**Caregiver's Self-Confidence in Food Resource Management Is Associated with Lower Risk of Household Food Insecurity among SNAP-Ed-Eligible Head Start Families**

**Lamis Jomaa 1,2, Muzi Na 2, Sally G. Eagleton 2,3, Marwa Diab-El-Harake <sup>1</sup> and Jennifer S. Savage 2,3,\***


Received: 26 June 2020; Accepted: 29 July 2020; Published: 31 July 2020

**Abstract:** Food resource management (FRM) behaviors are key components within nutrition education programs designed to help food insecure households maximize their food dollars. However, little is known about the association between FRM self-confidence and financial practices with household food insecurity (HFI) among families with young children. Using a sample of SNAP-Ed-eligible Head Start families, this study examined associations between FRM self-confidence, FRM behaviors and financial practices by HFI. A needs assessment survey was conducted with caregivers of Head Start children (*n* = 365). HFI was measured using the US Household Food Security Survey Module. Chi-square and logistic regression analyses were conducted to examine if FRM self-confidence, FRM behaviors, and financial practices differed by HFI. Participants with high FRM self-confidence had lower odds of HFI (OR = 0.54, 95%CI: 0.33, 0.87), yet FRM behaviors, financial practices, and HFI were not related after adjusting for covariates. All FRM self-confidence questions significantly differed by HFI, whereas only one of six FRM behaviors and two of three financial practices differed by HFI (all *p*-values < 0.05). Promoting caregivers' self-confidence in FRM skills within nutrition education programs may be explored as a potential strategy to assist low-income households to stretch their food dollars in an attempt to address HFI.

**Keywords:** food resource management; food insecurity; self-confidence; nutrition education; financial practices; SNAP-Ed; Head Start; young children

#### **1. Introduction**

Household food insecurity (HFI), defined as "the inability to provide enough food for a healthy and active lifestyle for all household members [1]", remains a serious social and public health problem in the US [2]. Food insecurity is especially prevalent among low-income families with children. In 2018, 13.9% of American households with children were food insecure, and the prevalence of HFI reached 14.3% among households with children 6 years of age or younger [3]. Food insecurity is associated with a range of negative health outcomes among infants and young children, including poor physical health, increased risk of infections, micronutrient deficiencies [4,5], suboptimal sleep quality [6], adverse

behavioral, mental, and academic behaviors [5,7,8], as well as obesity and other chronic conditions during childhood and later in life [7,9].

Federal assistance programs that provide monetary benefits along with nutrition education to low-income households have been shown to alleviate HFI [10]. These nutrition education programs, such as the Supplemental Nutrition Assistance Program Education (SNAP-Ed) and the Expanded Food and Nutrition Education Program (EFNEP), provide participants with trainings on how to maximize the use of their food dollars while providing healthy foods to their families and children [11]. An integral component of these nutrition education programs is to teach individuals how to acquire food resource management (FRM) skills and behaviors defined as "the handling of all foods and the resources that may be used to acquire foods by an individual or family [12]." In addition, FRM trainings cover topics such as meal planning, shopping strategies, food selection, budgeting, food preparation, and cooking strategies to maximize nutrition under resource constraints [12]. Previous studies indicate that integrating FRM within nutrition education (e.g., food preparation tips, healthful food selection, and budgeting) improves the food security status of low-income households [10,13], including those with children [14,15].

Although food assistance programs, such as SNAP and SNAP-Ed, focus on behavioral change in FRM, less emphasis has been placed on assessing participants' self-efficacy and confidence in their FRM skills. Few studies, to date, have reported how nutrition education interventions targeting self-efficacy and confidence in FRM can improve food security [15,16]. Perceived self-efficacy represents a key construct in behavioral change theories, as it refers to an "individual's confidence in their ability to plan and follow through with a series of actions that will result in desired outcomes or achievements" [17]. Research studies examining the effect of self-efficacy on behavioral change related to nutrition, exercise, and weight loss [18], as well as the prevention of chronic diseases [19], have demonstrated the pivotal role that self-efficacy plays in improving health. Knowing that families experiencing food insecurity may face various challenges affecting their confidence in managing their budgets to maintain food sufficiency [20,21], it is integral to further examine the association between FRM self-confidence and HFI [16].

Food insecurity is linked to income [1]; however, food insecurity is not the outcome of income alone. Instead, it is influenced by a myriad of other demographic, environmental, and financial factors [22,23]. To further examine the determinants of HFI, a growing body of literature has been exploring the association between financial management skills and food insecurity [22,24]. It was previously suggested that good financial management practices may safeguard certain households from food insecurity, whereas those with less effective financial skills may be at increased risk of food insecurity [22,25]. To our knowledge, the associations between FRM, financial practices, and HFI have not been adequately explored in the literature, particularly among households with young children. To address this research gap, the present study aimed to first examine the associations between FRM self-confidence and FRM behaviors by HFI status using a sample of SNAP-Ed-eligible Head Start families. A secondary objective of the study was to explore the association between financial practices of caregivers and HFI status in the study sample. Head Start is a federally-funded program that serves just over 900,000 low-income preschool children in the US to optimize their health and nutrition. The Head Start program also provides balanced snacks and meals to children through the Child and Adult Care Food Program [26]. Although previous studies have shown that Head Start programs can help alleviate HFI and improve nutrition outcomes of children [27,28], none, to our knowledge, have examined the potential associations between caregiver's FRM self-confidence and behaviors by HFI. We hypothesized that (1) caregivers with higher self-confidence and better FRM skills would have lower risk of being food insecure; and (2) caregivers with good financial practices would report lower levels of food insecurity.

#### **2. Materials and Methods**

#### *2.1. Sampling and Recruitment*

Caregiver-child dyads in the present study were recruited from Head Start preschool classrooms in four rural counties in central Pennsylvania. Data used in the present study were drawn from a needs assessment survey that was designed to characterize the home environments of low-income families with young children and to better inform future nutrition education programming for the Head Start participants. The survey was distributed through classrooms to 1297 Head Start families. If parents had more than one child enrolled in Head Start, they were instructed to complete the survey for their oldest child enrolled in the program. Of the 1297 distributed surveys, 379 (30%) were returned in the mail. Caregivers received a \$25 gift card for their participation. Data collection spanned May 2017 to May 2018. Among nine families, a survey was completed for two children in the home, thus we excluded the survey for the younger of the two children. Four children were excluded because they were outside the age range of Head Start eligibility, resulting in a final study sample of 365. For the purpose of the present study, a minimum sample size of 134 participants was required to test for the associations between our main variables of interest (FRM behavior, FRM self-confidence, and HFI) at 80% power and with 95% confidence interval. The sample size calculations were done using data from previous studies that examined similar associations [10,16]. Informed consent was obtained from subjects prior to their participation in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Institutional Review Board at the Pennsylvania State University (00007467).

#### *2.2. Caregiver and Household Characteristics*

The survey included questions related to the caregiver characteristics, such as age and sex, ethnicity, education, employment, and marital status. As for household characteristics, questions included child's age, number of children in the household, number of people supported by household income, participation in assistance programs in the past 12 months (e.g., Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) and Supplemental Nutrition Assistance Program (SNAP)), and household income. Household income was missing in seventy-four of 365 households (20.2%). Missing income was imputed based on WIC and SNAP status, parent education, marital status, and employment using PROC MI in SAS software (SAS Institute Inc., Cary, NC, USA).

#### *2.3. Household Food Insecurity Status*

Household food insecurity (HFI) experienced during the previous 12 months was measured using the 18-item US Household Food Security Survey Module [1]. The food security status of households was determined by the number of food-insecure conditions and behaviors the household reports. Households were classified as 'food secure' if participants responded affirmatively to two or fewer items on the 18-item scale and as 'food insecure' if the affirmative responses were on three or more items, such as "cutting the size of meals or skipping meals because there wasn't enough money for food during the past 12 months" or "losing weight because there wasn't enough money for food".

#### *2.4. Food Resource Management (FRM) Self-Confidence and Behaviors*

FRM self-confidence and FRM behaviors of caregivers were assessed in the present study using two sets of questions derived from the SNAP-Ed evaluation framework guide and toolkit [11]. These questions were previously used and validated in other studies assessing the impact of nutrition education programs targeting low-income adults, including SNAP-Ed, Cooking Matters, and Expanded Food and Nutrition Education Program (EFNEP), on participants' FRM skills [10,14,29] and confidence [16,29].

The caregivers' self-confidence in FRM abilities (in the past 12 months) was assessed in the present study using five questions. Three questions assessed caregiver confidence to "choose the best-priced form of fruits and vegetables", "buy healthy foods on a budget", and "cook healthy foods on a budget"; and two questions were related to caregiver's confidence in their ability to "make a shopping list and stick to it" and "compare prices of similar foods to find the best value". Responses for these questions were measured using a 4-point Likert scale that ranged from 1 (*not very confident*) to 4 (*very confident*). An average FRM self-confidence score was calculated for each participant based on their responses to the five questions, and a binary score was later created for FRM self-confidence to classify participants into two groups (low/high): participants with scores less than the median were categorized as "low" FRM self-confidence, whereas participants with scores greater than or equal to the median score were categorized as "high" FRM self-confidence. A high FRM self-confidence indicated a greater self-confidence in shopping, preparing foods, and managing food resources on a budget.

The FRM behaviors of participants in the present study were assessed using six questions from the SNAP-Ed evaluation framework and toolkit, asking how often do caregivers "plan meals before shopping", "prepare shopping list", "compare prices before buying", "use grocery store flyers", and "identify foods on sales or use coupons" [11]. A 5-point response scale (1 = *never*, 2 = *rarely*, 3 = *sometimes*, 4 = *usually*, 5 = *always*) was used for each of the FRM behavior items. An average FRM behavior score was first calculated, then a binary score was created to classify participants into two groups: participants with scores less than the median were categorized as "low" FRM behaviors, whereas participants with scores greater than or equal to the median score were categorized as "high" FRM behaviors. A high FRM behavior indicated better practices in meal planning, shopping with a grocery list, and comparing prices.

#### *2.5. Financial Situation, Financial Practices, and Di*ffi*culties*

To assess the financial situation, respondents were asked to describe their own financial situation with responses including: 1 = "*Very comfortable and secure*", 2 = "*Very comfortable and secure*", 3 = "*Occasionally have some di*ffi*culty making ends meet*", 4 = "*Tough to make ends meet but keeping head above water*", and 5 = "*In over your head*". As for financial difficulties, these were evaluated based on 5 questions from the USDA national food study [30] to assess difficulties that individuals had in meeting their essential household expenses, such as mortgage or rent payments, utility bills, or important medical care during the past six months.

Financial practices of the caregivers were also assessed using 3 questions that were derived from the USDA national food study [30]. Caregivers were asked to report how frequently they adopted the following practices during the past 6 months: "review your bills for accuracy", "pay your bills on time", and "pay more than the "minimum payment due" on your credit card bills". Response options ranged from 1 = never to 5 = always. An average financial practices score was calculated for each participant based on their responses to the five questions, and a binary score was later created (low/high): participants with scores less than the median were categorized as "low", whereas participants with scores greater than or equal to the median score were categorized as "high", referring to those with better financial practices.

