Advertisement

Food Purchasing Selection Among Low-Income, Spanish-Speaking Latinos

      Background

      In the U.S., poverty has been linked to both obesity and disease burden. Latinos in the U.S. are disproportionately affected by poverty, and over the past 10 years, the percentage of overweight U.S. Latino youth has approximately doubled. Buying low-cost food that is calorie-dense and filling has been linked to obesity. Low-income individuals tend to favor energy-dense foods because of their low cost, and economic decisions made during food purchasing have physiologic repercussions. Diets based on energy-dense foods tend to be high in processed staples, such as refined grains, added sugars, and added fats. These diets have been linked to a higher risk of obesity, type 2 diabetes, and cardiovascular disease.

      Purpose

      This pilot study conducted ethnographic qualitative analyses combined with quantitative analyses to understand grocery shopping practices among 20 Spanish-speaking, low-income Latino families. The purpose was to analyze food selection practices in order to determine the effect of nutrition education on changes in shopping practices to later develop educational tools to promote selection of healthier food options.

      Methods

      Participants received tailored, interactive, nutrition education during three to five home visits and a supermarket tour. Grocery store receipts for grocery purchases collected at baseline and at the end of the project were analyzed for each family to extract nutritional content of purchased foods. Nutritional content was measured with these factors in mind: quantity, calories, fats, carbohydrates, fiber, protein, and percentage of sugary beverages and processed food. Data were collected in 2010–2011 and analyzed in 2011–2012.

      Results

      After receiving between three and five home-based nutrition education sessions and a supermarket tour over a 6-month period, many families adopted instructions on buying budget-friendly, healthier alternative foods. Findings indicate that participating families decreased the total number of calories and calories per dollar purchased from baseline to post-education (median total calories: baseline, 20,191; post-education, 15,991, p=0.008); median calories per dollar: baseline, 404; post-education, 320, p=0.008). The median grams of carbohydrates per dollar (baseline, 66, post-education, 45) and median calories from processed food (baseline, 11,000, post-education, 7845) were not reduced (p=0.06).

      Conclusions

      This pilot study demonstrated that grocery shopping practices are an important factor to address in nutrition education among Spanish-speaking, low-income individuals, and that there may be ways to encourage low-income, Latino families to purchase healthier foods. Findings challenged arguments suggesting that such an approach is not possible because of the high cost of healthier foods.

      Introduction

      Latino Children and Obesity

      One half of all Latino children born in or after 2000 will develop diabetes in their lifetime, and overweight and obesity have been prominently implicated as contributing factors.
      • Narayan K.M.
      • Boyle J.P.
      • Thompson T.J.
      • et al.
      Lifetime risk for diabetes mellitus in the U.S..
      Over the past 10 years, the percentage of overweight Latino youth in the U.S. has approximately doubled, and 38.2% of Latino children aged 2–19 years are now overweight or obese.
      • Ogden C.L.
      • Carroll M.D.
      • Curtin L.R.
      • Lamb M.M.
      • Flegal K.M.
      Prevalence of high body mass index in U.S. children and adolescents, 2007–2008.
      Many of these children have substantial metabolic and vascular abnormalities that predispose them to both type 2 diabetes and cardiovascular disease.
      • Caballero A.E.
      • Bousquet-Santos K.
      • Robles-Osorio L.
      • et al.
      Overweight Latino children and adolescents have marked endothelial dysfunction and subclinical vascular inflammation in association with excess body fat and insulin resistance.
      Genetic predisposition and environmental, economic, cultural, and community factors that affect diet all may contribute to Latinos' heightened risk for obesity and type 2 diabetes.
      • Caballero A.E.
      Diabetes in Hispanics/Latinos: challenges and opportunities.
      These factors include poor access to healthy food, family budgets, food preferences, and fewer opportunities for physical activity. Obesity and diabetes together are associated with cardiovascular disease and orthopedic problems, among other conditions.
      • Perrin J.M.
      • Bloom S.R.
      • Gortmaker S.L.
      The increase of childhood chronic conditions in the U.S..

      The Economics of Obesity in the U.S.

