Grummon consumption. Grummon et al reported thatGrummon consumption. Grummon et al reported that

et al focused on both food and beverage purchases of SNAP participants and non-participants,
while Park et al focused exclusively on sugar-sweetened soft drink consumption.
Grummon et al reported that in unadjusted analyses, households purchased a mean
of 1,400–1,600 kcal person, and that most households, regardless of SNAP or
income status, purchased less healthful foods and nutrients overall. Adversely,
before matching for Park et al, it was reported that SNAP recipients
significantly differed from non-recipients in terms of all covariates at the
significance level of 0.05. Continuing with unadjusted analyses, Grummon et al
found that all households purchased a mean of 51–89 kcal per person for sugar
sweetened beverages, 453–476 kcal per person of junk foods, total sodium purchases
were high at 2,400–2,700 mg per person, and mean store purchases of total
saturated fat were 23–27 g per person compared to the recommended daily
allowance of 22 g/day for a 2,000 calorie diet. Furthermore, before matching, Park
et al reported that non-recipients (n=807) consumed sugar-sweetened soft drinks
often. 209 (25%) individuals consumed SSD 1-3 times per month, 243 (30.1%) individuals
consumed SSD 1-6 times per month, and 355 (44.0%) individuals consumed SSD ? 1 time per day (Park et al., 2017). On the other
hand, there was a significant difference between SNAP recipients (n=393). 71
(18.1%) individuals consumed SSD 1-3 times per week, 92 (23.4%) individuals
consumed SSD 1-6 times per month, and 230 (58.5%) consumed SSD ? 1 time per day (Park et al., 2017). The total P
value was <0.01.               In multivariate adjusted analyses for Grummon et al, there were purchase similarities and differences between SNAP participants and nonparticipants. For several food groups, nonparticipating households purchased higher amounts of healthful foods and lower amounts of unhealthful foods than SNAP participating households did (Grummon et al., 2017). Similarly, after adjustment for Park et al, it was concluded that SNAP recipients were more likely to report higher levels of SSD consumption, compared with non-recipients. Grummon et al ultimately found that income-eligible nonparticipating households purchased significantly greater calories from fruit than SNAP households did (mean difference: +4.35 kcal per person (P<0.0001) as did higher-income nonparticipants (mean difference: +4.46 kcal per person (P<0.0001). Income-eligible nonparticipants purchased fewer calories from processed meat (mean difference: 28.41 (P<0.0001) and from sweeteners and toppings (mean difference: 210.81 kcal per person (P = 0.0001) than SNAP participants (Grummon et al., 2017).     However, the opposite pattern occurred for salty snacks. All nonparticipants purchased significantly more calories from salty snacks than SNAP participants did (P<0.0001). Higher-income nonparticipants purchased slightly more calories from non-starchy vegetables than SNAP participants did (mean difference: +1.58 kcal per person (P<0.0001) but slightly fewer calories from starchy vegetables (P<0.0001) (Grummon et al., 2017) There were no significant differences across groups in purchases of total vegetables, legumes, nuts, other dairy, desserts, sweet snacks, candy and gum, or junk foods (Grummon et al., 2017).   In regards to sugar sweetened beverages for Grummon et al, and Park et al, both found that SNAP participants purchased and consumed significantly more then non-participants. Grummon et al reported a mean difference of 214.98 kcal per person (P<0.0001) for income eligible non-participants, and a mean difference of 220.52 kcal per person (P<0.0001) for higher-income nonparticipants. Park et al reported that after adjustment, out of the non-recipients (n=393), 78 (19.9%) individuals consumed sugar-sweetened soft drinks 1-3 times per month, 125 (31.8%) individuals consumed SSD 1-6 times per week, and 190 (48.3%) individuals consumed SSD ? 1 time per day. For SNAP recipients (n=393) , it was reported that 71 (18.1%) individuals consumed SSD 1-3 times per month, 92 (23.4%) individuals consumed SSD 1-6 times per month, and 230 (58.5%) consumed SSD ? 1 time per day. The total P value was 0.02 (Park et al., 2017).   