#### *2.6. Statistical Analyses*

Descriptive statistics were reported in the present study as frequencies and proportions for categorical variables and as medians and interquartile ranges (IQR) for non-normal continuous variables. Chi-square tests and Mann-Whitney U tests were conducted to explore differences between categorical variables and non-normal continuous variables by HFI status (food secure vs. food insecure households), respectively. Simple and multiple logistic regression analyses were also conducted to examine the association between FRM self-confidence, FRM behaviors, and financial practices by HFI status. Variables included in the multiple logistic regression models were those found to have a significant bivariate relationship with HFI and were statistically significant in the simple logistic models (*p* < 0.05). Sensitivity analyses were also conducted to assess the validity of findings by: (1) adjusting for significant and non-significant sociodemographic variables as potential confounders

in the logistic regression models, (2) running linear regression models with HFI and other variables of interest (FRM behavior, FRM self-confidence and financial practices) as continuous variables, and (3) running models using imputed and non-imputed income data. For the models with non-imputed income, we excluded subjects with missing income in the sensitivity analysis. Results from the logistic regression models were expressed as odds ratios with 95% confidence intervals. Statistical analyses were conducted using Stata/MP version 15.1 (StataCorp. College Station, TX, USA). A *p*-value of 0.05 was used to detect significance in all analyses used in the present study.

#### **3. Results**

#### *3.1. Descriptive Characteristics of the Study Sample*

The majority of caregivers in our study sample were females (96%), White non-Hispanic (98%), and completed high school education level or less (61%). The median age of caregivers was 30 (IQR = 9) years old. More than half of study participants were married or partnered (57%) and unemployed (54%). In addition, almost three quarters of participants were receiving SNAP benefits (75%) and WIC (70%). The median number of children in the household was 2, and the prevalence of HFI was 37% (see Table 1).

Caregiver and household characteristics of the study sample were also presented by HFI in Table 1. Participation in the SNAP/Food Stamps program was significantly greater among food insecure households compared to food secure ones (84% vs. 69%, *p* = 0.001), whereas participation in WIC was less common among food insecure households (64% vs. 74%, respectively, *p* = 0.041). No other significant associations were noted between HFI and demographic characteristics in the present study.




**Table 1.** *Cont.*

<sup>1</sup> Categorical variables were presented as *n* (%) and non-normal continuous variables were presented as medians and interquartile ranges (IQR). IQR represents the difference between the upper and lower quartiles (Q3−Q1). <sup>2</sup> Chi-square tests were conducted to determine differences between categorical variables and binary food security status. <sup>3</sup> Mann-Whitney U tests were used to determine differences between non-normal continuous variables and binary food security status. <sup>4</sup> Households with low and very low food security status were categorized as food insecure and those with marginal or high food security were classified as food secure [1]. <sup>5</sup> SNAP, Supplemental Nutrition Assistance Program; TANF, Temporary Assistance for Needy Families; WIC, The Special Supplemental Nutrition Program for Women, Infants, and Children.

#### *3.2. Food Resource Management and Household Food Insecurity*

Table 2 presents FRM self-confidence and FRM behaviors of caregivers in the study sample and by HFI. Results showed that almost three-quarters of caregivers were *moderately* to *very confident* in choosing best priced food items, comparing food prices for best values, and cooking healthy food items on a budget. In addition, slightly greater than two-thirds of participants were *moderately* or *highly confident* in "buying health foods for their families on a budget" and "making a shopping list and sticking to it". The proportion of participants reporting *usually* or *always* adopting FRM behaviors ranged between 31% and 79%. The less adopted FRM behaviors included "using grocery store flyers to plan meals" (31%), "planning of meals prior to grocery shopping" (57%), and "identifying foods on sale or using coupons to save money" (57%).

**Table 2.** Food resource management (FRM) self-confidence and FRM behaviors of Head Start caregivers in the study sample by household food insecurity, (*n* = 365) 1.



**Table 2.** *Cont.*

<sup>1</sup> Chi-square test was conducted to determine differences between categorical variables and binary food security status. \* For expected cell counts less than 5, *p*-value from Fisher's exact test was reported.

Significant differences were observed between food secure and food insecure households for all FRM self-confidence items (*p*-value < 0.05). More specifically, caregivers in food secure households were more likely to report being very confident in their abilities to "choose best priced fruits and vegetables" (42% vs. 27%), "buy healthy foods for their families" (41% vs. 21%), "cook healthy foods on a budget" (45% vs. 25%), "make a shopping list and stick to it" (43% vs. 26%), and "compare prices of similar foods when shopping to get the best value" (47% vs. 32%) when compared to their food insecure counterparts. On the other hand, only one item from the FRM behaviors was found to be significantly different between food secure and food insecure households in our study sample. A greater proportion of caregivers in food secure households reported that they always "use a shopping list when grocery shopping" as compared to their food insecure counterparts (49% vs. 33%, Table 2).

#### *3.3. Financial Situation, Practices, and Di*ffi*culties and Household Food Insecurity*

When caregivers were asked to describe the household's financial situation, 37% of the total sample reported being "very comfortable and secure" or "able to make ends meet without much difficulty", 34% "occasionally have some difficulty making ends meet", and the remaining 29% reported it is "tough to make ends meet but keeping your head above water" or they are "in over their heads". In terms of financial practices, the majority of caregivers in the study sample responded they *usually*

*or always* "review bills for accuracy" (75%) and "pay bills on time" (79%), yet less than one-third of participants responded they "pay more than the "minimum payment due" on credit card bills" as frequently. With respect to financial difficulties, 39% of caregivers in our study reported going through a time "when they could not pay mortgage or rent, electricity or gas utilities, or important medical expenses", and 44% reported going through periods when they "could not pay the full amount of gas, oil, or electricity bills" (Table 3).

**Table 3.** Financial situation, practices and difficulties of Head Start caregivers in the study sample and by household food insecurity (*n* = 365) 1.


\* For cells with counts less than 5 in the chi-square analysis, *p*-value from Fisher's exact test was reported.

In addition, Table 3 presents the financial situation, difficulties, and financial practices of caregivers in the study sample by HFI. Overall, food insecure households were more likely to report their financial situation as "occasionally have some difficulty making ends meet" (40% vs. 31%) or "tough to make ends meet but keeping head above water" compared to their food secure counterparts (45% vs. 14%). In terms of financial practices, a higher proportion of caregivers in food secure households reported they *always* "pay bills on time" (50% vs. 25%) and "pay more than the minimum payment due on credit card bills" (20% vs. 11%) compared to their food insecure counterparts. On the other hand, food insecure households were significantly more likely to report facing financial difficulties compared to food secure ones: "has there been a time when you could not pay your mortgage or rent, electricity or gas utilities, or important medical expenses?" (63% vs. 24%) and "has there been a time when you could not pay the full amount of gas, oil, or electricity bills" (64% vs. 32%), *p*-value < 0.001.

#### *3.4. Food Resource Management, Financial Practices, and Household Food Insecurity*

The associations between FRM self-confidence, FRM behaviors, and financial practices with HFI were also explored in the present study (Table 4). Results from the logistic regression analyses showed that caregivers with high FRM self-confidence had lower odds of HFI (OR = 0.54, 95% CI: 0.33, 0.87, *p* = 0.012), even after adjusting for financial practices and participation in food assistance programs (SNAP and WIC). Although the association between financial practices and HFI was significant in the simple regression analysis, this association lost its statistical significance in the adjusted model (OR = 0.77, 95%CI: 0.46, 1.3, *p* = 0.338). Results from the models remained robust after conducting sensitivity analyses and adjusting for significant and non-significant sociodemographic variables, including parent's age, employment, household income (imputed and not imputed values), and participation in food assistance programs in the past 12 months (Supplemental tables—Tables S1 and S2).


**Table 4.** Simple and multiple logistic regression analyses of food resource management (FRM) self-confidence, FRM behaviors, and financial practices of Head Start caregivers with household food insecurity (*n* = 365).

<sup>1</sup> The model was adjusted for socio-economic characteristics found to be significant correlates of household food insecurity, namely participation in any assistance program (in the past 12 months) including SNAP/Food Stamps or WIC.

#### **4. Discussion**

Food insecurity remains a social and public health problem for low-income families with young children in the US that has serious consequences on children's overall health and wellbeing. To our knowledge, the present study is the first to examine the associations between FRM self-confidence, FRM behaviors, and financial practices by HFI status in a sample of low-income households with young children. Using a sample of SNAP-Ed-eligible Head Start families, our study findings showed that caregiver's self-confidence in their FRM was associated with lower odds of HFI. Nevertheless, the associations between the FRM behaviors and financial practices of Head Start caregivers by HFI were not statistically significant in the adjusted models.

As hypothesized, caregivers with high FRM self-confidence had lower odds of HFI in the present study, even after adjusting for other correlates including FRM behaviors, financial practices and participation in other federal assistance programs. When individual FRM questions were explored, all FRM self-confidence questions were also found to significantly differ by HFI status. More specifically, caregivers in food secure households were more likely to report being "*very confident*" in their abilities to choose the best priced fruits and vegetables, compare prices of similar foods when shopping to get the best value, as well as buy and cook healthy foods for their families on a budget as compared to their food insecure counterparts. These results were in concordance with those reported earlier by Begley et al. (2019) showing that food secure participants, who were assessed at the enrollment stage of an adult food literacy program in Australia, reported being "*always confident*" about managing money for healthy food compared to food insecure participants (41.2% vs. 9%) and "*always confident*" in their ability to cook a variety of healthy meals (21.9% vs. 15.4%) [31]. Our findings were also consistent with a few studies conducted to date that highlight how greater self-efficacy in shopping and preparing healthy food, based on nutrition education programs targeting low income adults, has been associated with lower risk of food insecurity [16,29]. According to Martin et al. (2016), self-efficacy in managing food resources was found to be associated with a decrease in very low food security levels among food pantry clients participating in the *Freshplace* intervention. This was an 18-month innovative food pantry intervention that combined several strategies to boost the confidence of participants, such as motivational interviewing and serving food in client-choice format to increase their confidence in planning meals ahead of time, making a shopping list before going to the store, and making food money last all month [16]. Another study evaluating the impact of *Cooking Matters for Adults* nutrition education program showed significant improvements in the FRM skills and self-confidence in managing food resources of low-income households up to six months after the program completion. In addition, participants in the Cooking Matters intervention were worried less that food would run out before they could get money to buy more [29]. It is worth noting that these nutrition education programs were focused primarily on improving the self-efficacy of low-income adults as integral components for the uptake and maintenance of FRM skills to maximize the use of limited food dollars.

Although self-efficacy represents a key construct within theories of behavioral change and has been shown to be effective in promoting healthy behaviors for weight loss, exercise, and chronic disease management [32,33], only a few studies to date, as described earlier, have explored the association between self-confidence in FRM with food insecurity among low-income households [16,29]. To our knowledge, the present study is the first to examine these associations in low-income households, focusing primarily on those with young children. Our study findings suggest that increased confidence in resource management skills among caregivers may be associated with lower risk of HFI. These results may be promising for families with young children, who may have increased concerns about smart shopping, stretching their food dollars, as well as cooking tasty and low-cost food to feed their children [20,34]. Food insecure individuals may be also influenced by financial, social, and personal stressors that can further affect their confidence in their ability to shop, prepare, and plan a healthy meal on a limited budget [35,36]. Thus, federal assistance and nutrition education programs targeting families with young children, such as Head Start and SNAP-Ed, may need to give particular attention to strategies that can help improve the self-confidence and efficacy of caregivers in their resource management skills. These programs can also help participants in accessing community-level resources and in overcoming common misconceptions and barriers to enrolling in other federal assistance programs, including WIC [37].

