consuming five or more servings of fruits and vegetables per day. This surprising finding merits additional qualitative and quantitative research to examine the complex relationships among perceived access, availability, and consumption of healthy foods as well as improved measures of fruit and vegetable consumption.

Aim: To determine whether the tuck shop purchasing habits of Grade 4 learners were contributing towards the development of childhood overweight and obesity. Objectives: To assess the nutritional quality of the food and beverages available for learners to purchase; items regularly purchased from the tuck shop as well as factors influencing the learner’s decision to purchase these items; the anthropometric and socio-demographic characteristics of grade 4 learners as well as their nutrition knowledge related to the tuck shop items purchased. Method: A survey administered to 11 tuck shop managers, a questionnaire administered to 311 Grade 4 learners and two single-sex focus groups of 5 learners each were conducted. Results: Fifty six percent of the sample were female (n=173) and 44% were male (n=138). Twenty seven percent of the study sample was overweight (n = 83) and 27% were obese (n = 85). Eighty six percent of learners (n = 266) claimed to buy from their school tuck shop. Twenty two percent of learners purchased from their tuck shop at least three times per week (n =58). Learners who purchased from the tuck shop had a significantly higher BMI than those who did not (p = 0.020). Learners who purchased from the tuck shop spent on average R8,38 per day with a minimum of R1 and a maximum of R40 (standard deviation R5.39). The most popular reasons for visiting the tuck shop included “this is my favourite thing to eat or drink” (66.5%, n = 177) and “I only have enough money to buy this item” (47.0%, n = 125). Savoury pies were the most popular "lunch" item for all learners for both food breaks (45%, n = 5 schools and 27.3%, n = 3 schools) selling the most number of units (43) per day at eight of the eleven schools (72.7%). Iced popsicles were sold at almost every school, ranked as the cheapest beverage and also sold the most number of units (40.7). Healthy beverages sold included canned fruit juice and water, while healthy snacks consisted of dried fruit, fruit salad, bananas, yoghurt and health muffins. The average healthy snack contained almost half the kilojoules of its unhealthy counterpart (465kJ vs 806kJ). Nutritional analyses of the healthy lunch options revealed total fat contents that exceeded the DRI and South African recommended limit. Perceived barriers to stocking healthy items included cost and refrigeration restrictions. The average score for the food groups was only 33% indicating that learners were not familiar with the Food Based Dietary Guidelines (FBDG). Further analyses showed that the total knowledge scores of those learners that reported to buy from the tuck shop frequently, was significantly lower when compared to the total knowledge scores of those learners who bought from the tuck shop less frequently (13.0 ± 3.9 and 11.6 ± 3.1, respectively; p < 0.05). Logistic regression analysis confirmed that the total knowledge of a learner could be used to predict whether he or she is more likely to make purchases from the tuck shop (significance = 0.017). Focus group results revealed that learners are aware of “healthy” and “unhealthy” tuck shop items. Most learners stated that they would continue to purchase items from their tuck shop if all “unhealthy” items were removed. Conclusion: Primary school tuck shops of well resourced schools in Pietermaritzburg are contributing to childhood overweight and obesity through a combination of factors. These include the poor nutritional quality of the items stocked at the tuck shop as well as the poor tuck shop purchasing practices. Much consultation is required amongst dieticians, school principals and privatised tuck shop managers to overcome barriers to stocking healthy items. School management and government have an important role to play in imposing restrictions on the sale of unhealthy items; along with improving the quality of the nutrition education curriculum to ensure that learners are able to translate their knowledge into healthier purchasing practices. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.