Dissimilarly from Park et al, Grummon et al also took nutrient groups into consideration. Multiple significant differences in nutrient group purchases were found (Grummon et al., 2017). All nonparticipating households purchased fewer total calories than SNAP participants (income eligible non-participants mean difference: 263.08 kcal per person (P = 0.0002) and higher income non-participants mean difference: 269.97 kcal per person (P<0.0001). Both groups of nonparticipants also purchased fewer grams of sugar than did current SNAP participants (mean differences: 26.73 g per person for income-eligible nonparticipants (P<0.0001); 28.08 g per person for higher income nonparticipants (P<0.0001) and fewer milligrams of sodium (mean difference:2170.34 mg per person for income eligible nonparticipants, P<0.0001; 2194.80 mg per person for higher-income nonparticipants, P<0.0001). Lastly, income eligible nonparticipant households purchased more grams of fiber than did current SNAP households (mean difference: +0.52; P = 0.0002) as did higher-income nonparticipants (mean difference: +0.52; P<0.0001) (Grummon et al., 2017).               In similar way to Grummon et al, and Park et al, the two other studies that are focused on in this section also examine shopping habits on individuals on food assistance programs. However, Andreyeva et al and Fernandes et al, looked solely at food participant recipients and their beverage purchases and consumption.  The first study found a positive association between SNAP and WIC participants and unhealthy beverage choices (Andreyeva et al., 2012). On the other hand, the second study did not find an association between SNAP participants and unhealthy beverage choice (Fernandes et al., 2012).  Andreyeva et al examined the beverage purchases of carbonated soft drinks, bottled water, 100& fruit juice, fruit drinks, energy drinks, sports drinks, ready-to-drink tea, ready-to-drink coffee, flavored water, powdered drinks, sugar-sweetened beverages, diet beverages, unsweetened beverages, and less-sweetened beverages. Moreover, Fernandes et al had a much more narrow focus, looking only at consumption of soft drinks, 100% fruit juice, and milk. Andreyeva et al reported that               Andreyeva et al reported that on average, SNAP households purchased 689 oz. of refreshment beverages per month, including 399 oz. of sugar-sweetened beverages or 58% of beverage volume. WIC only households purchased less of all refreshment beverages (352 oz.) and fewer sugar-sweetened beverages: 169 oz. or 48% of beverage volume (Andreyeva et al., 2012). Both SNAP and WIC groups similarly purchased carbonated soft drinks more than any other beverage. Purchases of bottled water, sports drinks, ready-to-drink tea/coffee, and flavored water were identical between the two groups (Andreyeva et al., 2012). However, almost half of fruit-based beverages purchased by SNAP households were less nutritious fruit drinks, while WIC only households purchased 100% juice (Andreyeva et al., 2012). SNAP households also had higher monthly spending on refreshment beverages than WIC only households ($17 vs. $9) (Andreyeva et al., 2012). On the contrary, Fernandes et al found that milk was the most frequently consumed beverage in the 8th grade (8.0 times per week). Soft drinks (6.0 times per week in 8th grade) and fruit juice (5.6 times per week in 8th grade) were consumed less frequently (Fernandes et al., 2012). SNAP participation was associated with 0.6 fewer episodes of soft drink consumption, but was not found to be statistically significant (Fernandes et al., 2012). Similarly, SNAP participation was also associated with a greater frequency of fruit juice consumption but was not found to be statistically significant (Fernandes et al., 2012).               