Nevertheless, when exploring FRM behaviors, only one of the six behaviors of caregivers were shown to differ significantly by HFI in the present study. In addition, the association between FRM behaviors and HFI was not found to be statistically significant in the regression models. Contrary to our study findings, food secure families were previously observed to have overall better FRM skills, such as shopping for sales, researching for best prices on particular products, traveling to multiple stores, and planning meals around their limited budgets [35,38]. According to Begley et al. (2019), individuals who reported at the onset of a food literacy program a low frequency of adopting certain planning and food preparation behaviors, such as planning meals ahead of time and making a list before they shop, were significantly more likely to be food insecure than those who reported adopting more frequently these behaviors [31]. The limited differences in FRM behaviors by HFI, as observed in the present study, can be attributed in part to the overall low proportion of caregivers who reported planning their meals prior to grocery shopping, using grocery store flyers to plan their meals, or identifying foods on sale and using coupons to save money. Another reason could be differences in questions raised when assessing caregivers' FRM confidence and behaviors in the present study. For example, questions relevant to buying and cooking healthy foods were only present in the FRM self-confidence questionnaire, whereas questions related to using shopping lists and planning meals prior to shopping were common among both scales. Caregivers participating in the present study may have also received family-centered services that cover topics related to child nutrition, growth, and development as part of the Head Start programs [39–42], which could have influenced their perceived confidence in providing healthy foods for their children. Nevertheless, confidence alone might be insufficient to alleviate HFI, and households with higher confidence may not be able to adopt adequate FRM behaviors when other environmental, financial, and personal barriers exist, such as limited availability and/or access to food stores with healthy and nutritious food, lack of kitchen appliances, as well as time and money constraints [31,38,43]. Poor physical and mental health can also affect the FRM skills and capabilities of food insecure individuals [38,44] and are worth further exploration when examining the association between resource management skills and HFI.

A growing body of evidence suggests that households facing economic hardships and with limited knowledge of basic financial concepts (i.e., financial literacy) are also more likely to experience food insecurity compared to those with higher financial management skills [22,25]. In line with former research, results from the present study showed significant differences in the financial situation, difficulties, and financial practices of caregivers by HFI status. Compared to caregivers from food insecure households, those from food secure households were more likely to report better financial situation and lower financial difficulties reflected through their ability to pay their mortgages or other basic expenses (such as rent, electricity, gas, and medical expenses). In addition, caregivers from food secure households were also more likely to report frequently adopting certain financial practices, such as paying bills on time and paying more than the "minimum payment due" on credit card bills. Nevertheless, the association between higher financial practices and HFI lost its statistical significance in the adjusted models. These results may be explained by the lower income levels of households enrolled in federal assistance programs, such as SNAP and WIC, who represent approximately three-quarters of the study sample, and who may be facing heightened financial hardships that could have attenuated the association between caregivers' better financial management practices and their HFI status. Our study findings highlight the need to further explore the association between financial literacy (knowledge and capabilities) and HFI, particularly in low-income households with children. The latter group may be at increased risk of facing economic hardships, and thus may adopt risky coping strategies that can further increase their risk for HFI and its adverse health consequences [22,45].

#### *Strengths and Limitations*

To our knowledge, this study is the first to explore the associations between FRM self-confidence, FRM behaviors and financial practices by HFI among a sample of low-income Head Start households with young children. Nevertheless, the present study has a number of limitations worth considering. First, the study is cross-sectional in nature, thus causality cannot be determined when exploring the associations between FRM self-confidence, FRM behaviors, and financial practices by HFI. The association between FRM self-confidence and food insecurity, as reported in the present study, may have been bidirectional in nature. Caregivers in food insecure households may have poor conditions

that affect their self-confidence in their resource management skills as compared to food secure households; on the other hand, having higher self-confidence may also improve one's capabilities to access and utilize food, which can influence their food security and feeling of self-sufficiency [31]. Another limitation of the present study is that data were self-reported, thus we cannot rule out response bias. Our study findings may also have limited representativeness with a moderate survey response rate (30%) and the study population limited to only four rural areas in central Pennsylvania. Thus, results cannot be generalizable and the external validity of our findings may be limited to certain low-income families. Albeit modest, the response rate in the present study was still similar to other surveys conducted with rural Head Start families in Colorado (28.5%) and Appalachian Ohio (42%) [46,47]. Future research considering more diverse and larger samples of Head Start families are still needed to further examine the associations explored in the present study.

#### **5. Conclusions**

Our study findings suggest that increased self-confidence in FRM among caregivers of young children is associated with lower odds of HFI among low-income Head Start families. Nutrition and health education programs, such as SNAP-Ed and WIC, that are designed to assist low-income households in alleviating their HFI status may need to give more emphasis to the self-efficacy and confidence of caregivers in stretching their food dollars and adopting adequate FRM skills. The strategies may help caregivers in offering healthy food and improve the food choices offered to their children. Caregivers can also play a pivotal role in structuring their children's early experiences with food through child feeding practices, social modeling of healthy eating behaviors, and regulating the quality and quantity of food provided to the child [48–50]. Thus, future research should examine the extent to which nutrition education programs that focus on improving FRM self-confidence and behaviors can contribute (directly or indirectly) to the feeding practices of caregivers and, subsequently, to the diet quality and nutrition outcomes of young children in low-income households. It is also important to further investigate the role that financial literacy and practices of caregivers can play in improving the food security of low-income households.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2072-6643/12/8/2304/s1. Table S1. Sensitivity analysis to evaluate the association between food resource management (FRM) self-confidence with household food insecurity adjusting for significant and non-significant sociodemographic variables as potential confounders in the logistic regression models, and using imputed and non-imputed income values. Table S2: Sensitivity analysis to evaluate the associations between food resource management (FRM) self-confidence, FRM behaviors, and financial practices of Head Start caregivers by household food insecurity using simple and multiple linear regression analyses.

**Author Contributions:** Conceptualization, L.J., M.N., S.G.E., J.S.S.; methodology, L.J., M.N., S.G.E., M.D.-E.-H., J.S.S.; software, S.G.E., M.D.-E.-H.; validation, L.J., S.G.E., M.D.-E.-H.; formal analysis, L.J. and M.D.-E.-H.; investigation, J.S.S.; resources, J.S.S.; data curation, S.G.E.; writing—original draft preparation, L.J.; writing—review and editing, L.J., M.N., S.G.E., M.D.-E.-H., J.S.S.; visualization, L.J. and M.D.-E.-H.; supervision, J.S.S.; project administration, J.S.S.; funding acquisition, J.S.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This project was funded by USDA's Supplemental Nutrition Assistance Program (SNAP) through the PA Department of Human Services (DHS).

**Conflicts of Interest:** The authors declare no conflicts of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Supplemental Nutrition Assistance Program-Education Improves Food Security Independent of Food Assistance and Program Characteristics**

#### **Heather A. Eicher-Miller 1,\*, Rebecca L. Rivera 1,2, Hanxi Sun 3, Yumin Zhang 3, Melissa K. Maulding 4,5 and Angela R. Abbott <sup>4</sup>**


Received: 4 August 2020; Accepted: 27 August 2020; Published: 29 August 2020

**Abstract:** The purpose of this project was to determine whether consistent food assistance program participation or changes in participation over time mediated or moderated the effect of federal nutrition education through the Supplemental Nutrition Assistance Program-Education (SNAP-Ed) on food security and determine the associations of SNAP-Ed program delivery characteristics with change in food security. This secondary analysis used data from a randomized controlled trial from September 2013 through April 2015. SNAP-Ed-eligible participants (*n* = 328; ≥18 years) in households with children were recruited from 39 counties in Indiana, USA. The dependent variable was one year change in household food security score measured using the United States Household Food Security Survey Module. Assessment of mediation used Barron-Kenny analysis and moderation used interactions of food assistance program use and changes over time with treatment group in general linear regression modeling. Program delivery characteristics were investigated using mixed linear regression modeling. Results showed that neither consistent participation nor changes in food assistance program participation over time mediated nor moderated the effect of SNAP-Ed on food security and neither were SNAP-Ed program delivery characteristics associated with change in food security over the one year study period. SNAP-Ed directly improved food security among SNAP-Ed-eligible Indiana households with children regardless of food assistance program participation and changes over time or varying program delivery characteristics.

**Keywords:** supplemental nutrition assistance program-education; SNAP-Ed; nutrition education; food assistance; SNAP; food stamps; WIC; food security; food pantry; emergency food programs

#### **1. Introduction**

Members of low-income households face a high burden of food insecurity, poor nutrition, and undesirable health outcomes [1–5]. The Supplemental Nutrition Assistance Program-Education (SNAP-Ed) is a program of the United States Department of Agriculture (USDA) Food and Nutrition Service (FNS) that offers education on nutrition, budgeting, and resource management to low-income households to improve dietary intake and food security [6,7]. SNAP-Ed has been shown to improve

household and adult food security in previous longitudinal randomized controlled trials [8,9]. Approximately 73% of households interested in receiving SNAP-Ed also report participating in at least one of three other food assistance programs [9] directed to alleviate food insecurity in qualifying low-income households [10], including the Supplemental Nutrition Assistance Program (SNAP), the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), and The Emergency Food Assistance Program (TEFAP). SNAP and WIC provide financial and food resources to help individuals and families obtain foods to supplement their nutritional needs [11,12] while TEFAP provides foods to state agencies who partner with private and local organizations to distribute emergency foods to food banks and food pantries where individuals in need may access foods at no cost [13]. Mutual participation in SNAP-Ed and SNAP is not required; some SNAP-Ed participants may not qualify for SNAP benefits or choose not to participate in SNAP. Further, sometimes SNAP-Ed lessons are used to fulfill WIC education requirements.

Previous evidence of improvement in food security because of SNAP [2], WIC [14], and associations with emergency program use [15], taken with knowledge of the common practice of simultaneous participation in food assistance programs and nutrition education programs, suggests that the changes observed in food security previously attributed to nutrition education [9] may actually be accounted for by participation or change in participation of food assistance (mediation). It may also be likely that the effect of nutrition education on food security may be differential by food assistance participation or changes in food assistance (moderation). SNAP-Ed educators commonly help participants with eligibility and encourage their application for local, state, and federal food assistance as part of the resource management education offered, making salient the reality that participation status in food assistance programs may frequently change during nutrition education participation [16]. Previous investigation of nutrition education program effectiveness on food insecurity has focused on singular program use and has not considered mediation or moderation by food assistance participation or changes in their use, specifically regarding the three most common food assistance programs, SNAP, WIC, and TEFAP [17]. Only one previous non-experimental short-term study evaluated joint use of two of these programs and showed that SNAP-Ed participants who were also receiving SNAP benefits and made more improvement in resource management skills, reported the greatest decrease in running out of food (measured by only one question) compared with participants who were not receiving SNAP benefits and who had less improvement in resource management skills [18]. Additional factors of relevance in SNAP-Ed effect on participant food security improvement are SNAP-Ed program delivery characteristics, such as the number of lessons, group or individual lessons, or SNAP-Ed educator. In Indiana, over sixty educators deliver up to ten SNAP-Ed lessons using group and individual lesson delivery. Program variability presented by these characteristics are inherent to SNAP-Ed and may potentially be associated with an effect on food security. For example, food security improvement may be influenced by participants receiving 10 rather than 4 lessons, individualized compared with group lessons, or by interaction with a particular SNAP-Ed educator.

Therefore, determining the potential mediating or moderating role of food assistance participation and changes in participation over time on nutrition education program participation would clarify knowledge of impacts to food security. Examination of the role of SNAP-Ed program characteristics number of lessons, delivery format, and variability of educator to food security improvement would inform program and policy of important programmatic aspects of success. The objectives of this paper were investigated among adults ≥18 years from Indiana in a dataset where a decrease of 1.2 ± 0.4 (mean ± SE) units in household food security score over the one year study period, indicating a meaningful longitudinal improvement in food security among the intervention compared to the control group, was previously discovered [9], and included:

1. Determine whether participation and changes in participation status in food assistance programs SNAP, WIC, and food pantries over one year mediated the effect of a SNAP-Ed intervention on one year change in household food security.