      In the U.S., poverty has been linked to both obesity and disease burden,
      • Drewnoski A.
      Obesity and the food environment: dietary energy density and diet costs.
      and some have suggested that obesity in the U.S. is an economic issue.
      • Drenowski A.
      • Darmon N.
      The economics of obesity: dietary energy density and energy cost.
      As of 2010, the poverty rate among Latinos was 26.6% compared to 15.1% for the total U.S. population.
      • DeNavas-Walt C.
      • Proctor B.D.
      • Smith J.C.
      U.S. Census Bureau, current population reports, P60-239, income, poverty, and health insurance coverage in the U.S.: 2010.
      About 27% of children from families with incomes below the federal poverty level are obese, compared to about 10% of children in households with incomes at or above 400% of the poverty level.
      DHHS, Health Resources and Services Administration, Maternal and Child Health Bureau
      Child health USA 2010.
      A persuasive explanation for the relationship between poverty and obesity is economic; low-income individuals tend to buy low-cost food that is more calorie-dense and filling
      • Drewnoski A.
      Obesity and the food environment: dietary energy density and diet costs.
      • Hofferth S.L.
      • Curtin S.C.
      Poverty, food programs, and childhood obesity.
      because energy-dense foods are more affordable. Diets based on energy-dense foods tend to be high on processed staples. Processed foods are more palatable and easier to prepare than nonprocessed foods
      • Rolls B.
      • Barnett R.A.
      Volumetrics.
      ; and high consumption of them has been linked to obesity, type 2 diabetes, and cardiovascular disease.
      • Liu S.
      Intake of refined carbohydrates and whole grain foods in relation to risk of type 2 diabetes mellitus and coronary heart disease.
      Nonprocessed foods provide more health benefits than processed foods.
      WHO, Food and Agriculture Organization
      Report of the joint WHO/FAO expert consultation on diet, nutrition, and the prevention of chronic diseases.
      For low-income individuals, the decisions made tend to favor energy-dense foods because of their low cost, and these decisions have physiologic repercussions.
      • Drenowski A.
      • Darmon N.
      The economics of obesity: dietary energy density and energy cost.
      Although these decisions are greatly influenced by purchasing power, the kind of food individuals have access to also influences what they consume. Low-income neighborhoods tend to offer limited choices of food outlets and healthy food.
      • Wang Y.
      • Beydoun M.A.
      The obesity epidemic in the U.S.—gender, age, socioeconomic, racial/ethnic, and demographic characteristics: a systematic review and meta-regression analysis.
      Studies
      • Hersey J.
      • Anliker J.
      • Miller C.
      • et al.
      Food shopping practices are associated with dietary quality in low-income households.
      • Powell L.M.
      • Chaloupka F.J.
      Food prices and obesity: evidence and policy implications for taxes and subsidies.
      • Powell L.M.
      • Zhao Z.
      • Wang Y.
      Food prices and fruit and vegetable consumption among young American adults.
      have shed light on the relationships among food shopping practices, prices, and diet quality. But little is known about the role of nutrition education on changes in shopping practices, and the need to evaluate nutrition education that is effective in addressing the challenges that low-income families experience has been noted.
      • Hersey J.
      • Anliker J.
      • Miller C.
      • et al.
      Food shopping practices are associated with dietary quality in low-income households.
      The current study analyzed food shopping practices among Spanish-speaking, low-income, Latino families to determine the effect of nutrition education on changes in food shopping practices.

      Methods

      The study used a qualitative research design that included baseline semi-structured interviews, participant observations, home visits, as well as quantitative nutritional analyses of grocery store receipts of food purchased by participants. The methods were reviewed and approved by the University of Massachusetts Boston IRB. The intervention with the families occurred between May 2010 and September 2011. The families received a $50 incentive at the end of the study. Data were collected in 2010–2011 and analyzed in 2011–2012.