In terms of mean ounces purchased per month of all beverages, Andreyeva et al found that for WIC households, 33% of purchases was carbonated soft drinks, 15% was bottled water, 18% was 100% fruit juice, 10% was fruit drinks, 0% was energy drinks, 7% was sports drinks, 5% was ready-to-drink-tea, 0% was ready-to-drink coffee, 2% was flavored water, 10% was powdered drinks, 48% was sugar-sweetened beverages, 16% was diet beverages, 34% was unsweetened beverages, and 2% was less-sweetened beverages. For SNAP households, 34% of purchases was carbonated soft drinks, 14% was bottled water, 15% was 100% fruit juice, 13% was fruit drinks, 0% was energy drinks, 6% was sports drinks, 4% was ready-to-drink-tea, 0% was ready-to-drink coffee, 2% was flavored water, 12% was powdered drinks, 58% was sugar-sweetened beverages, 12% was diet beverages, 29% was unsweetened beverages, and 1% was less-sweetened beverages (Andreyeva et al., 2012). There were no P values given. Adversely, Fernandes et al compared participation on frequency of soft drink, 100% fruit juice, and milk consumption in one week between children in grades 5 and 8 whose families participate in the SNAP program. It was concluded that for soft drinks, participants had a -0.58 frequency of consumption (P value = 0.29), for 100% fruit juice, the participants had a 0.64 frequency of consumption (P= 0.48), and a 0.51 frequency of consumption for milk (P value = 0.53).   Food Security Status and Food Purchasing & Consumption Habits Eleven of the fifteen studies that were used to write this literature review examined the food purchasing and consumption habits of food insecure and low-income individuals. Six of the studies analyzed were cross-sectional studies (Turrell et al., 2006, Lombe et al., 2016, Smith, et al., 2013, Ricciuto et al., 2006, Dammann et al., 2010, Dachner et al., 2010), four of the studies were qualitative (Evans et al., 2015, Wigg et al., 2008, Darko et al., 2013, Cortés et al., 2013), and one study was quasi experimental (Vaughan et al., 2017). Five of the cross-sectional studies had subject counts ranging from 165-2,71, while the sixth focused on 10,924 households studies (Turrell et al., 2006, Lombe et al., 2016, Smith, et al., 2013, Ricciuto et al., 2006, Dammann et al., 2010, Dachner et al., 2010). Three of the qualitative studies had a subject count of 72-148 individuals, with the fourth qualitative study using 20 families as their subject participants (Evans et al., 2015, Wigg et al., 2008, Darko et al., 2013, Cortés et al., 2013). The quasi-experimental study had a subject number of 1,372 households (Vaughan et al., 2017). All of the eleven studies focused on in the section of the results found significant positive relationships between food security status and food purchasing and consumption habits.     Two of the eleven studies paid particular attention to the food purchasing habits of low-income women (Wigg et al., 2008, Dammann et al., 2010). Dammann et al found that rates of less healthy food group purchases among the sample were higher compared to healthy food group purchases. The frequency and percentage of purchases made by the participants in one month reported the following: (general food group women spent the most money on) 52.1% on beef/pork (P=0.000), 23.7% on chicken/turkey (P=0.000), 5.6% on sweet snack (P=0.031), 3.6% on pop (P=0.129), 3.4% on fruits (P=0.079), 2.9% on vegetables ((P=0.393), 2.5% on dairy (P=0.020), 1.8% on eggs/beans/nuts (P=0.349), 1.6% on Kool-Aid/fruit drinks (P=0.109), 1.4% on grains (P=0.898), and 1.4% on salty snacks (P=0.198) (Dammann et al., 2010). Meanwhile, Wigg et al focused on similar food groups and reported 80.4% on meat, poultry, fish and eggs, 69.6% on fruits, 64.1% on cereal, bakery, bread, rice, and pasta, 59.8% on vegetables, 56.5% on dairy products, 35.9% on Kool-Aid, juice, fruit punch, and lemonade, 35.9% on salty snacks, 29.3% on pop, 27.2% on sweets, and 12.0% on fats (Wigg et al., 2008). P values were not given. Through out the subject interviews, the women explained that meat was the basis for most grocery shopping and meal planning simply because it was filling, although they only had the money to spend on high fat, cheaper cuts of meat (i.e. ground beef, hot dogs) (Wigg et al., 2008). Starches (e.g. rice, noodles) were also common inexpensive purchases and vegetables were frequently mentioned as part of the main meal, yet foods from this group, except for potatoes, failed to make up a significant portion of the participants' budgets (Wigg et al., 2008).               