#### **2. Materials and Methods**

#### *2.1. Study Population*

For this secondary data analysis, all data were obtained from The Indiana SNAP-Ed Long-term Study, a longitudinal (one year) parallel-arm randomized controlled nutrition education intervention trial conducted between August 2013 and April 2015 [9]. Thirty-five county-level Indiana SNAP-Ed nutrition education paraprofessionals (SNAP-Ed educators) recruited adult participants (*n* = 575) aged ≥18 years from August 2013 to March 2014 and administered baseline assessments. Participants were recruited from locations such as WIC clinics, food pantries, or Indiana Cooperative Extension county offices. The one year follow-up assessments were completed from September 2014 through April 2015. Data to address the hypotheses of this study are expected to maintain relevance to current program and participants as food insecurity in Indiana from 2013–2015 was not statistically significantly different from 2016–2018 estimates [5], and the data represent a unique opportunity to comprehensively address hypotheses using a singular sample. Only participants who completed the study (i.e., baseline and one year follow-up assessments) were included in the analysis presented here (total *n* = 328, control *n* = 163, intervention *n* = 165). SNAP-Ed educators were trained to determine participant study eligibility and randomly assigned participants to either the non-active control group or intervention group using an allocation ratio of ~1:1. A random number allocated the first participant or group recruited simultaneously (to prevent knowledge of different treatment) to the intervention or control group and then an alternating assignment was followed. After treatment group assignment, SNAP-Ed educators delivered lessons to the intervention group participants as per program protocol over the following four to ten weeks, at approximately 1 lesson per week, and facilitated all survey assessments to both treatment groups. Eligible study participants included Indiana adult residents who had one or more children living in the household, had not received a SNAP-Ed lesson in the past one year, were able to speak, read, and write in English, and were willing to wait one year to receive nutrition education lessons.

#### *2.2. Intervention*

The intervention consisted of the first four (out of ten) lessons in the Indiana SNAP-Ed curriculum [19] as these lessons comprise SNAP-Ed guidance and cover the USDA key behavioral outcomes of maintaining caloric balance over time for a healthy weight and consumption of nutrient-dense foods and beverages. Additionally, lessons included instruction on budgeting food resources through the following lesson topics: applying USDA MyPlate to build healthy meals, using food labels to make healthy choices, identifying the importance of whole grains, and adding more fruits and vegetables to meals [19,20]. The Purdue Institutional Review Board approved the trial protocol and all participants provided written informed consent. The trial was registered at www.clinicaltrials.gov as NCT03436589.

#### *2.3. Food Security Measures*

Household food security score was measured using the 18-item USDA U.S. Household Food Security Survey Module (US HFSSM) with scores ranging from 0 (food secure) to 18 (very low food secure) and a 12-month reference period [21,22]. Categorical classification of food security at baseline was also constructed as food secure, marginally food secure, and food insecure according to prior guidance [22]. Change in food security score was the response variable in this secondary data analysis to determine a more specific change compared with using food security categories, and was quantified by subtracting the baseline score from the one year follow-up score for each participant.

#### *2.4. Food Assistance Program Measures Used in Objectives 1 and 2*

Study participants self-reported participation status in SNAP, WIC, and food pantries over the 30 days prior to both baseline and one year follow-up assessments because the food assistance provided through these programs are generally distributed on a monthly basis. One month or 30 days was considered the minimal amount of time that these programs may exert influence on a participant household and on SNAP-Ed effectiveness. Missing values were 8% (*n* = 27) at baseline and 15% (*n* = 50) at follow-up. A sensitivity analysis was conducted where missing values were coded as participation and compared to coding values as non-participation. The results did not change so coding as non-participation was applied. Participation in local, state, or national food assistance programs other than SNAP, WIC, or food pantries was not recorded.

Three individual four-level categorical variables referred to as "change in one year participation status" were created for SNAP, WIC, and food pantries, respectively, to represent any changes or no changes in food assistance participation status between the 30 days prior to baseline and the 30 days prior to one year follow-up assessments. "Change in one year participation status" variables were created by concatenating the baseline and one year follow-up binary variables to simultaneously represent the participation status for each of the food assistance programs at baseline and at one year follow-up in addition to change in participation status if it occurred (00 = no participation; 10 = participation at baseline only; 01 = participation at one year follow-up only; 11 = participation at both baseline and one year follow-up). These variables were used as a categorical independent variable to address the first and second research objectives, whether change in food assistance program participation status or consistency mediated or moderated the impact of SNAP-Ed on one year change in food security score.

#### *2.5. SNAP-Ed Program Characteristics Measures Used in Objective 3*

The number of SNAP-Ed lessons a participant received, the lesson delivery format, and which SNAP-Ed educator delivered the lessons were investigated as the SNAP-Ed program characteristics among intervention group participants who completed the required four lessons to address the third research objective. Participants assigned to the intervention group that did not complete the four required intervention lessons, lost contact with SNAP-Ed educators, or did not follow the study protocol were considered withdrawn from the study (*n* = 87). The number of lessons (4–10 lessons) a participant received was recorded by the SNAP-Ed educator at each lesson and summed at the one year follow-up assessment. Lesson delivery format was a categorical variable with three levels representing how the participant received lessons (one-to-one lessons, group lessons, combination of one-to-one and group lessons) and was based on the preference of the participant to attend group lessons, educator facilitation, and the schedule of group or individual lessons. Assignment of SNAP-Ed educator (*n* = 37) was determined by the participant's county of residence at recruitment.

#### *2.6. Other Covariates*

A binary variable for treatment group (control, intervention) was used to address the first and second research objectives. Time was included as a binary variable in mixed regression modeling (baseline, follow-up) to address the third research objective. Self-reported baseline participant characteristics identified as potential confounders through Chi-square comparisons between the intervention and control groups were investigated: sex (female, male); age in years (18–30, 31–50, ≥51); race/ethnicity (non-Hispanic white, other); highest level of education among the household (no high school diploma, high school diploma, or General Educational Development certification indicating high school level skills; some college/associate's degree; ≥bachelor's degree); marital status

(living with partner/married, never married, divorced/separated/widowed); household employment (household member employed, no household member employed); household poverty status (<federal poverty guideline, ≥federal poverty guideline); household size (two, three, four, or ≥five household members); SNAP, WIC, or food pantry participation status 30 days prior to baseline (not participating, participating), and food security category at baseline (food secure, marginally food secure, food insecure). Two categories for race/ethnicity were used in this study because reports other than non-Hispanic white were very few: 3 participants reported American Indian, 1 reported Asian, and 7 reported non-Hispanic black. Maintaining separate categories would threaten the robustness of the analysis and model fit so categories were combined to a single category.

#### *2.7. Statistical Methods*

To address the first research objective, the Baron-Kenny causal mediation approach was used to investigate whether the suspected mediator "change in one year participation status" in SNAP, WIC, or food pantries mediated the effect of the exposure, SNAP-Ed intervention, on the outcome, change in household food security score over the one year study period [23]. Additional covariates are not included in the Baron-Kenny three variable system regression approach (Figure 1, below); investigation of the role of other covariates are outside of the scope of the hypotheses of this paper.

Step 1. Determine c: Results described in text in Section 3.1, Step 1. Step 2. Determine a: Results described in text in Section 3.1, Step 2 and in Tables 1 and 2. Step 3. Determine b: Results described in text in Section 3.1, Step 3. Step 4. Determine c and b if there are significant relationships from Steps 1-3: Mediation is supported when the effect of the suspected mediator is significant after controlling for the exposure. Results described in text in Section 3.1, Step 4.

**Figure 1.** Hypothesized Baron-Kenny causal mediation model of the Supplemental Nutrition Assistance Program-Education (SNAP-Ed) intervention effect by the "change in one year participation status" in Supplemental Nutrition Assistance Program (SNAP), Women, Infants, and Children (WIC), or food pantries on the change in one year food security score among Indiana SNAP-Ed Study participants. a = the relationship of the exposure on the suspected mediator using regression, b = the relationship of the suspected mediator on the outcome using regression, c = the relationship of the exposure on the outcome using regression.

To address the second research objective, interactions between "change in one year participation status" (SNAP, WIC, and food pantries) and treatment group variables were used in general linear regression modeling to determine whether the change in food assistance program participation, consistent participation, or non-participation moderated the effect of SNAP-Ed on the change in food security score over the one year study period. SNAP, WIC, and food pantry interactions with treatment group were investigated in separate models; the reference group was consistent non-participation during the 30 days prior to baseline and one year follow-up. Other participant characteristics (sex, age, race/ethnicity, education, poverty status, employment status, marital status, household size) were initially included in the models as potential confounders but removed because they were not influential (*p* < 0.2). Statistical power to detect a difference at a significance level of α = 0.05 with power at 0.90, for a one unit improvement in food security based on previous study data [9,18,24] was confirmed using a power analysis procedure for general linear regression models. A treatment effect of one unit on the food security scale was chosen for the power analysis because of the practical relevance and

potential of a one unit decrease to transition a participant between two food security statuses and the associated positive benefit. In addition, an approximate one unit change was discovered in the study from which this data was derived and considered reasonable. Tukey adjustment for multiple comparisons was applied.

To address the third research objective, a mixed linear regression model was used to determine the association of the number of lessons, lesson delivery format, and variability between SNAP-Ed educators with change in food security score over one year among the intervention group (*n* = 165). Time, number of lessons, and lesson delivery format were included as fixed effects in the model. Participants and SNAP-Ed educator were considered random effects. The covariance structure was specified as compound symmetry after using the Sawa Bayesian information criterion (BIC) to compare various covariance structures. None of the potential participant characteristic confounders were found influential (*p* > 0.2), except for age (*p* = 0.02) which was included as a covariate in the model. Statistical power to detect a difference at a significance level of α = 0.05 with power at 0.90 and one unit improvement in food security was confirmed using a power analysis procedure for mixed linear regression models.

Model assumptions were checked by plotting residuals against predicted means, Q-Q plots, and histograms of residuals for general and mixed linear regression modeling and applied to each study objective. All analyses were completed using SAS® software version 9.4 (SAS Institute Inc., Cary, NC, USA).

#### **3. Results**

The characteristics and food security of participants in the intervention and control groups are shown in Table 1.


**Table 1.** Comparison of baseline sociodemographic characteristics by treatment group of Indiana SNAP-Ed participants among households with children using Chi-Square analysis.


**Table 1.** *Cont*.

Values are counts, percentages, and *p*-values from Chi-square comparisons of the distributions among sociodemographic characteristics between control and intervention group participants. Total numbers do not always add to sample size due to missing values and percentages do not always add to 100 due to rounding. \* *p* ≤ 0.05. Abbreviations: SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP, Supplemental Nutrition Assistance Program; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

Participation in WIC, food pantries, and employment were the only characteristics with significantly different distributions among intervention and control groups at baseline.

#### *3.1. Research Objective 1: Test for Food Assistance Program Mediation of SNAP-Ed E*ff*ect on Food Security*

Step 1: Food security score did not differ between treatment groups at baseline using regression (β = −0.4, SE = 0.3, *p* = 0.4). The SNAP-Ed treatment group exposure had a significantly improved food security change from baseline to 12 months later (β = 1.2, SE = 0.4, *p* = 0.001).