      Participants

      A purposive sample of 20 low-income families was recruited in Massachusetts if they met the following selection criteria: Spanish-speaking with children aged <18 years. All parents were foreign-born: 65% came from the Dominican Republic, 17.5% came from Puerto Rico, and 17.5% came from El Salvador. In 65% of the families, the children were born in the U.S., whereas the remaining 35% were born in the Dominican Republic. There were 95 family members who made up the 20 families, with family sizes ranging from three to nine members and a mean family size of 4.8 (SD=1.5). The number of children aged <18 years varied from one to five with a mean of 2.3 children (SD=1.0).
      Monthly family income ranged from $800 to $4500 with a mean of $1613 (SD=$602) and monthly food expenditures fluctuated from $300 to $1000 with a mean of $514 (SD=$204) for an average of 33% of the income going to food. This amount is consistent with national data, showing that low-income families' share of annual household income spent on food is greater than that of high-income families.
      U.S. Department of Labor, Bureau of Labor Statistics
      Consumer expenditure survey.
      Half of the families received Supplemental Nutrition Assistance Program (SNAP) benefits, and these were included in the families' total food expenditures. In terms of medical history, 85% of the families reported a family history of diabetes, and 65% had a family history of obesity and cardiovascular disease.

      Interviews and Participant Observations

      After consenting to participate, baseline demographic data, food purchasing, and consumption were collected using a semi-structured interview developed by the research team. Questions also assessed nutrition knowledge areas that participants sought to improve through this project. After completion of the baseline interview, participants received $50, with instructions to purchase food for breakfast, lunch, dinner, and snacks as usual, and to provide the food store receipts to the research team. The study team evaluated whether transportation was a barrier to accessing the supermarkets and provided transportation to the families in need for food purchasing at baseline and at the end of the intervention, and to complete the supermarket tour. The nutritional value of the grocery store receipts were analyzed (using the U.S. Department of Agriculture's A National Nutrient Database for Standard Reference
      U.S. Department of Agriculture, Agricultural Research Service
      2011. USDA national nutrient database for standard reference, release 24. Nutrient Data Laboratory.
      ), and that information was shared with the families.
      Participant observations were conducted during home visits and shopping trips to the store where families purchase their groceries. At participants' homes, observations were made around the kitchen (e.g., content in pantry and refrigerator) and living room areas. During shopping trips, participants were observed as they made their way through the store aisles.

      Nutrition Education

      Participants received tailored (i.e., based on knowledge gaps and areas of interest noted at baseline) interactive, nutrition education during an average of three to five home visits and a supermarket tour. The nutrition education was based on principles of low health literacy, which included limiting the scope of the topic of interest, using visuals to convey information, and responding to specific learning experiences and needs. A bicultural/bilingual dietician provided the education to each family.
      The content and delivery of the nutrition education sessions were informed by social learning theory.
      • Bandura A.
      Social learning theory.
      The application of principles of social learning theory in the educational sessions allowed addressing the environmental and psychological factors that are antecedents of food purchasing and consumption behaviors. This was achieved by making sure each participant remembered basic, important nutrition information (retention); was able to practice behaviors learned through the educational session (reproduction); and developed a desire or good reason to practice the newly learned behaviors (motivation). This approach was guided by the following rationale: Newly acquired knowledge, combined with the level of motivation for change, could influence participants' perceptions of their community's food environment. This, in turn, could facilitate their choosing healthier food options, despite limited environmental choices and budgets.
      Another educational strategy used to develop the content of nutrition education was informed by a patient-activation framework that has shown to be effective in teaching individuals how to become agents of change.
      • Cortés D.E.
      • Mulvaney-Day N.
      • Fortuna L.
      • et al.
      Patient/provider communication: understanding the role of patient activation for Latinos in mental health treatment.
      Instruction was delivered based on the assumption that participants have the ability to learn how to identify healthier choices that are readily available within their grocery stores when the instruction takes into account their immediate surroundings. This is important, because findings suggest that changes in price or income do not automatically translate into changes in food purchasing patterns.
      • Blisard N.
      • Stewart H.
      • Joliffe D.
      Low-income households' expenditures on fruits and vegetables. U.S. Department of Agriculture, Agricultural Economic Report No. AER833, May 2004.
      The content of the educational sessions focused on teaching participants how to choose affordable and nutritious foods using a family-centered approach. They covered a wide array of learning activities: supermarket tour, how to plan a healthy meal using the plate method, how to read food labels using the family's favorite packaged foods and beverages, the benefits of drinking water, the relationship between sugary drinks and tooth decay, learning how to tweak the favorite family recipe, and how to find low-cost healthy foods. Education also included how to prepare food in a more time- and cost-efficient fashion, as well as the importance of planning for grocery shopping, food preparation and menu planning using the plate method approach. An important factor in obesity is excessive consumption of processed foods in contrast to fruits and vegetables
      • Drenowski A.
      • Darmon N.
      The economics of obesity: dietary energy density and energy cost.
      ; thus, purchases of seasonal vegetables and fruits, frozen vegetables, and dry beans were recommended.
      When teaching participants how to select healthier food items, recommendations were framed to take into account five important factors involved in the process of choosing and purchasing food: taste, cost, convenience, health, and variety.
      • Glanz K.
      • Basil M.
      • Maibach E.
      • Goldberg J.
      • Snyder D.
      Why Americans eat what they do: taste, nutrition, cost, convenience, and weight control concerns as influences on food consumption.
      An important aspect of the educational intervention was the use of hands-on activities, inspecting the content of the kitchen pantry and refrigerator, learning from the labels of the families' favorite foods, and looking for healthier options during the supermarket tour.
      Also, because 85% of the families reported a family history of diabetes, emphasis was placed on consuming whole foods. Each family received an audio-novela, which includes a chapter on diabetes prevention that was developed by the Latino Diabetes Initiative at the Joslin Diabetes Center.
      The Latino Diabetes Initiative at Joslin Diabetes Center
      Finally, each family received the results of the nutritional analysis of their baseline grocery store receipts: the list of purchased food items with calories; fiber and macronutrient content (i.e., fats, carbohydrates, and protein); and calories for the total purchase. This information was used as an educational tool to inform participants about the nutritional value of the foods they purchased at baseline.

