Eicher-Miller et al, 2009, in a randomized controlled trial (RCT) (quality rating, positive), studied 236 low-income women who were head of households [Experimental Group (EG)=137, Gontrol Group (CG)=82; 98.6% non-Hispanic white, 2.3% black; EG=52% with food stamps; CG=45% with food stamps] to determine the effect of the Food Stamp Nutrition Education (FSNE) in Indiana on participants' food insecurity and food insufficiency. The FSNE in Indiana operates independently of the Food Stamp Program and does not directly offer supplemental food or food assistance, but provides nutrition education in the form of educational lessons, free of charge, to low-income participants.
The FSNE assistants randomly assigned participants into the EG or CG. Clients arriving together were assigned to the same group to prevent knowledge of a difference in treatment. At time of assignment, participants' food insecurity was assessed with the six-item United States Household Food Security Scale. Food insufficiency was quantified with the United States Department of Agriculture Food Insufficiency Question. 40% of participants were classified as food-insecure.
EG participants received one FSNE lesson, then took a pre-test. They completed four additional lessons over a five-week period, then took a post-test.
CG participants received one FSNE lesson, then took a pre-test. After a five-week period during which no nutrition education was provided, a post-test was administered. Following administration of the post-test to the CG, the four additional FSNE lessons were provided to CG participants.
The FSNE lessons provided to EG participants targeted food insecurity and nutrition. EG participants spent time with an FSNE assistant, a peer educator and a paraprofessional trained in nutrition education. The FSNE assistant modeled food preparation, healthful food selection and cooking skills, often in the client's home. Lessons included interactive presentation of the ideas discussed (e.g., preparing a meal, learning how to read food labels, visual examples of more and less healthful foods, practical tips on budgeting and thrifty food purchasing). The lessons covered five topical domains: MyPyramid, food groups, food safety, shopping behaviors and resource management and wellness. Lessons were tailored to each client based on household composition (i.e., young participant heading household with children) and age of participant (a single elderly) by choosing one lesson from the one to four lesson choices provided for each of the five topical domains.
The investigators found that food security in the EG compared to the CG was significantly improved (R2=0.3073, β -0.37±0.17, P=0.03) after adjusting for pre-test score and employment. Food insufficiency in the EG compared with the CG was significantly improved (R2=0.4384, β -0.08±0.04, P=0.04) after adjusting for pre-test score. There were no significant (NS) differences between CG and EG in self-reported characteristics of CG and EG, in initial food security or food insufficiency status and in the proportion of individuals classified as food-secure, food-insecure without hunger or food-insecure with hunger.
The investigators concluded that FSNE was successful in improving participants' food insecurity and food insufficiency that employed FSNE participants or those with employed partners were able to improve household food security more than households without employment and that nutrition education is an effective intervention for food insecurity.
Study results may not apply to a more ethnically- or racially-diverse population.
Goodner et al 2000, in a cross-sectional study (quality rating, neutral), studied a low-income population to determine if the provision of food stamps without nutrition education improves food intake patterns. A total of 208 persons were included in the sample [151 food-stamp (FS) recipients, 57 not food-stamp (NFS) recipients; mean age 34±12 years; 33% white, 66% black, 1% other; no high school diploma, 60%; high school diploma, 26%; beyond high school, 13%; 84% had income under $10,000 per year; BMI, 31.5±9.7]. FS recipients were significantly older (P<0.01), were less likely to have completed high school (P<0.01) and were more likely to have incomes less than $10,000 per year (P<0.001) than NFS participants. However, NFS participants met food stamp eligibility requirements. All participants were from South Carolina.
Data were collected at one time point from eligible participants and then data were compared between those who received food stamps and those who did not. Data were collected regarding 24-hour diet recalls, demographics, food behavior and knowledge data, anthropometrics, blood pressure and self-reported physical activity. Data were collected by Registered Dietitians (RDs) and by graduate students supervised by RDs.