Step 2: Participation status in WIC and food pantry use, but not for SNAP, 30 days prior to baseline differed (*p* < 0.01) between the intervention and control groups in Chi-square analyses (Table 1). Additionally, "change in one year participation status" (30 days prior to baseline and one year follow-up) in WIC and food pantry use differed (*p* = 0.03) between the intervention and control groups using Chi-square analysis (Table 2), but again, not for SNAP (*p* = 0.3). Logistic regression showed similar results of an association with treatment group and the potential for mediation for WIC (*p* = 0.04) and food pantry use (*p* = 0.05) but not SNAP (*p* = 0.3) (Table 2).

Step 3: Using general linear regression modeling, "change in one year participation status" in SNAP (*p* = 0.3), WIC (*p* = 0.4), or food pantry use (*p* = 0.5) were not associated with the long-term change in food security score.

Step 4: Since significant relationships were present in steps 1 and 2, multiple linear regression modeling of the relationship of treatment group and "change in one year participation status" in SNAP, WIC, and food pantries on the outcome was completed. Results showed that neither SNAP (*p* = 0.2), WIC (*p* = 0.2), nor food pantries (*p* = 0.3) were significant after treatment group was included in the model, yet treatment group remained significant (*p* ≤ 0.001).

In conclusion of research objective 1, no mediation was found between the SNAP-Ed intervention and "change in one year participation status" in SNAP, WIC, or food pantries on the change in food security score over the one year study period in the intervention compared to the control group using the Baron-Kenny causal mediation approach.


**Table 2.** Change in one year participation status comparison of SNAP, WIC, and food pantries by treatment group among Indiana SNAP-Ed participants using Chi-Square and logistic regression.

Values are counts, percentages, and *p*-values from Chi-square and logistic regression comparisons of the distributions among "change in one year food assistance participation status" between control and intervention group participants. Total numbers do not always add to sample size due to missing values and percentages do not always add to 100 due to rounding. Reference period for one year participation status covered the 30 days prior to baseline and 30 days prior to one year follow-up. \* *p* ≤ 0.05. Abbreviations: SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP, Supplemental Nutrition Assistance Program; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

#### *3.2. Research Objective 2: Test for Food Assistance Program Moderation of SNAP-Ed E*ff*ect on Food Security*

The interactions of "change in one year participation status" in SNAP, WIC, or food pantries with the treatment group did not moderate the mean difference (mean ± SEM) in food security scores in the intervention compared to the control over the one year study period using general linear regression modeling (SNAP −0.8 ± 0.4, *p* = 0.2; WIC −1.1 ± 0.5, *p* = 0.1; food pantries −1.2 ± 0.8, *p* = 0.7) (Table 3).


**Table 3.** Change in food security score over one year study period for the interaction of "change in one year participation status" and treatment group among Indiana SNAP-Ed participants using general linear regression modeling.

Least squares means were calculated using general linear regression models with change in food security as the response variable. SNAP, WIC, and food pantries were investigated in separate models including interactions with treatment group. ‡ A decrease in food security score from baseline to 1 year follow-up indicates improved food security. § Tukey adjustment for multiple comparisons in stratified analyses in each model. Interactions of each food assistance program with treatment were significant when interaction term *p* ≤ 0.05. Abbreviations: SE, Standard Error of the Least Squares Mean; SNAP-Ed, Supplemental Nutrition Assistance Program-Education; SNAP, Supplemental Nutrition Assistance Program; WIC Special Supplemental Nutrition Program for Women, Infants, and Children.

#### *3.3. Research Objective 3: Test for SNAP-Ed Program Characteristics Relationship with SNAP-Ed on Food Security*

The majority of intervention group participants (*n* = 165, 78%) received more than the minimum of four lessons with a mean of 6.8 lessons (Table 4). Approximately half of participants (*n* = 85, 57%) received lessons in a one-to-one or individualized format, followed by group (*n* = 38, 26%), and combination the two types (*n* = 25, 17%). There was no statistical evidence of an association between lesson delivery format (*p* = 0.3), the number of lessons received (*p* = 0.6), or variation between SNAP-Ed educators (*p* = 0.4) and the mean increase in food security score over time using a mixed multiple linear regression model.


**Table 4.** Evaluation of lesson delivery format, SNAP-Ed educator, and number of lessons received by Indiana SNAP-Ed Study participants on change in food security score over one year study period using mixed multiple linear regression modeling.

Lesson delivery format was reported at baseline assessment. Number of lessons was reported at the one year follow-up assessment. The control group did not receive lessons. A minimum of 4 lessons was required to have completed the intervention. Only treatment group participants were included in the mixed multiple linear regression modeling. Cells do not always add to total sample size due to missing data. § *p*-values reported for lesson delivery format and number of lessons are from the type 3 test of fixed effects. The *p*-value reported for SNAP-Ed educator is from the random effect covariance parameter estimate. Abbreviations: SNAP-Ed, Supplemental Nutrition Assistance Program-Education.

#### **4. Discussion**

The major finding from this secondary data analysis indicated an improvement in household food security among the SNAP-Ed intervention group compared to the control group regardless of participation and changes in participation in food assistance programs SNAP, WIC, or food pantries 30 days prior to baseline and one year after the intervention. The mediation and moderation analyses addressing research objectives one and two revealed that SNAP-Ed directly improved food security rather than exerting or magnifying improvement through food assistance participation or changes in participation over one year.

One previous study found greater improvements in food security among SNAP-Ed participants who also received SNAP [18] indicating that for certain populations and shorter time periods, SNAP may assist SNAP-Ed to further improve food security. However, the present results using experimental data, determined no significant difference between the treatment groups for change in food security across the four types of one year SNAP participation status. Together, previous and current study results build evidence that SNAP-Ed is effective in directly improving food security over a one year period [9].

In addition to improving food security, SNAP-Ed may have caused changes in participation status in food assistance programs throughout the study period for the following reasons. As part of the normal program delivery, SNAP-Ed educators may have encouraged and assisted intervention group participants who were not receiving food assistance at baseline to apply for financial benefits through SNAP or WIC or to maximize nutrition resources available through food pantries or other resources. On the other hand, improvements in food security directly from SNAP-Ed may have led intervention group participants who reported receiving food assistance at baseline to attain and maintain sufficient nutrition resources and withdraw participation in SNAP, WIC, or use of food pantries by the one year follow-up. Alternatively, participation in other local, state, or federal food assistance programs or resources that were not recorded in this study may have impacted food security. For example, policy, systems, environment, and other nutrition and lifestyle related resources may be influential in the success of SNAP-Ed and should continue to be investigated in the future [25]. Investigation to the reasons for changes in food assistance participation were outside of the scope of this research but present an opportunity for the future. Due to the observational nature of food assistance designation in this study, the results do not provide causal evidence of SNAP-Ed influence on changes in food assistance participation status. This limitation provides an important research opportunity, yet ethical constraints may hinder randomization of food assistance resources and require pragmatic study designs in future research [18].

In addition to finding no mediation or moderation of changes or consistency in food assistance program participation on SNAP-Ed effectiveness on food security, nutrition education program characteristics such as the number of lessons, delivery format (group or individual lessons), and SNAP-Ed educator were not associated with the magnitude of SNAP-Ed effectiveness on food security. A study describing the effect of online compared to in-person SNAP-Ed lesson delivery [26], on nutrition knowledge, intentions to change behavior, and self-efficacy, is the only previously published SNAP-Ed study to evaluate similar SNAP-Ed program characteristics. No previously published studies have addressed the question of a dose-response effect of the number of SNAP-Ed lessons on food security. In the study described herein, more than four lessons did not result in a significantly larger improvement in food security. The minimum lessons comprising SNAP-Ed guidance, four in this case, were a sufficient intervention to improve food security, reinforcing the notion that these limited lessons cover the most important behavioral recommendations for SNAP-Ed set by the USDA FNS at least in regard to food security [21]. The results suggest that participation in the minimally adherent intervention lessons is more critical to food security gains than the frequency and amount of additional time spent in lessons. Other beneficial outcomes that were not quantified here, such as sustainability of food security gains over a period longer than one year, increased nutrition knowledge, or dietary changes, may potentially be influenced by additional lessons; however, those outcomes have yet to be investigated.

The format of lesson delivery was also not significantly associated with change in food security over the one year study period among the intervention group. A current Indiana SNAP-Ed priority set forth by the USDA FNS encourages a transition to mostly group lesson delivery format rather than one-to-one format. This policy decision is supported by these study results in regard to food security improvements. Group lessons reach a greater number of participants at less cost and time, and, in this study, were as effective as individual lessons. Yet, reach to participants with special needs was not evaluated here and the provision of individual lessons may remain relevant for this group.

The third program characteristic assessed in this study, variability in one year food security score due to different educators, was not statistically significant. Variable characteristics inherent to the educator that may potentially influence outcomes include age, race, ethnicity, language, gender, education level, years of experience, depth of nutrition education knowledge, personality, knowledge and connection with community resources, among many others. These characteristics may affect the delivery and acceptance of the program to participants by potentially influencing SNAP-Ed educators' and participants' abilities to connect and relate to each other. Investigating the educator as a random effect in the model did not allow for comparisons specifically based on the educator characteristics mentioned or between specific educators yet, did allow insight to educator significance with regard to SNAP-Ed effectiveness. The study results suggest that the SNAP-Ed educators delivered a program effective at improving participants' household food security irrespective of educator.

A few studies have evaluated the impact of SNAP-Ed on food security; however, there is a paucity of SNAP-Ed literature specifically evaluating the impact of program delivery characteristics on food security outcomes [8,9,18]. A small body of literature has evaluated a second federally-supported nutrition education program, the Expanded Food and Nutrition Education Program (EFNEP) [27–30]. Since the two programs are similar in terms of aligning program goals with the Dietary Guidelines for Americans and target population, research results from EFNEP provide relevant background. Studies

evaluating EFNEP reported an increase in food security using a variety of food security measures including one survey question [27] and the 6-item [28] and 18-item [30] US HFSSM. The number of lessons needed to increase food security greatly varied across the studies. In one study, program completers (mean number of lessons 8.5 ± 0.02) compared to drop-outs (mean number of lessons 6.8 ± 0.11) showed a positive dose-response in food security with increasing number of lessons [27]. Additionally, food security was higher in participants who received lessons in a one-to-one format compared to those who received lessons in a group format or a combination of group and individual lessons [27]. In other studies, participants improved food security after receiving seven EFNEP lessons [28] or with just two or more lessons compared to a comparison group receiving one or no lessons [30]. Lesson delivery format was not always defined in these studies. The results of the present study strengthen the evidence that effectiveness of nutrition education to improve food security does not depend on the number of lessons exceeding the program completion criteria, nor format of lessons (group or one-to-one), despite the mixed results from the small body of EFNEP and SNAP-Ed literature.

Results from the present study provide a foundation for further research that improves upon some limitations, but others are presented. Treatment groups were not originally designed to test participation in singular or concurrent food assistance programs or program characteristics as main effects in the analysis. The implication of the simple randomization technique in conjunction with the large number of potential confounding characteristics presents a possibility for uneven distribution of characteristics across treatment groups, which could result in overestimation their effects. Although no significant effect was detected in this study, designing future studies to further stratify the control and intervention groups by food assistance participation status may enhance evaluation of simultaneous food assistance program participation and changes in participation and nutrition education on target outcomes. Potential for misclassification was present; however, non-response was low (baseline 8% (*n* = 27), follow-up 15% (*n* = 50)) and did not influence the results based on the sensitivity analysis, but the hesitation for some participants to answer these types of sensitive survey questions is important to consider when calculating future study sample sizes and mitigation of bias. Specifically, responses on the HFSSM were made for the entire household by one adult in the household (as per guidance [22]) and entail the reporting adult's perceptions on the other household member's food security. The 30-day reference periods before baseline and one year follow-up may not have captured all changes in food assistance. Collecting additional information on the consistency and timing of food assistance use in future studies could elucidate the temporality of the relationship between food assistance program participation, SNAP-Ed, and food security improvement. Interpretation of the results should be carefully limited to the hypothesis focused on SNAP-Ed as the main independent variable and do not inform the role of SNAP as the main independent variable on food security status.