      Grocery Store Receipt Analysis

      Quantitative data from grocery store receipts for food purchases collected at baseline and at the end of the project were analyzed for each family to extract nutritional content: quantity, calories, fat, carbohydrates, fiber, protein, and the percentage of sugary beverages and processed food. The following indices were created for each family: total calories, calories per dollar, total grams of macronutrients, and macronutrients per dollar, and calories from processed foods and sweetened beverages. Indices were created for percentages of macronutrients, processed foods, and sweetened beverages.
      There were 17 families with grocery receipt data at both baseline and post-education (one family did not complete the post-education shopping activity, and two other receipts were lost when the dietician's wallet was stolen during a supermarket tour). Median values and interquartile ranges were calculated for the nutritional values of the food items for each family before the intervention and then after the education. The Wilcoxon signed-rank test, a nonparametric test for paired data, was used to identify differences between baseline and post-education (significant if p<0.05). Differences in median values for the nutritional values (total value, value per dollar, and percentage) between those that had a reduction in calories with the intervention and those that did not were tested using the Wilcoxon Mann–Whitney test. All analyses were completed in SAS, version 9.2.

      Results

      Baseline Interview

      Data collected at baseline indicated that 63% (n=12) of the families used a shopping list when doing grocery shopping. Close to one third (n=6) of the families reported some sporadic difficulties getting to a large supermarket. They all shopped at large grocery stores. Half (n=10) of the families indicated planning family menus ahead of time.