Three of the eleven studies had a strong focus on the relationship between household income and food cost concern and how this effected their food purchasing choices (Turrell et al., 2006, Darko et al., 2013, Lombe et al., 2016). Turrell et al reported a strong association between household income and food-cost concern, with respondents from low-income households being significantly more likely to report that food costs created a barrier for purchasing healthy food. In terms of food cost concern for Turrell et al, it was shown that those whose household income was $36,400-51,999 had a confidence interval of -8.29, -2.24 (scale measuring food cost concern ranged from 0-100), those who had a household income of $20,800-36,399 had a confidence interval of -9.75, -4.01, lastly, those who had a household income of ? $20,799 had a confidence interval of -12.48, -5.90 (Turrell et al., 2006). Food-cost concern was independently related with food purchase: a one-unit increase on the concern index was associated with an average decrease of -0.11 (95% CI 20.18, 20.04) units on the purchasing measure (P=0.0015) (Turrell et al., 2006).   Continuing on, Lombe et al looked at effects of food insecurity and food choices on health outcomes. It was reported that household insecurity compared with diet related health risk had a CI of .92-1.64. SNAP participation compared with diet related health risk had a CI of .92-1.63. Informal food supporters compared with diet related health risks had a CI of .88-1.90 (Turrell., et al 2006). In regards to food choices compared with diet related health risks, food pyramid awareness had a CI of .59-.95, MyPyramid use had a CI of 1.13-2.40, and food insecurity + awareness had a CI of 0.26. Finally, Darko et al had similar findings with Turrell et al and Lombe et al. It was reported that most participants agreed that economics played a substantial role on shopping behaviors, and that these behaviors were often influenced by the timing of receiving employment wages or SNAP/WIC benefits (Darko et al., 2013). Throughout a 1-month period, some participants stated that their shopping and eating habits changed because they bought a greater variety of food when funds were available. However, toward the end of the month or when economic resources were running low, they relied more on carbohydrate-rich, canned, and packaged food (Darko et al., 2013). In terms of food cost, most participants agreed that food prices affected their shopping habits and that they relied on shopping at stores that they thought had the lowest prices (Darko et al., 2013). These results were solely based on focus groups.   Two studies focused on how access to healthy foods played a major role in purchasing habits. These qualitative and quasi-experimental studies used focus groups, household interviews and 24 diet recalls to assess their participants (Vaughan et al., 2017, Evans et al., 2015). Evans et al reported that despite the high level of knowledge about the components of a healthy diet, participants expressed multiple barriers in order to consume healthful foods. The most common influences included high cost of healthful foods, inadequate geographical access to healthful food, and poor quality of available healthful food (Evans et al., 2015). Participants living in areas with no supermarket consistently stated that the solution to increase access to more  healthful foods was to build a conveniently located supermarket offering a wide variety of quality items in a convenient location. For those participants already living close to a supermarket, solutions focused on increasing the quality of food available in the store and on improving the overall quality of the store (Evans et al., 2015). When participants were specifically asked about their use of convenience stores for food purchasing, the overall sentiment was very negative. Convenience stores were typically perceived to be too expensive (Evans et al., 2015). Convenience stores also were perceived as having limited and very low-quality food products, especially produce (Evans et al., 2015).   