The researchers found that mean total energy intakes for both groups fell below 100% (FS, 82%; NFS, 74%) of the RDA for women 25 years to 50 years of age. Food intake reported by both groups exceeded the RDA for vitamin C, thiamin, riboflavin and niacin. Mean intake of vitamin B12 was significantly higher among FS recipients than NFS recipients. Intakes of vitamins A, E and B6 and iron were below 100% of the RDA; mean intake of vitamin D was below two-thirds of the RDA for FS and NFS recipients. The NFS recipients had mean intakes less than two-thirds of the RDA for vitamin E and zinc. Mean intake of zinc was significantly lower in the NFS recipients than FS recipients (P<0.05). Intakes of calcium and folate were NS different between groups. However, when compared to the DRIs, intakes of calcium for FS and NFS were below two-thirds (53% and 59%, respectively) and intakes of folate were below 50% (43% and 49%, respectively) of the recommended value. Reported fruit and vegetable consumption (FS, 2.24; NFS, 2.92) for both groups was below the Healthy People 2000 objective of five fruits or vegetables per day. Mean scores for all participants on the brief nutrition knowledge pre-test were below 50% (40±24%). The FS recipients scored significantly lower (P<0.01) than the NFS recipients (37±24% and 47±25%, respectively) on the behavior and nutrition knowledge pre-test. Analyses of food intake data adapted from Food Guide Pyramid categories indicated inadequate intakes for both groups for recommended numbers of servings in all categories except meats.
Analysis of food intake data adapted from Food Guide Pyramid categories indicated inadequate intake for both groups for recommended number of servings in all categories except meats.
The researchers concluded that although food stamp benefits increased food purchasing power, food stamps did not appear to ensure consumption of nutritionally adequate diets. In addition, the significantly lower income and educational levels of FS recipients may have contributed to their less than optimal dietary habits. Results of the present study suggest that low-income individuals in South Carolina would benefit from nutrition education.
The study groups differed in age, education and income levels.
Kennedy et al, 2009, in an RCT (quality rating, positive), studied 40 African-American women from East Baton Rouge Parish in Louisiana to test the feasibility of the Rolling Store (RS), an innovative food delivery intervention, along with a nutrition education program to increase the consumption of healthy foods to prevent weight gain in African-American women.
Subjects were randomized to either the Intervention Group (IG, N=20) or the Control Group (CG, N=20). Subjects' ages ranged from 22 years to 66 years and average BMI was defined as Obesity Class I.
The intervention was provided by trained peer educators and trained RS operators. The CG met with the peer educator once each month for six months to measure blood pressure (BP) and weight. At the time of those visits, nutrition and physical activity information were given to each participant in the Control Group to take home and read.
The IG received the same nutrition and physical activity information that were given to the Control Group; however the lessons were delivered in a group class-setting that included cooking demonstrations. All participants were provided fresh fruits and vegetables each week for 24 weeks from the Rolling Store, as well as recipes for healthy preparation techniques utilizing the fruits and vegetables. IG participants were asked to keep a seven-day food and exercise diary during one week of each month for six months. The diaries were used to provide feedback and guidance based on current recommendations to prevent weight gain.
The Rolling Store was a truck with a detachable camper shell with banners reading "Rolling with Healthy Choices." The Rolling Store was parked outside the community center from 2 p.m. to 6 p.m. on the same day of each week and IG participants received nine to 14 choices of fruits and vegetables. Fruits and vegetables varied each week and each week participants received one or more new choices. The principal investigator and RS operator were responsible for stocking the store each week.
The investigators noted a study retention rate of 93%. Participants in the IG lost a mean weight of 2.0kg, while participants in the CG gained a mean weight of 1.1kg at six months. Overall participants showed a mean decrease in weight of 0.4kg [standard deviation (SD), 3.0kg], but the IG lost significantly more weight and had decreased BMI at six months. In the IG, the average number of servings consumed per day of fruits, fruit juice and vegetables significantly increased at six months. Waist circumference and systolic blood pressure were lowered in the IG, although the difference did not reach significance.
Participants in the study were provided with incentives as well as $10 worth of free fruits and vegetables each week. The community voluntarily participated and purchased fruits and vegetables from the Rolling Store.
The investigators concluded that the RS, at least on the small scale on which it was implemented, is a feasible approach to producing weight loss and improvements in healthy eating when combined with an educational program in a small community center. In order for communities to undertake such an effort, leaders would be needed to organize the cooperative buying at less than retail value, provide a distribution system and manage finances.
The study was limited by a small sample size.
Pempeck et al 2009, in a cross-sectional study (quality rating, neutral), studied 30 children (15 female; mean age, nine years six months ±11 months; 100% African-American; 100% resided in low-income areas of Washington, DC) to determine whether playing an advergame that promoted less healthy foods such as those typically seen in food advertisements directed at children, could affect children's behavior, as measured by what children selected and ate as a snack and to determine whether playing an advergame could promote healthier food choices and consumption patterns. Participants were blinded to the purpose of the trial.
The children were randomized to one of three groups:
- Healthier advergame
- Less healthy advergame
- Control group.