      Qualitative Findings

      Much of the food purchased at baseline was calorie-dense, low in fiber, and high in fat and carbohydrates (Table 1). Purchases included malt beverages; cold cereals high in sugar; sugary drinks; fruit drinks; instant ramen noodles; and salami, among other less healthy foods. Offering feedback to families about their food receipts' nutritional analyses provided an opportunity to recommend changes in purchasing patterns.
      Table 1Nutritional value of food items purchased at baseline and for the total household (n=17), median (IQR) unless otherwise noted
      BaselinePost-educationp
      Total calories20,191 (16,966–23,062)15,991 (15,093–22,196)0.008
      Total calories per dollar404 (339–555)320 (302–444)0.008
      Total fat (grams)477 (328–792)468 (359–654)0.52
      Total fat (grams) per dollar10 (7–16)9 (7–13)0.5
      Calories from fat (%)24 (16–32)24 (18–31)0.9
      Calories from cholesterol (%)62 (48–75)58 (47–65)0.33
      Total carbohydrates (grams)3,298 (1,857–5,011)2,240 (1,688–2,791)0.08
      Total carbohydrates (grams) per dollar66 (37–100)45 (34–56)0.06
      Total fiber (grams)182 (146–267)193 (160–278)0.95
      Total protein (grams)689 (585–929)781 (694–929)0.58
      Total protein (grams) per dollar14 (12–19)15 (14–18)0.98
      Calories from protein (%)13.6 (11–18)18 (16–21)0.06
      Total calories from food16,356 (14,869–21,605)15,093 (12,561–16,565)0.03
      Total calories from processed food11,000 (7,384–16,495)7,845 (4,124–11,328)0.06
      Calories from processed foods (%)48 (40–77)51 (27–66)0.38
      Total calories from sugary beverages1748 (901–3962)923 (0–3099)0.27
      Calories from sugary beverages (%)8 (3–14)4 (0–15)0.67
      Note: Boldface indicates significance (p<0.05).
      IQR, interquartile range
      Participants often were surprised to learn about the low nutritional value of many foods they had purchased, and many asked for recommendations they could use to make appropriate changes within their budget constraints. One participant was surprised when she realized that the bread she had bought was not whole wheat. She had purchased “brown” bread with zero fiber content. She made this discovery after she deciphered the nutritional information label, and said: “I am surprised that the foods I bought are deficient in fiber. What I bought is making us sick.” After baseline, many families reported applying lessons learned from the nutrition education to buy budget-friendly, healthier foods.