Similarly, Vaughan et al reported that nearly all store types emphasized unhealthy over healthy foods. In regards to frequency of food shopping at different store types participants rated each type on a scale from 1 (never) to 4 (often). Full service supermarkets had a mean number of 3.6, supercenters had a mean number of 2.5, dollar stores had a mean number of 2.5, fruit and vegetables stores had a mean number of 2.1, discount grocery stores had a mean number of 2.1, wholesale clubs had a mean number of 1.9, convenience stores had a mean number of 1.8, neighborhood stores had a mean number of 1.8, and specialty grocery stores had a mean number of 1.5 (Vaughan et al., 2017). Shopping more frequently at unhealthy and moderate food stores was associated with unhealthy diet (Vaughan et al., 2017). Shopping at convenience stores was significantly associated with greater consumption of added sugars; buying food more often at neighborhood stores predicted significantly greater intake of SSBs and unhealthy fats, and buying food more often at supercenters was significantly associated with greater intake of unhealthy fats (Vaughan et al., 2017). Conversely, shopping more often at specialty grocery stores and fruit and vegetable stores was significantly associated with greater fruit and vegetable consumption (Vaughan et al., 2017).   The last four studies focused entirely on the relationship between specific food group purchases and food security status. Smith et al stated that households reporting low food security compared to households with moderate food security spent less money on food overall (t-test, P = 0.026) and on the food groups fruit and vegetables (Wilcoxon rank-sum test, P = 0.005) and cereals (Wilcoxon rank-sum test, P = 0.006). Households reporting low food security compared to moderate food security to spent less on milk (Wilcoxon rank-sum test, P = 0.056) (Smith et al., 2013) .For those with low food security, Smith et al reported that median amount spent on fruits and vegetables was $3.94 (P=0.004), $5.29 on meat, fish, and poultry (P=0.441), $3.77 on cereals and bread (P=0.008), $4.97 on snacks, cakes and biscuits (P=0.805), $5.33 on dairy, eggs and milk (P=0.113), $1.81 on beverages (P=0.485), $4.21 on other grocery food (P=0.228) and $1.69 on ready to eat food (P=0.123) (based of $36.51 spent total) (Smith et al., 2013). Similarly to Smith et al, Dachner et al reported that among respondents who reported usually purchasing a given indicator food, price was the most salient factor influencing decisions. The greater the severity of food insecurity, the more likely families were to report price and price as the sole factor in their purchasing decisions of each food group (Dachner et al., 2010).   In terms of  proportion of usual purchasers reporting price as the sole factor underlying purchasing decision by food-security status, Dachner at al found that 47.9% of  participants who were severely food insecure purchased margarine (P=0.0032), 33.7% purchased cereal (P=0.0001), 31.4% purchased fruit (P=<0.0001), 31/3% purchased vegetables (P=0.0002), and 51.9% purchased meat (P=<0.0001). For margarine, this simply meant purchasing "the cheapest," or  "whatever is on sale", but for perishable items like fruit, vegetables, and meat, the durability of an item and other considerations such as whether the product would get eaten or go to waste were factored into the cost of an item (Dachner et al, 2010). Some respondents reported buying frozen vegetables and meat because they were cheaper and lasted longer, and  many reported buying cheap cuts, especially chicken legs and wings and ground beef (Dachner et al, 2010).   Cortes et al focused on food purchases based off of total calories. It was reported that total fat calories purchased was 477(P=0.52), cholesterol calories was 62 (P=0.03), total carb calories was 3,298(P=0.08), total fiber calories was 182 (P=0.95), total protein calories was 689 (P=0.58), total calories from processed food was 16,356 (P=0.03), and total calories from sugary beverages was 1,784 (P=0.27) (Cortes et al., 2013). The median calories purchased by families at baseline were 20,191(median 404 calories per dollar) (Cortes et al., 2013). When compared to Ricciuto et al, similar results were found. The largest average share of income was allocated to the meat and alternatives food group (3.7%), which may be because of the high cost of these foods, as suggested by the high level of expenditure and low quantity purchased, relative to the rest of the food groups (Ricciuto et al., 2006). In regards to reported purchase and average weekly expenditure for the major food groups, Ricciuto found that the median expenditure on grain products was $7.07, $13.35 on vegetables and fruit, $9.20 on milk products, $17.93 on meat and alternatives, and $13.43 on other foods including oils, sugars, desserts, snacks and beverages (Ricciuto et al., 2006).