A major strength of this study was the use of longitudinal data derived from a randomized controlled impact evaluation showing an improvement in one year food security due to SNAP-Ed [9]. Participants included in these analyses represented the greater Indiana SNAP-Ed population except for less racial diversity (89% of Indiana SNAP-Ed participants compared to 95% of study participants were non-Hispanic White; Chi-square *p* < 0.01). This difference in racial diversity is likely due to not having SNAP-Ed educators from more racially diverse geographic areas volunteer to assist with the study. Participants who withdrew from the trial were less likely to be married or living with a partner, resided in smaller households, and reported lower incomes compared to study completers [9]. The results of this study may not be generalizable to SNAP-Ed participants who have similar characteristics as the participants who withdrew from the trial and do not classify themselves as non-Hispanic white. Quantification of the change in food security score using the US HFSSM contributed a second major strength to the study. This tool is considered to be the gold standard that is used in national surveys and other research studies, permitting comparisons of results across other populations and enhancing external validity. Use of the score allows a more specific understanding of the change in food security and relationships evaluated.

#### **5. Conclusions**

This study highlights nutrition education as a critical, independent component to improving food security in the US low-income population by showing SNAP-Ed directly and sustainably improves food security with or without the presence of food assistance. Neither group, individual or mixed type lessons nor SNAP-Ed educator were related to the effectiveness of SNAP-Ed on food security. Neither were provision of lessons additional to those fulfilling SNAP-Ed guidance related to the magnitude of SNAP-Ed effectiveness. The current study results, along with previous documentation of food assistance effectiveness on food security, support a need for future investigation into the longitudinal effect of participation in multiple food assistance programs, including SNAP-Ed, to maximize improvements in food security and other USDA FNS targeted health outcomes.

**Author Contributions:** Conceptualization, R.L.R. and H.A.E.-M.; methodology, R.L.R., H.A.E.-M., H.S., Y.Z.; formal analysis, R.L.R. and H.A.E.-M.; data curation, R.L.R., H.A.E.-M., M.K.M., A.R.A.; writing—original draft preparation, R.L.R.; writing—review and editing, H.E.-M., R.L.R., H.S., Y.Z., M.K.M., A.R.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Purdue University Frederick N. Andrews Fellowship, Purdue University Nutrition Education Program, the USDA National Institute of Food and Agriculture Hatch Project grant (IND030489), a North Central Nutrition Education Center of Excellence grant (IND030473G), a U.S. National Library of Medicine training grant (T15LM012502), and the University of Kentucky Center for Poverty Research (UKCPR) through funding by the U.S. Department of Agriculture, Food and Nutrition Service (AG-3198-S-12-0044). The opinions and conclusions expressed herein are solely those of the author(s) and should not be construed as representing the opinions or policies of the UKCPR or any agency of the Federal Government.

**Acknowledgments:** The authors would like to thank Bruce Craig from the Purdue Department of Statistics for providing oversight for the analyses.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Review* **Fruit and Vegetable Incentive Programs for Supplemental Nutrition Assistance Program (SNAP) Participants: A Scoping Review of Program Structure**

#### **Katherine Engel and Elizabeth H. Ruder \***

Rochester Institute of Technology; Rochester, NY 14620, USA; katherine.engel4@gmail.com **\*** Correspondence: ehrihst@rit.edu; Tel.: +1-585-475-2402

Received: 29 April 2020; Accepted: 1 June 2020; Published: 4 June 2020

**Abstract:** The low intake of fruits/vegetables (FV) by Supplemental Nutrition Assistance Program (SNAP) participants is a persistent public health challenge. Fruit and vegetable incentive programs use inducements to encourage FV purchases. The purpose of this scoping review is to identify structural factors in FV incentive programs that may impact program effectiveness, including (i.) differences in recruitment/eligibility, (ii.) incentive delivery and timing, (iii.) incentive value, (iv.) eligible foods, and (v.) retail venue. Additionally, the FV incentive program impact on FV purchase and/or consumption is summarized. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for scoping reviews, a search of four bibliographic databases resulted in the identification of 45 publications for consideration; 19 of which met the pre-determined inclusion criteria for full-length publications employing a quasi-experimental design and focused on verified, current SNAP participants. The data capturing study objective, study design, sample size, incentive program structure characteristics (participant eligibility and recruitment, delivery and timing of incentive, foods eligible for incentive redemption, type of retail venue), and study outcomes related to FV purchases/consumption were entered in a standardized chart. Eleven of the 19 studies had enrollment processes to receive the incentive, and most studies (17/19) provided the incentive in the form of a token, coupon, or voucher. The value of the incentives varied, but was usually offered as a match. Incentives were typically redeemable only for FV, although three studies required an FV purchase to trigger the delivery of an incentive for any SNAP-eligible food. Finally, most studies (16/19) were conducted at farmers' markets. Eighteen of the 19 studies reported a positive impact on participant purchase and/or consumption of FV. Overall, this scoping review provides insights intended to inform the design, implementation, and evaluation of future FV incentive programs targeting SNAP participants; and demonstrates the potential effectiveness of FV incentive programs for increasing FV purchase and consumption among vulnerable populations.

**Keywords:** incentive programs; Supplemental Nutrition Assistance Program (SNAP); fruits and vegetables; low-income; farmers' markets; dietary quality; produce intake; produce purchasing

#### **1. Introduction**

Eating sufficient amounts of fruits/vegetables (FV) is vital for a healthy dietary pattern associated with a lower risk of cardiovascular disease and certain cancers [1]. However, Americans do not consume enough FV; only 12.2% and 9.3% of US adults meet the *2015–2020 Dietary Guidelines for Americans*' recommendations for daily fruit and vegetable consumption, respectively [2]. Among Americans, lower income groups consume less FV than higher income groups, and this is a key socioeconomic disparity in overall dietary

quality [2–4]. Thus, it is important that low-income participants in federal food assistance programs in the United States, such as the Supplemental Nutrition Assistance Program (SNAP), have access to these foods.

SNAP is the largest federal food assistance program in the United States. It functions by providing participants with food purchasing resources in the form of an electronic benefit transfer (EBT, an electronic system that allows a recipient to authorize transfer of their government benefits from a federal account to a retailer) on a monthly cycle. Although SNAP eligibility requirements vary from state to state, households that are SNAP eligible have gross incomes of less than 130% of the federal poverty line [5]. Unlike other U.S. food assistance programs, like the Supplemental Nutrition Assistance Program for Women, Infants, and Children (WIC), SNAP benefits can be used for most food products with few exceptions (such as hot foods and foods that are intended to be eaten in stores) [6,7]. In contrast to SNAP, WIC benefits are limited to foods such as milk, cheese, yogurt, FV, canned fish, tofu, breakfast and infant cereal, whole wheat breads and grains, eggs, peanut butter, infant formula, and jarred baby foods [8]. Thus, although SNAP plays an integral role in ensuring that millions of people have the resources they need to access sufficient amounts of food, it lacks specific restrictions that dictate the nutritional quality of foods that participants can purchase. Importantly, it has been shown that individuals who receive SNAP benefits have poor diets relative to the overall population and other income-eligible non-participants [3]. In some cases, SNAP participation has been associated with negative health outcomes and inversely correlated to self-assessed health status [9]. Given the evidence that WIC participation is associated with health benefits [10], one proposed alteration to the SNAP program is creating restrictions around which foods can be purchased with benefits. However, key constituencies, ranging from members of U.S. Congress to hunger relief organizations, have rejected these proposals for reasons including concerns about limiting participants' ability to exercise autonomy in food choice and administrative burdens [11]. Moreover, restrictions on the types of food eligible for SNAP could contribute to worsening food security in areas where a variety of healthful foods is not sold by food retailers. Another alteration to the SNAP program that has been suggested is FV incentives, which provide participants with considerable autonomy in deciding what foods to purchase [12].

FV incentives include a variety of inducements to offer low-income participants funds to purchase these foods. They are potentially appropriate for improving dietary quality, because they are a tool for facilitating behavior change. The theory that incentives serve as a strategy for inducing changes in behavior centers on the standard direct price effect [13]. The standard direct price effect makes the incetivzed behavior more attractive by providing a financial reward. As a result, incentives have the capacity to instill new, positive habits, as well as end pre-existing, negative habits. Thus, when applied on a large enough scale, incentives may have the ability to shift cultural norms [13]. Incentives may be particularly useful for promoting healthy behaviors, such as consuming more FV, because the benefits of healthy behaviors are often uncertain and delayed, while the cost of these behaviors is immediate. Consumers tend to value current costs and benefits more than future costs and benefits, which in turn can lead to choosing not to engage in healthy behaviors, since the present value of these behaviors is low. Incentives create an immediate benefit because they lower the cost of healthy foods for consumers. By creating short-term rewards for healthy behaviors, incentives serve to make these behaviors more appealing by increasing their present value [14]. In general, the cost of food plays a critical role in how people make food choices [15,16]. Glanz and colleagues [17] found that behind taste, price is the second most important influence on food choice. For SNAP participants specifically, it has been demonstrated that the cost of healthy foods is a barrier for improving dietary quality [18,19]. Incentives expand the financial resources participants have available to purchase healthy foods, and thus address the barrier that the cost of these food poses to dietary quality [18–20].

FV incentive programs have been designed and implemented for a number of different populations, including WIC and SNAP participants, and venues, such as farmers' markets and grocery stores. In addition,

the types and value of incentives that have been developed vary widely from point-of-sale (POS) discounts to coupons, vouchers, and tokens. A preliminary search for existing scoping reviews on this topic was conducted by searching the Cochrane Database of Systematic Reviews, Google Scholar, ProQuest, PubMed, and Sage Journals Online. No scoping reviews on this topic were identified. Given the emerging evidence related to FV incentive programs among SNAP participants and the diversity of structural factors within these programs, a scoping review was selected as the appropriate method. The objective of this scoping review is to characterize the factors in program structure which may impact the effectiveness of incentive programs. The scoping review research question is, "What are the differences in structural factors, including recruitment and eligibility criteria, delivery and timing of incentives, financial value of incentives, foods eligible for incentive redemption, and type of retail venue reported among FV incentive programs?" Finally, this review summarizes the outcomes of existing FV incentive programs with respect to the purchase and/or consumption of FV among SNAP participants, with specific attention to the quality of the assessment methods for FV purchase and/or consumption. This work provides insight intended to inform the design, implementation, and evaluation of future FV incentive programs targeting SNAP participants.

#### **2. Materials and Methods**

A scoping review was undertaken to systematically synthesize factors in program structure which may impact the effectiveness of FV incentive programs. This review was conducted as per the Arksey and O'Malley framework for scoping reviews [21] and integrated with the guidance from the Joanna Briggs Institute (JBI) [22] and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for scoping reviews [23]. A protocol document is publically available online at: https://figshare.com/articles/Protocol\_Document\_pdf/12380669, and a completed PRISMA ScR checklist is included as Supplementary Table S1.