      Quantitative Findings

      Families decreased the total number of calories (p=0.008) and calories per dollar (p=0.008) purchased between baseline and post-education (Table 1). The median calories purchased by families at baseline was 20,191 (median 404 calories per dollar), which decreased to a median of 15,991 calories purchased at post-education (372 calories per dollar). The total calories of food purchased decreased from a median of 16,356 at baseline to 15,093 at post-education (p=0.03); however, the total calories from beverages purchased did not decrease (p=0.3). For families that saw a reduction in calories (Table 2), each of the following medians moved in the desired direction, but were just short of a significant change (p=0.06 for all): median grams of carbohydrates per dollar purchased (baseline: 66, post-education: 45); percentage of calories from protein (baseline: 14%, post-education: 18%); and calories from processed food (baseline: 11,000, post-education: 7845).
      Table 2Nutritional value of food items purchased at baseline compared to post-education,
      For those that reduced the total calories purchased from baseline to post-education (n=13) versus those that did not reduce the total calories purchased (n=4)
      median (IQR) unless otherwise noted
      Reduced calories, n=13Did not reduce calories, n=4p
      For difference in baseline values for reduced versus not reduced
      BaselinePost-educationpBaselinePost-educationp
      Calories22,213 (18,846–30,997)15,852 (13,433–16,874)<0.0113,990 (13,070–17,114)19,336 (15,768–25,339)0.130.03
      Calories per dollar444 (377–620)317 (269–337)<0.01280 (262–342)387 (316–507)0.130.03
      Fat (grams)556 (378–792)452 (281–631)0.09333 (323–693)587 (449–1,087)0.130.54
      Fat (grams) per dollar11 (8–16)9 (6–13)0.087 (7–14)12 (9–22)0.130.54
      Calories from fat (%)19 (13–29)24 (18–26)0.8928 (23–40)33 (21–50)0.630.35
      Calories from cholesterol (%)67 (48–76)58.3 (53–65)0.3753 (37–60)49 (34–62)0.880.18
      Carbohydrates (grams)3,831 (2,948–5,057)2,243 (1,899–2,791)<0.011,549 (1,158–2,009)1,929 (1,521–3,570)0.130.02
      Carbohydrates (grams) per dollar77 (59–101)45 (38–56)<0.0131 (23–40)39 (31–71)0.250.03
      Fiber (grams)210 (146–286)185 (160–278)0.30129 (68–166)224 (169–266)0.130.10
      Protein (grams)694 (645–929)769 (690–929)0.84600 (521–1,025)845 (738–1,033)0.630.47
      Protein (grams) per dollar14 (13–19)15 (13–16)0.6712 (11–21)17 (15–21)0.750.47
      Calories from protein (%)12 (9–14)19 (14–22)0.0119 (18–24)17 (16–19)0.500.04
      Calories from food20,191 (15,760–29,997)14,623 (12,490–16,479)<0.0113,076 (11,706–15,037)18,199 (13,827–24,474)0.130.03
      Calories from processed foods14,106 (9,900–18,215)7,498 (4,124–11,328)<0.015,003 (2,789–5,730)8,812 (4,799–13,782)0.250.01
      Calories from processed foods (%)64 (48–82)49 (27–66)0.0733 (19–38)56 (29–70)0.380.02
      Calories from sugary beverages1,748 (1,000–3,962)1,684 (0–3,099)0.271,365 (451–2,991)865 (403–2,403)0.750.58
      Calories from sugary beverages (%)8 (3–14)9 (0–15)0.8410 (4–17)3 (1–14)0.750.91
      Note: Boldface indicates significance (p<0.05).
      IQR, interquartile range
      a For those that reduced the total calories purchased from baseline to post-education (n=13) versus those that did not reduce the total calories purchased (n=4)
      b For difference in baseline values for reduced versus not reduced
      There were no changes among families that did not reduce their calories of purchased food from baseline to post-education. To account for differences in family size, food composition variables were also analyzed per family member in the household (e.g., number of calories purchased per family member). There was no difference in findings when analyzed per family member versus for the total household (not shown).
      Thirteen of the 17 families (76%) reduced the number of calories purchased after the educational intervention from a median baseline of 22,213 calories to a median, post-education value of 15,852. In general, families that reduced their number of calories with the intervention were more likely to begin at baseline with food choices of poorer nutritional value than those families that did not see a change from baseline to post-education. Changes included higher median baseline values of total calories per purchase (p=0.04); calories per dollar (p=0.04); calories from carbohydrates (p=0.03); carbohydrates per dollar (p=0.03); calories from processed food (p=0.008); and percentage of calories from processed food (p=0.03), as well as a smaller percentage of calories from protein (p=0.04).
      These 13 families not only reduced the number of calories purchased but also decreased the number of calories from carbohydrates (median baseline: 3831, median post-education: 2243; p=0.006); carbohydrates per dollar (median baseline: 77, median post-education: 45; p=0.006); and calories from processed food (median baseline: 14,106, median post-education: 7498; p=0.003). Further, they decreased the total calories purchased from food (p=0.0005), but the total calories purchased from beverages was not reduced post-education. Each family also increased the percentage of calories from protein (median baseline: 12%, median post-education: 19%; p=0.01). These findings were similar when analyses were run for nutritional values per family member to account for family size.

      Discussion

      This study demonstrated that focusing on food shopping practices is an important area within nutrition education among low-income, Spanish-speaking individuals.
      • Hersey J.
      • Anliker J.
      • Miller C.
      • et al.
      Food shopping practices are associated with dietary quality in low-income households.
      The current healthcare system could benefit from family- and community-based interventions
      • Millán-Ferro A.
      • Caballero A.E.
      Cultural approaches to diabetes self-management programs for the Latino Community.
      to help the most vulnerable populations at high risk for obesity, type 2 diabetes, and cardiovascular disease. Proper nutrition, particularly in young populations, is a key factor to be addressed.
      • Malik V.S.
      • Fung T.T.
      • van Dam R.M.
      • Rimm E.B.
      • Rosner B.
      • Hu F.B.
      Dietary patterns during adolescence at risk of type 2 diabetes in middle aged women.
      Findings from the present study showed that there are ways to encourage low-income, Latino families to purchase healthier foods. Findings also challenged arguments suggesting that such an approach is not possible because of the high cost of healthier foods.
      • Drenowski A.
      • Darmon N.
      The economics of obesity: dietary energy density and energy cost.
      The educational strategies used in this study to address the economics of food choice
      • Philipson T.
      The world-wide growth in obesity: an economic research agenda.
      suggested that using easy-to-understand nutrition information leads to reduced purchasing of processed foods.
      Although the fact that energy-dense foods cost less seems to be an important driver in the consumption of these foods, and even though this study did not measure food consumption, the findings suggest the role that other factors may play in low-income individuals' ability to adopt a healthier diet. One of these factors is education focused on food purchasing behavior, which could attenuate the impact of economic and environmental forces. The use of patient-activation strategies helped change food shopping practices: participants, as a group, identified healthier choices that were readily available within their community food stores. The current results also echoed findings indicating that individuals can increase the overall nutritional value of purchased food without increasing their food budgets.
      • Katz D.L.
      • Doughty K.
      • Njike V.
      • et al.
      A cost comparison of more and less nutritious food choices in U.S. supermarkets.