The two intervention groups received instruction in how to play their respective games, played the games twice, filled out a questionnaire and were given an opportunity to choose a snack and beverage following the game. The snacks and beverages offered were the same as those used in the games. Children in the Control Group were given an opportunity to choose a snack and beverage, then played the healthier advergame and filled out a questionnaire.
Nutritious snacks included orange juice, apples, bananas and baby carrots. Less healthy snacks included soda, potato chips, chocolate chip cookies and chocolate candy bars. The game was electronic, with points awarded to a Pac-Man character, according to whether the child was playing the healthier or less healthy game.
The investigators found a main effect of treatment on snack selections (P<0.007). Children in the healthier advergame treatment selected and ate a greater number of healthier snacks than those in the less healthy game condition [1.40 (0.24) vs. 0.20 (0.24)]. The number of healthier snacks selected and eaten by children in the Control Group fell between the two treatment groups and did not significantly differ from either group [0.90 (0.24)]. When asked what Pac-Man wanted them to eat, 25 of 28 children answered correctly. When asked what Pac-Man wanted them to drink, 27 of 28 children answered correctly. There was no effect of the game by gender on snack selection, but girls were more likely than boys to visit food websites when using the internet (P<0.01). 27 of 28 children said they "really liked" the game. Participants played the game an average of nine minutes 32 seconds (SD, two minutes 22 seconds). The children reported weekly internet usage with the most popular activity being gaming.
The investigators concluded that with less than 10 minutes of exposure, the results revealed that participants selected and ate whatever snacks were being marketed by the advergame, healthy or not, and that advergames can be used to promote healthy food choices. 36% of the girls visited food websites, compared to 0% of the boys, indicating that girls may be more at risk for exposure to marketing practices that promote foods that can lead to obesity and poor health.
Inclusion and exclusion criteria were implied, not stated. No anthropometric data were provided. The comprehensiveness of this study was limited by the use of children from only one ethnic group.
Wiig et al 2009, in a cross-sectional study (quality rating, positive), examined grocery shopping behaviors and other potential factors that could influence food choice and food stamp usage among low-income women with children in their home using both qualitative and quantitative methods of data collection. A total of 92 women participated in the study (36.6±8.0 years; 51% African-American, 27% Native American, 13% Caucasian; 84% on the Food Stamp Program; 70% were unemployed; 66% had incomes under $10,000 per year; 49% had a GED or high school equivalent and the rest had some education beyond high school; approximately one-third were from a homeless shelter; mean BMI, 32.9±8.9kg/m2; residents of Minneapolis-St. Paul).
Participants were recruited mainly by posted flyers. Participants were assigned among 14 focus groups. The focus groups lasted 90 minutes and were audio taped. Questions examined personal, behavioral and environmental influences on grocery shopping and food choices and were framed in the context of Social Cognitive Theory constructs. Participants also provided demographic information and completed a written grocery shopping activity. The purpose of the activity was to transition women into talking about how they prioritize food purchases, given limited resources. Height and weight were measured following a standard protocol and used to calculate BMI. A cash stipend was given to participants to compensate for their time.
The investigators identified major themes from the focus groups:
- Store location, a critical factor since most women did not have their own car and relied on alternative forms of transportation
- Participants reported shopping at a variety of places including: retail grocery stores, meat markets, discount stores, wholesale stores and corner stores
- The frequency of shopping varied from daily to once per month, depending on the availability of transportation
- Most participants prioritized meat as their most important food purchase and allocated nearly 50% of their food money for meat. Meat was the basis for most grocery shopping and meal planning.
- Per the grocery shopping activity, particpants spent most of their money on high-fat cheaper cuts of meat, such as ground beef and hot dogs
- Other choices for food purchases were based on what the household needed and what items could be obtained through food assistance programs such as dairy from WIC and canned goods from local food shelves and food pantries
- Most women reported that their food stamp allotments lasted two to three weeks, depending on how they spent them and where they shopped
- Fresh fruits and vegetables were perceived as high cost
- Perishable items such as milk and dairy products were considered expensive and difficult to keep around, since family members quickly consumed them.
The investigators concluded that low-income women's grocery shopping was driven by their families' personal preferences and by their economic and environmental situations. Despite attempts to have nutritious meals, participants felt that they could not purchase ideal foods for a good diet. The participants' tendency to purchase high-fat foods like meat and less fruits and vegetables may explain the high incidence of obesity. Nutrition education that teaches food budgeting skills and meal preparation strategies involving less meat and more fruits and vegetables could be useful in assisting low-income families use their food dollars wisely and could potentially have healthful impacts.
The results may have limited generalizability to the population at large, because the data collection was focused on one state.