#### *2.1. Search Strategy*

Focused searches were conducted by one author (K.E.) using Google Scholar, ProQuest, PubMed, and Sage Journals Online. The search terms that were used include "SNAP incentives," "WIC incentive," "food benefits incentive," and "food assistance incentive." Results were limited to English language publications and indexed up to 7 November 2019. In addition to the use of these search terms, papers were identified by examining the articles cited by the papers found in the preliminary search.

#### *2.2. Study Selection*

Full-text articles identified in the search were imported into Mendeley reference management software and duplicates were manually removed. In total, 45 unique publications were identified and both authors independently reviewed the full-text documents for pre-determined inclusion/exclusion criteria. The inclusion criteria included: full-length publication in a peer-reviewed journal or government report, quasi-experimental design, and targeted focus on verified, current SNAP participants studies that solely examined the use of FV vouchers as part of the WIC foods package were excluded for the following reasons: (1.) FV vouchers became a standard part of the WIC Food Package following a final rule published in May 2014 [24] and (2.) WIC FV vouchers can only be used for FV and therefore are not used to incentivize the purchase of FV over other foods within the WIC Food Package. The authors conferenced regularly to ensure agreement and talked through any inconsistencies. Due to the relative lack of research on this topic, papers were not excluded based on their publication date.

#### *2.3. Data Charting*

Data were extracted from eligible papers into a standardized Google Doc chart developed by both authors. The two authors independently charted the data, discussed the results and continuously updated the data collection chart. The data ultimately collected included: study authors, year of publication, study objective, population and sample size, methodology, incentive program structure characteristics (participant eligibility and recruitment, delivery and timing of incentive, foods eligible for incentive redemption, type of retail venue), and study outcomes related to FV purchases and FV consumption. In addition, study methods for the assessment of FV purchase and/or consumption were charted. The charted data were summarized as counts where applicable.

#### **3. Results and Discussion**

Of the 45 publications initially reviewed, *n* = 6 were excluded for not falling within the scope of the review, *n* = 8 were excluded for not employing a quasi-experimental design, *n* = 6 were excluded because participation was not focused on current, verified SNAP participants, and *n* = 2 were excluded for being solely related to WIC FV vouchers prior to the implementation of the 2014 WIC Food Package. In addition, two poster presentations were excluded and two publications were excluded because they presented preliminary data that was included in a subsequent publication. In total, 19 publications were included in the final review (Table 1).

#### *3.1. Incentive Program Structure*

A variety of types of incentive programs have explored approaches for increasing the purchase and consumption of FV by SNAP participants. The following section details ways in which eligible individuals become participants in incentive programs, the delivery and timing of incentives, and differences in the financial value of incentives to participants.


 and enabling

Supplemental

 Nutrition Assistance

**Table 1.**

Summary of studies on fruit and vegetable incentives as an approach for encouraging




*Nutrients* **2020**, *12*, 1676


FV: fruits and vegetables, WIC: Special Supplemental

 Nutrition Program for Women, Infants, and Children, POS: point of sale.

#### 3.1.1. Recruitment and Eligibility of Incentive Program Participants

In the studies under review, individuals became incentive program participants in a multitude of ways. Eleven programs had an enrollment process through which individuals had to complete some type of informal or formal sign-up process for the program to receive the incentive [7,26–28,31–33,36,38,40,41], while eight studies had no enrollment process and provided the incentive when participants visited and/or made a purchase at a retailer and provided evidence of their SNAP participation [25,28,30,34,35,37,39,42]. Bowling et al. provided all SNAP participants shopping at participating markets with an incentive and provided an additional incentive to a subset of this population that had specifically enrolled in the program [27]. It is important to note that the inclusion of an enrollment process may create additional administrative challenges, as well as barriers for participation. However, as enrollment processes often included a pre-test survey and/or a method of tracking participants' transactions throughout the implementation period, these programs may provide opportunities for more rigorous evaluation and therefore greater insight regarding the impact of incentive programs of FV purchases and consumption. One study assessed programs in which participants were given the incentive after visiting a health clinic [33]. Similar to the challenges with enrollment processes, this requirement may create a participation barrier, but may have a greater impact on dietary quality and health, as it is part of a broader focus on the health status of federal food assistance program participants.

#### 3.1.2. Delivery and Timing of Incentive Benefits

Table 1 summarizes the types of incentive benefits that have been granted to participants. Two programs were structured such that the incentive was provided at the point-of-sale (POS) [7,31]. For the purposes of this review, POS incentives are defined as those that immediately discount participants' FV purchases at checkout. In contrast to this model, 17 programs provided participants with coupons, vouchers, or tokens [25–30,32–42]. The delivery of coupons, vouchers, or tokens (hereafter referred to as incentives) varied by program. Some programs provided the incentives when the participant enrolled in the program or following their enrollment, such as when they visited a farmers' market [28,32,35,41]. Other programs provided the incentives following or in conjunction with the purchase of FV [7,25,26,28,30,31,33,36–38,40]. Moreover, some programs required incentives to be redeemed immediately upon receipt [7,31], but others allowed the incentive to be used for a future transaction. Importantly, allowing participants to save the incentive and choose when they use it may be beneficial, due to the monthly "SNAP-cycle" spending pattern, where the majority of recipients spend most of their monthly benefits within two weeks after receiving them [43,44]. In all cases, the intent of these benefits is to induce participants to increase their FV purchases by providing them with financial rewards for these purchases and/or resources that enable them to purchase these foods at a lower price.

Most incentive programs included in this review required participants to make an FV purchase in order to "trigger" the delivery of the incentive benefit [7,25,26,28,30,31,33,36–38,40]. However, the types of FV that qualify as trigger foods differ. For example, the Healthy Incentives Pilot, a federally funded FV incentive program administered in Hampden County, MA, distributed incentive benefits after participants purchased targeted FV, which were defined as any fresh, canned, frozen, and dried fruit or vegetable FV without any added sugars, fats, oils, or sodium. In addition, the pilot excluded fruit juice, mature legumes, and white potatoes. These specifications were selected to mirror the restrictions of WIC-eligible produce items [7]. In contrast, for programs held at farmers' markets, participants often had to purchase fresh FV in order to receive the incentive.

A few programs had multiple points and locations at which incentive benefits were distributed to participants. In the program evaluated by Young et al. [42], a \$2 bonus incentive coupon was provided for every \$5 in SNAP benefits used at a farmers' market. Additionally, coupons were distributed at community

organizations that serve SNAP-eligible populations, absent of any initial purchase by the participant. The program examined by Olsho et al. [34], was structured similarly, in that some participants received the incentive through a match after they made a purchase, while others received the incentive from community-based organizations absent of any purchase, usually after they attended a nutrition workshop or other health and fitness program. Similarly, two incentive distribution methods were employed in the program examined by Savoie-Roskos et al. [39]; participants received either "regular incentives", which were distributed at regular intervals without any purchase requirement, or matched incentives. Bowling et al. [27] employed both POS incentives and tokens; all SNAP recipients shopping at participating markets received a matched incentive when they used their EBT card at these markets, which could not be saved for future use, but at every third market, participants also received \$20 in "Bonus Buck" tokens.

#### 3.1.3. Financial Value of Incentive to Participants

The value of incentive benefits to participants differed widely. As stated previously, most programs required a purchase to receive the incentive [7,25–27,30–32,34,36–38,40,42]. In these programs, the value of the benefit was either pre-determined or determined by the value of the participants' purchases. For example, in the incentive program studied by Freedman et al. [31], participants received benefits valued at \$5 regardless of the cost of their initial purchases. However, many incentive programs functioned such that the value of the benefit was determined by the magnitude of the participants' spending [7,25–27,29,30,32,34,36–40,42]. In these cases, the value of the benefit was either equal to the participants' spending or a percentage of their spending. In many instances, 100% of the participants' spending was matched, meaning that the value of the benefits was equal to the amount of money spent by participants [25,26,30,32,36–40]. In some cases, the value of the benefit was adjusted based on the size of participating families, as families with children were given additional value [38].

Among incentive programs that provided a match to participant spending, there was frequently a ceiling on the value of the match. In all, 11 of the studies [7,25–28,30,32,36–38,41] reviewed had some type of ceiling. For instance, in a Utah-based farmers' market incentive program, participants received \$1 in incentives for every \$1 they spent in SNAP benefits, with individuals and couples receiving \$10 worth of incentives each week and families receiving an additional \$5 per child, up to \$30 each week [39]. Another type of ceiling was demonstrated, where participants could receive an extra \$20 in bonus tokens every third farmers' market visit but were limited to receiving \$120 of these bonus tokens during the program's implementation [27]. Other programs provided benefits that were valued as a percentage of the participants' spending. For instance, the Health Bucks and Philly Food Bucks programs provided \$2 vouchers for every \$5 participants spent, and thus acted as a 40% match of the participants' spending [32,42,45]. Notably, incentives that are granted in proportion to the participants' spending are designed to encourage participants to purchase more fruits and vegetables, because with these programs, the more participants spend on these foods, the more they are rewarded.

Some programs implemented multiple forms of incentives. For example, Savoie-Roskos et al. [39] provided one group of participants with incentive benefits that did not require them to make a purchase and another group with benefits, in the form of spending matches, that augmented the incentive. A comparison of the outcomes between the groups was not reported.

Overall, programs that match participants' spending may provide incentive benefits that have greater financial value than those that provide a benefit of a fixed value. In addition, programs with ceilings may create less value for participants than those without ceilings. Thus, certain programs may be more effective in inducing participants to purchase and consume more FV, because they expand participants' purchasing power to a greater degree. Differences in value may also be important, given that program retention is a challenge across the literature, and programs that provide less value may be less effective in encouraging ongoing participation. Notably, Wetherill et al. [41] posited that low incentive redemption rates may be tied to perceived differences in the value of different kinds of incentives, as incentives that function as discounts and expand buying power may be less valuable to participants than incentives that provide free products.

#### 3.1.4. Eligible Foods

The foods that were eligible for purchase using incentive benefits also differed. While some of the programs provided benefits that could be utilized to purchase only FV, others were triggered by an FV purchase, but provided benefits redeemable for a diverse range of foods, such as any SNAP-eligible food [7,36,39]. A drawback of awarding incentives that can be used for any SNAP-eligible food is the possibility that incentive benefits are used to purchase foods with low nutrient density. For example, in the Healthy Incentives Pilot, an additional \$0.30 was added to participants' EBT cards for every \$1 of SNAP benefits spent on FV. There are few limitations on the types of foods and beverages that can be purchased with EBT, and some evidence suggests that reducing the price of healthful foods may result in the increased purchase of energy, which could contribute to obesity [46].

In contrast, other FV incentive benefits could be used only to purchase locally grown FV [27,29,31, 32,34,37,40,47]. In some cases, the foods included in the incentive program varied based on participant eligibility. For example, in the Fresh Funds program [36], participants could use the tokens they purchased with their SNAP benefits, and the tokens that they received as a match, to purchase fresh produce or packaged foods, such as jams/spreads, breads, eggs, pasta, cheese, and fish; however, the tokens they purchased using WIC benefits could only be spent at vendors selling fresh produce.

The characteristics and needs of the recipients must be considered when designing incentive programs and the types of FV eligible for incentive redemption. For example, FV may not be an appealing incentive to people with limited facilities and equipment for food preparation. However, for participants with access to food preparation facilities, frozen, canned, and dried FV have a longer shelf-life than fresh FV and may be useful for prolonging food security throughout the month and between monthly SNAP benefit distributions. Likewise, the capacity of the retail environment must also be considered. SNAP vendor eligibility implemented in January 2018 requires vendors to stock FV, but does not require those FV to be fresh if other perishable foods are stocked (i.e., meat or dairy), and only one type of perishable food needs to be offered (i.e., selling just one type of fruit would fulfill the fresh FV requirement) [48]. Low income communities tend to have more convenience stores and small markets [49,50] where the availability of FV tends to be lower [51–53]. Therefore, the retail capacity, including the availability of freezers/refrigeration, must be considered when designing fresh FV incentive programs.