      Limitations

      The study presents some limitations to be overcome in future studies. For example, although an objective assessment of food purchasing was conducted, no data were collected on other eating behaviors such as eating out and the nutritional value of those meals. Also, data were collected using a nonvalidated interview schedule, and the study used a quasi-experimental design. Finally, the fact that transportation was provided both at baseline and at the end of the study limited the influence of this potentially confounding factor.

      Conclusion

      The ethnographic nature of the current study helped to uncover and better illuminate factors involved in the decision-making processes that some low-income, Latino consumers follow when purchasing food for their families (i.e., price, lack of knowledge about the nutritional value of certain foods, family members' preferences for certain foods, emotional ties to children, and advertisement strategies). It was evident that to address obesity prevention, it is important to understand the decision-making processes of individuals when purchasing food. At a micro-level, the qualitative data contain specific messages that could help inform policymakers about important policy areas, such as pricing strategies to promote the purchase of healthier foods.
      Quantitative findings and participant observations also suggest the need to develop tailored social-marketing messages and other communication strategies to promote healthy food purchasing and consumption practices among Latinos. Specifically, there is a need to deliver messages that are developed with an understanding of the challenges that low-income families face when shopping for healthier food on small budgets. Also, food literacy could be improved with the use of visual, multimedia materials, such as photographs or videos. Feedback generated from the nutrition analysis of food shopping receipts also proved helpful.
      Publication of this article was supported by the Robert Wood Johnson Foundation.
      This study was funded by the Robert Wood Johnson Foundation through its national program, Salud America! The RWJF Research Network to Prevent Obesity among Latino Children (www.salud-america.org). Salud America!, led by the Institute for Health Promotion Research at The University of Texas Health Science Center at San Antonio, Texas, unites Latino researchers and advocates seeking environmental and policy solutions to the epidemic.
      No financial disclosures were reported by the authors of this paper.

