ABSTRACT Fruit and vegetable intake is not uniform across levels of socioeconomic status (SES) and researchers have identified low SES as a risk factor for poor intake of fruits and vegetables. In an effort to eliminate public health disparities and increase fruit and vegetable intake, the Women, Infant, and Children (WIC) program implemented additional food assistance programs, with a specific emphasis on fresh fruits and vegetables. The Farmers' Market Nutrition Program (FMNP) provides pre-existing WIC clients with coupons to purchase fresh, locally grown produce at farmers' markets. In addition, Congress also approved the WIC Cash Value Voucher (CVV) program, which provides WIC participants with vouchers to purchase fresh fruits and vegetables at farmers' markets or grocery stores. The purpose of this thesis was to investigate the relation of FMNP coupon use with accessibility and WIC CVV redemption rates at farmers' markets. Furthermore, this thesis addressed whether WIC shoppers redeemed a higher percentage of their WIC CVV value at farmers' markets or grocery stores. WIC CVV and FMNP issuance and redemption data were analyzed to establish overall redemption rates and total perecent of WIC CVV value redeemed. Accessibility was assessed using the Geographic Information System, which allowed me to calculate the distance that WIC participants would have to travel to redeem their FMNP coupons at FMNP-approved farmers' markets. The results showed that less than 1% of WIC shoppers redeem their WIC CVVs at farmers'markets in Arizona. However, the redemption of WIC CVV was significantly higher during the months when shoppers had the option of using both WIC CVV and FMNP coupons at farmers' markets. Furthermore, the percent of total CVV value redeemed at farmers' markets was 99%, significantly higher than grocery stores (93.5%). Average FMNP coupon redemption rates for 2008-2010 was 43.3%, well below the national average of 59%. However, my spatial analysis revealed that there was no significant association between the distance traveled to farmers' markets and FMNP redemption rates. This indicates that the distance traveled to farmers' markets is not a major barrier to redemption of FMNP coupons in Arizona.

Mexican Americans have an increased risk for type 2 diabetes and premature cardiovascular disease (CVD). The association of hyperglycemia with traditional CVD risk factors in this population has been established, but there is limited data regarding other non-traditional CVD risk factors. Thus, this cross-sectional study was conducted to evaluate CVD risk among Mexican Americans by measuring concentrations of lipids, high-sensitivity C-reactive protein (hsCRP), and cholesterol in low-density-lipoprotein (LDL) and high-density-lipoprotein (HDL) subfractions. Eighty overweight/obese Mexican-American adults participating in the Maricopa Insulin Resistance Initiative were randomly selected from each of the following four groups (n = 20 per group): nomolipidemic
ormoglycemic controls (NC), dyslipidemic
ormoglycemic (DN), dyslipidemic/prediabetic (DPD) and dyslipidemic/diabetic (DD). Total cholesterol (TC) was 30% higher among DD than in NC participants (p<0.0001). The DPD group had 27% and 12% higher LDL-C concentrations than the NC and DN groups, respectively. Similarly, LDL-C was 29% and 13% higher in DD than in NC and DN participants (p=0.013). An increasing trend was observed in %10-year CVD risk with increasing degree of hyperglycemia (p<0.0001). The NC group had less cholesterol in sdLDL particles than dyslipidemic groups, regardless of glycemic status (p<0.0001). When hyperglycemia was part of the phenotype (DPD and DD), there was a greater proportion of total and HDL-C in sHDL particles in dyslipidemic individuals than in NC (p=0.023; p<0.0001; respectively). Percent 10-year CVD risk was positively correlated with triglyceride (TG) (r=0.384, p<0.0001), TC (r=0.340, p<0.05), cholesterol in sdLDL(r=0.247; p<0.05), and TC to HDL-C ratio (r=0.404, p<0.0001), and negatively correlated with HDL-C in intermediate and large HDL(r=-0.38, p=0.001; r=0.34, p=0.002, respectively). The TC/HDL-C was positively correlated with cholesterol in sdLDL particles (r=0.698, p<0.0001) and HDL-C in sHDL particles (r=0.602, p<0.0001), and negatively correlated with cholesterol in small (r=-0.35, p=0.002), intermediate (r=-0.91, p<0.0001) and large (r=-0.84, p<0.0001) HDL particles, and HDL-C in the large HDL particles (r=-0.562, p<0.0001). No significant association was found between %10-year CVD risk and hsCRP. Collectively, these results corroborate that dyslipidemic Mexican-American adults have higher CVD risk than normolipidemic individuals. Hyperglycemia may further affect CVD risk by modulating cholesterol in LDL and HDL subfractions.

Among its many roles in the body, ascorbic acid functions as a cofactor in carnitine and catecholamine synthesis, metabolites involved in fat oxidation and mood regulation, respectively. Given that fat oxidation and mood affect one's feelings of vigor, I hypothesized that those with lower levels of plasma ascorbic acid would be less likely to exercise at high levels than individuals with adequate or high levels of vitamin C. To test this, I conducted a double-blind, placebo-controlled intervention. A group of healthy, non-smoking males between the ages of 18 and 40 were put on a vitamin C-restricted diet for two weeks and then randomized to a control group that received placebo capsules for six weeks or an intervention group that received 500 mg of vitamin C daily for six weeks. The men were restricted from eating foods high in vitamin C, instructed to wear a pedometer daily and to record their step counts, and to take a pill daily (either the placebo or vitamin C supplement). Unexpectedly, the subjects receiving the intervention had lower step counts than the control group; the control group, rather than the vitamin C group, significantly (p=0.017) increased their steps at week 8 compared to week 2. However, I also estimated daily Metabolic Equivalent Tasks (METs), and subjects receiving the placebo had lower MET outputs than subjects receiving vitamin C at the end of the trial, in spite of having higher step counts. This means the intensity of their activity was higher, based on METs expenditure. Additionally, depression scores (POMS-D) as measured by the Profile of Mood States (POMS) questionnaire were significantly higher (p=0.041) among subjects receiving the placebo at the end of the study. These latter results are consistent with my expectations that subjects with higher levels of plasma vitamin C would have improved mood and higher energy output than subjects with low levels of vitamin C.