#### 3.1.5. Retail Venue

Table 1 illustrates that the majority (16/19) of the reviewed studies were implemented in part or in entirety at farmers' markets. Farmers' market incentive programs have the advantage of supporting local farmers and food vendors. Additionally, the literature indicates that shopping at farmers' markets positively impacts FV purchases and that by drawing participants to shop at these venues, incentive programs implemented at farmers' markets may positively impact FV purchase and consumption behaviors and attitudes beyond the time period in which the program is implemented [31,32,54].

Several studies indicate that farmers' market incentive programs attract SNAP participants who otherwise might not shop at these venues [30–32,35]. One incentive program study found that 57% of participants in a farmers' market incentive program had never been to a farmers' market [31]. Similarly, another study noted that SNAP participants' awareness of farmers' markets rose in relation to their exposure to the Health Bucks incentive program [35]. In addition, these researchers found that 54% of Health

Bucks participants who used their benefits at farmers' markets strongly agreed that "I shop at farmers' markets more often because of Health Bucks", and a Utah-based incentive program reported that 98% of baseline participants reported that the incentive made it more likely that they would shop at the farmers' market [30]. In the Farmers' Market Fresh Fund Incentive Program, 82% of participants had never attended a farmer's market prior to participating in the program, and 93% of participants reported that incentives were "important" or "very important" in their decision to shop at farmers' markets [32]. In addition to drawing more SNAP participants to farmers' markets, the Farmers' Market Fresh Fund Incentive Program demonstrated the potential to impact participants' long-term shopping behavior. In particular, the majority of participants reported that they would be "somewhat likely" or "completely likely" to shop at farmers' markets even without the continuation of the incentive program [32]. Increased awareness that EBT is accepted at many farmers' markets has also been noted among incentive program participants [39]. Accordingly, there is evidence that farmers' market incentive programs increase participants' exposure to markets as venues offering affordable, healthy food, and in turn have the potential to positively influence their long-term food purchasing behavior.

Another potential benefit of implementing incentive programs at farmers' markets is the potential for the increased consumption of FV. Shopping at farmers' markets is linked to increased FV consumption, and thus offering incentives at farmers' markets has the capacity to improve dietary quality beyond merely increasing the financial resources participants have to purchase FV [55]. Specifically, Olsho et al. [34] found that both incentive program participants and farmers' market shoppers who were not enrolled in the program reported higher FV consumption than other residents in their neighborhoods. However, incentive program participation per se was not related to an increase in daily FV servings.

Despite the potential benefits of implementing incentive programs at farmers' markets, it is important to consider access issues in this context. Specifically, farmers' markets are not as abundant as other types of food retailers, such as grocery stores, and may not exist in certain communities. Driving distance from residence to market has been inversely correlated with repeat use of farmer's market incentives [56]. However, other research suggests that the distance from food retailers does not significantly affect the extent to which incentives impact SNAP participants' FV spending [7,32]. Moreover, many markets are not open year-round and have limited hours of operation.

#### *3.2. Outcome Assessment*

All of the studies included in this review considered the impact of incentives on FV purchases and/or consumption. Four of the studies reviewed focused exclusively on FV purchases [25,30,36,39,41], five focused exclusively on FV consumption [28,29,34,37,40], and nine examined both FV purchase and consumption [7,24,26,27,31–33,35,38]. As explored later in this section, only one of the reviewed studies [41] did not report some positive impact on FV purchases and/or consumption in conjunction with incentive programs.

Studies employed a variety of approaches for measuring these outcomes, as shown in Tables 2 and 3. The majority of studies assessed FV purchase and/or consumption using a pre-/post-test design, where participants' FV purchase and consumption behaviors and attitudes were assessed prior to the implementation of the incentive program and then again at the program's conclusion [25–28,30–36,38,39,42]. In addition, the Healthy Incentives Program also assessed the program impact at various points throughout the implementation phase [7]. A few studies also used control or quasi-control groups to assess program impact [7,34,37,41,47]. The merits of quasi-control groups are somewhat limited if the comparison groups do not share important characteristics with the incentive program participants. For example, Olsho et al. [34], compared the FV purchase and consumption of incentive program participants with that of other non-participant neighborhood residents, but these residents were not necessarily federal food assistance

program participants. Although the groups may have shared relevant demographic characteristics, the comparison is problematic because federal food assistance program participants have unique circumstances that may make incentive programs particularly salient, such as the challenge of managing food-purchasing resources in conjunction with the monthly SNAP distribution cycle. Another study compared SNAP transaction data from participating grocery stores to that of nonparticipating stores to determine whether the percentage of dollars spent on fresh produce in total SNAP transactions is higher in stores that implement incentives than in stores that do not [37]. Control stores were selected using a coarsened exact matching and linear probability match to match on store characteristics and sociodemographics. While this approach is not as rigorous as randomizing stores to the incentive or control condition, the use of matched controls is preferred to non-matched controls. Wetherill et al. [41] employed a quasi-experimental design to compare two coupon interventions: basic information and plain coupon distribution compared to tailored, targeted marketing coupon intervention. However, low coupon redemption by either group made comparisons difficult.


**Table 2.** Assessment of fruit and vegetable purchases in nutrition incentive programs.



#### **Table 3.** Assessment of fruit and vegetable consumption in nutrition incentive programs.


Assessment of FV consumption varied in the quality of methods used for dietary assessment. Of the 15 studies which assessed the change in FV consumption (Table 3), five studies employed the validated Behavioral Risk Factor Surveillance System FV module [57], and two other studies used other validated assessment tools [7,29]. The validity of the dietary assessment methods for the remaining eight studies was not clear.

All studies under review noted some degree of positive impact, with the exception of Wetherill et al. [41]. In that study, participants were all recipients of Temporary Assistance for Needy Families (TANF), a cash-assistance program in the United States for very low-income families with children. Generally, TANF participation automatically qualifies a household for SNAP. Given the severe income restriction of TANF households, these participants may not be representative of the general SNAP population. Moreover, few participants in this study (*n* = 16, 6.3%) redeemed the incentive coupons; making outcome assessment difficult, although the authors did note that that education surrounding food preparation skills may be necessary, in conjunction with incentives to alter food purchasing behaviors at farmers' markets. Among the remaining studies that reported some positive impact, limitations in the impact of the incentive programs were identified. Conclusions from the Michigan farmers' market-based Double Up Food Bucks program included that the impact of incentive programs was unsustainable and minimal [40], and a Washington, DC-based farmers' market evaluation noted that although participants reported higher FV consumption, their intake still fell below recommended levels [35]. Olsho et al. [34], reported an increase in purchases but concluded that there was no observable difference in consumption between incentive program participants and non-participants, and similarly, Bowling et al. [27] observed that while participants reported increased fruit and vegetable consumption, they did not change the amount of their WIC/SNAP budget spent on these foods.

Several demographic factors have been linked to incentive program retention and use frequency and thus may be important when considering outcome assessment. Specifically, Dimitri et al. [29] noted that participants who were more reliant on food banks, very income restrained, and lived in areas where access to food was limited were more likely to drop out of the incentive program they studied [29]. These findings suggest that the presence of these factors may impact the effectiveness of incentive programs, as participant retention is essential for incentives to influence FV purchases and consumption. Ratigan et al. found that participants who had unhealthier diets at the beginning of the program were more likely to use incentives a greater number of times in the short term, but incentive use waned after six months [36]. In addition, Ratigan et al. noted that elderly and disabled individuals were more likely to use incentive programs in the long term than those who were younger and noted that ethnicity, type of government food assistance program participation, income, season of incentive program enrollment, and baseline FV consumption were related to the frequency of incentive utilization and total duration of their retention in the incentive programs [36]. They also noted that ethnicity, type of government food assistance program participation, income, season of incentive program enrollment, and baseline FV consumption correlate with both the number of times participants utilize incentives in a given period of time, as well as the total duration of their retention in incentive programs. Together, these results suggest that additional work is needed to identify the characteristics of subgroups who are most responsive to incentive programs in order to target incentive programs.

#### **4. Conclusions and Recommendations**

This scoping review highlights the wide range of FV incentive program structures and demonstrates that, in general, these programs may be an effective approach for increasing FV purchase and consumption by SNAP participants, while preserving autonomy in food choice. However, it is unclear whether the potential positive effects of these programs are substantial and sustainable. Moreover, the assessment methods employed to evaluate these programs have often relied on self-reports and lack sufficient rigor to assess program impact. Specifically, dietary assessment, when performed, frequently failed to utilize validated methods, such as the 24-h recall. Additionally, there are limitations to the scoping review process itself. Namely, a professional librarian was not consulted to assist with developing the search strategy and the protocol was not published early enough in the process to allow input from the greater

scientific community. These factors, in conjunction with the significant variation in program structure, makes it difficult to elucidate which programmatic elements may be most critical for designing and implementing effective programs.

Although the literature indicates that incentive programs may positively impact FV purchases and consumption by SNAP participants, several areas that require additional research in order to understand how to create effective programs are revealed. For instance, other interventions, such as nutrition education, cooking demonstrations, and food tastings, are often deployed in conjunction with incentives. These interventions not only equip participants with the knowledge they need to make healthy eating decisions and integrate healthy foods into their diets but may also contribute to participant use and retention. Participant use of available incentives and retention is a key determinant of program effectiveness, and additional research is needed to understand how to maximize participation and retention. Additional work is needed to elucidate how participant characteristics, such as food security status and demographics, may be associated with the use of incentives. Another area for future research is the impact of incentive program participation on objective measures of health. The studies reviewed in this scoping review demonstrated an improvement in program participants' perceptions of their health [7,32,33,36] and one study demonstrated an improvement in food security status [30], but none of the studies under review measured BMI or other health measures. Consequently, the actual impact of incentives on health remains unclear, and identifying the point at which incentives create a tangible difference in health outcomes is key for creating programs that promote participants' well-being. Lastly, more research is needed to understand the long-term effects of incentives. Some evidence suggests that the increases in FV purchases resulting from incentive program participation are not sustained following program termination [40]. Additionally, few studies have investigated the capacity of incentive programs to influence long-term food consumption and purchasing behavior. Thus, the long-term efficacy of incentives is uncertain [38,58]. Moreover, the research that has considered the long-term impacts of incentive programs has relied on self-reported predictions of future food purchasing behavior [32]. As no longitudinal studies of the impact of incentive programs have been performed, additional research is required to determine the long-term impact of these programs.

Overall, studies of FV incentive programs reveal a positive impact on both FV purchases and consumption. This scoping review provides insights intended to inform the design, implementation, and evaluation of future FV incentive programs targeting SNAP participants. Exploring these factors is critical for understanding how to effectively design and implement effective, sustainable incentive programs.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2072-6643/12/6/1676/s1, Table S1: Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRSIMA-ScR) Checklist.

**Author Contributions:** Conceptualization, K.E. and E.H.R.; methodology, K.E. and E.H.R., formal analysis, K.E. and E.H.R.; writing—original draft preparation, K.E. and E.H.R.; writing—review and editing, K.E. and E.H.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This manuscript grew from Katherine Engel's Thesis in Science, Technology and Public Policy at the Rochester Institute of Technology. The authors would like to thank Kathryn Faulring and Claire Cook for administrative support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**





© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Nutrients* Editorial Office E-mail: nutrients@mdpi.com www.mdpi.com/journal/nutrients

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18