      References

        • Narayan K.M.
        • Boyle J.P.
        • Thompson T.J.
        • et al.
        Lifetime risk for diabetes mellitus in the U.S..
        J Am Med Assoc. 2003; 290: 1884-1890
        • Ogden C.L.
        • Carroll M.D.
        • Curtin L.R.
        • Lamb M.M.
        • Flegal K.M.
        Prevalence of high body mass index in U.S. children and adolescents, 2007–2008.
        J Am Med Assoc. 2010; 303: 242-249
        • Caballero A.E.
        • Bousquet-Santos K.
        • Robles-Osorio L.
        • et al.
        Overweight Latino children and adolescents have marked endothelial dysfunction and subclinical vascular inflammation in association with excess body fat and insulin resistance.
        Diabetes Care. 2008; 31: 576-582
        • Caballero A.E.
        Diabetes in Hispanics/Latinos: challenges and opportunities.
        Curr Opin Endocrinol Diabetes Obes. 2007; 14: 151-155
        • Perrin J.M.
        • Bloom S.R.
        • Gortmaker S.L.
        The increase of childhood chronic conditions in the U.S..
        J Am Med Assoc. 2007; 297: 2755-2759
        • Drewnoski A.
        Obesity and the food environment: dietary energy density and diet costs.
        Am J Prev Med. 2004; 27: 154S-162S
        • Drenowski A.
        • Darmon N.
        The economics of obesity: dietary energy density and energy cost.
        Am J Clin Nutr. 2005; 82: 265S-273S
        • DeNavas-Walt C.
        • Proctor B.D.
        • Smith J.C.
        U.S. Census Bureau, current population reports, P60-239, income, poverty, and health insurance coverage in the U.S.: 2010.
        U.S. Government Printing Office, Washington DC2011
        • DHHS, Health Resources and Services Administration, Maternal and Child Health Bureau
        Child health USA 2010.
        DHHS, Rockville MD2010
        • Hofferth S.L.
        • Curtin S.C.
        Poverty, food programs, and childhood obesity.
        J Policy Manage. 2005; 24: 703-726
        • Rolls B.
        • Barnett R.A.
        Volumetrics.
        HarperCollins, New York2000
        • Liu S.
        Intake of refined carbohydrates and whole grain foods in relation to risk of type 2 diabetes mellitus and coronary heart disease.
        J Am Coll Nutr. 2002; 21: 298-306
        • WHO, Food and Agriculture Organization
        Report of the joint WHO/FAO expert consultation on diet, nutrition, and the prevention of chronic diseases.
        WHO, Geneva2003 (WHO Tech Report 916)
        • Wang Y.
        • Beydoun M.A.
        The obesity epidemic in the U.S.—gender, age, socioeconomic, racial/ethnic, and demographic characteristics: a systematic review and meta-regression analysis.
        Epidemiol Rev. 2007; 29: 6-28
        • Hersey J.
        • Anliker J.
        • Miller C.
        • et al.
        Food shopping practices are associated with dietary quality in low-income households.
        J Nutr Educ. 2001; 33: 16S-26S
        • Powell L.M.
        • Chaloupka F.J.
        Food prices and obesity: evidence and policy implications for taxes and subsidies.
        Milbank Q. 2009; 7: 229-257
        • Powell L.M.
        • Zhao Z.
        • Wang Y.
        Food prices and fruit and vegetable consumption among young American adults.
        Health Place. 2009; 15: 1064-1070
        • U.S. Department of Labor, Bureau of Labor Statistics
        Consumer expenditure survey.
        BLS, Washington DC2004-05
        • U.S. Department of Agriculture, Agricultural Research Service
        2011. USDA national nutrient database for standard reference, release 24. Nutrient Data Laboratory.
        • Bandura A.
        Social learning theory.
        General Learning Press, New York1977
        • Cortés D.E.
        • Mulvaney-Day N.
        • Fortuna L.
        • et al.
        Patient/provider communication: understanding the role of patient activation for Latinos in mental health treatment.
        Health Educ Behav. 2009; 36: 138-154
        • Blisard N.
        • Stewart H.
        • Joliffe D.
        Low-income households' expenditures on fruits and vegetables. U.S. Department of Agriculture, Agricultural Economic Report No. AER833, May 2004.
        • Glanz K.
        • Basil M.
        • Maibach E.
        • Goldberg J.
        • Snyder D.
        Why Americans eat what they do: taste, nutrition, cost, convenience, and weight control concerns as influences on food consumption.
        J Am Diet Assoc. 1998; 98: 1118-1126
        • The Latino Diabetes Initiative at Joslin Diabetes Center
        • Millán-Ferro A.
        • Caballero A.E.
        Cultural approaches to diabetes self-management programs for the Latino Community.
        Curr Diabetes Rep. 2007; 7: 391-397
        • Malik V.S.
        • Fung T.T.
        • van Dam R.M.
        • Rimm E.B.
        • Rosner B.
        • Hu F.B.
        Dietary patterns during adolescence at risk of type 2 diabetes in middle aged women.
        Diabetes Care. 2012; 35: 8-12
        • Philipson T.
        The world-wide growth in obesity: an economic research agenda.
        Health Econ. 2001; 10: 1-7
        • Katz D.L.
        • Doughty K.
        • Njike V.
        • et al.
        A cost comparison of more and less nutritious food choices in U.S. supermarkets.
        Public Health Nutr. 2011; 14: 1693-1699