Food Security Status and Produce Intake and Behaviors of Special Supplemental Nutrition Program for Women, Infants, and Children and Farmers’ Market Nutrition Program Participants

Methods 

The Institutional Review Board at Ohio University and the Ohio Department of Health approved this study prior to data collection. Surveys were mailed to the female head of household or adult woman managing the communication with the WIC office of all households participating in the Athens County, OH WIC program (n=1,076) in November 2005 (Farmers’ Market Nutrition Program vouchers had expired on October 31, 2005).

Two survey versions were used, one for those enrolled in WIC alone (WIC group) and one for enrolled in both WIC and Farmers’ Market Nutrition Program (Farmers’ Market Nutrition Program group). Both contained demographic questions, as well as previously validated survey tools [US Household 18-item Food Security Survey Module, to measure the level of household food security status (23), a 13-item Tool to Assess Psychosocial Indicators of Fruit and Vegetable Intake in Low Income Communities (22), a 7-item Food Behavior Checklist for a Limited Resource Audience (24), a 1-item perceived health question (25), and a 7-item measure of social capitol (26)]. The survey sent to Farmers’ Market Nutrition Program participants also included close-ended questions regarding Farmers’ Market Nutrition Program participation, satisfaction, and behavior. This report summarizes the food security, fruit and vegetable intake and behavior, and education level data.

A cover letter from the Director of the Athens County WIC program was sent with the surveys to describe the purposes and voluntary nature of the study, along with a postage-paid return envelope. Surveys were labeled and mailed from the Athens County WIC office by authorized WIC staff only, in order to ensure client confidentiality. No follow-up phone calls or postcards were sent, and participants were not compensated.

Data were entered into the Statistical Package for the Social Sciences (SPSS version 13.0, 2004, Chicago, IL). Missing data and nonapplicable responses were scored as missing data. A P value <0.05, set a priori, was considered statistically significant. Results should be considered exploratory in nature as P values are not adjusted for multiple testing. The χ2 test for independence was conducted when multiple answers to survey questions were possible to determine if group status was independent from response. With quantitative responses, independent sample t tests were performed to determine if a difference in means existed for WIC compared to Farmers’ Market Nutrition Program (two-sided P value reported.) Lastly, Spearman correlation was used for quantitative associations between two variables.

Results 

At the time of the study, 1,076 households (representing 1,742 Athens County residents) were receiving WIC benefits, with 23% (n=246) choosing to participate in Farmers’ Market Nutrition Program. Of those surveyed, 235 surveys were returned (22% overall response rate; Farmers’ Market Nutrition Program Group, n=65, 26.4% response rate; WIC Group, n=170, 20.4% response rate). Sample size is not consistent for all variables, however, because not all respondents answered all survey questions.

Table 2 summarizes participant characteristics. Of the single women with children, 51.5% (n=50) had another adult, over the age of 18 years, living in the household. Overall, average household size was 3.69±1.30 members. No differences (P>0.05) were noted between groups for characteristics measured.

Table 2. Participant characteristics by program participation group for women living in households participating in Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) Participation Alone or in Both WIC and Farmers’ Market Nutrition Program (FMNP) (WIC/FMNP)a

	Characteristic	Both Groups (n=235)		WIC (n=170)		WIC/FMNP (n=65)
		Frequency	%	Frequency	%	Frequency	%
	Marital status
	Married	130	55.3	91	53.5	39	60.0
	Single/never married	77	32.8	59	34.7	18	27.7
	Divorced	18	7.7	13	7.6	5	7.7
	Separated	9	3.8	6	3.5	3	4.6
	Pregnant	34	14.5	25	14.7	9	14.1
	Breastfeeding	25	10.6	14	8.3	11	17.2
	Married with children	122	51.9	84	49.4	38	58.5
	Singleb women with children	97	44.2	72	43.4c	25	38.5
	No other adult in the house	47	48.5	34	47.2	13	52.0
	Other adult in the house	50	51.5	38	52.7	12	48.0

a Variation in number of subjects is due to missing data. b Single includes women who are divorced, separated, or never married. c n=166, rather than 170, for subjects with both marital and number of children data.

Although ethnicity of respondents was not measured, the Athens County WIC office served mostly non-Hispanic white (93%) individuals in March 2006, along with African-American (3.9%), Asian (1.2%), and Hispanic (0.09%) individuals (Heidi Anderson, personal communication, March 2006). It is likely that our sample represented a similar ethnic profile.

Women participating in Farmers’ Market Nutrition Program reported higher levels of education. Almost all Farmers’ Market Nutrition Program respondents (95.2%, n=63) reported having at least a high school diploma or equivalent (general equivalency diploma), while 84% (n=169) of WIC respondents reported having a high school diploma or general equivalency diploma (P=0.027).

Table 3 summarizes the food security status of participant households (n=228). Groups did not significantly differ (χ2=2.117, P=0.548).

Table 3. Household food security and food security among children status of women living in households participating in Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) Participation Alone or in Both WIC and Farmers’ Market Nutrition Program (FMNP) (WIC/FMNP)

	Household Food Security
	Food security status level	Both Groups (n=228)a		WIC (n=167)		WIC/FMNP (n=61)
		Frequency	%	Frequency	%	Frequency	%
	Food security	61	26.8	48	28.7	13	21.3
	At risk for (marginal) food insecurity	47	20.6	34	20.4	13	21.3
	Low food securityb	75	32.9	51	30.5	24	9.3
	Very low food securityc	45	19.7	34	20.4	11	18.0

	Food Security among Children
		Both Groups (n=231)a		WIC (n=169)		WIC/FMNP (n=62)
		Frequency	%	Frequency	%	Frequency	%
	Food security among childrend	182	78.8	134	79.3	48	77.4
	Low food security among childrend	48	20.8	35	20.7	13	21.0
	Very low food security among childrend	1	0.4	0	0	1	1.6

a Varying participant numbers for food security and hunger measures are due to missing responses for some questions. b Low food security was previously categorized as “food insecure without hunger” (2, 23). c Very low food security was previously categorized as “food insecure with hunger” (2, 23). d Child food security status has been categorized into a dichotomous variable previously (2, 23), with households being categorized as “insufficient evidence of hunger/very low food security among children” or “clear evidence of hunger/very low food security among children,” based upon 0-4 and 5-8 positive responses, respectively, to the child items of the food security survey module (23). In this study, the food security among children, low food security among children, and very low food security among children groups were categorized, based upon 0-1, 2-4, and 5-8 positive responses, respectively, to the child items of the food security survey module (23).

Daily vegetable servings for women from the Farmers’ Market Nutrition Program group (2.23±1.18) was significantly greater than those for the WIC group (1.91±0.98; t test, P=0.040). Daily fruit intake did not differ (Farmers’ Market Nutrition Program, 1.69±0.97; WIC, 1.64±1.21; t test, P=0.769). No other variations in behaviors related to fruit and vegetable intake (fruit and vegetable variety, eating two or more servings of vegetables at a main meal, and eating fruits and vegetables as snacks) were significantly different (P>0.05).

Of the six constructs on Table 1, women from Farmers’ Market Nutrition Program showed higher scores in four (perceived benefit, perceived diet quality, and stages of change continuums for both fruit and vegetable intake). Perceived benefit of fruit and vegetable intake also significantly differed between groups (χ2=4.574, P=0.032), with 83.1% of women from Farmers’ Market Nutrition Program having a maximum perceived benefit score, while only 69.2% of the WIC group respondents had such a score. Both stages-of-change continuums with regard to fruit (χ2=12.171, P=0.007) and vegetable (χ2=10.238, P=0.017) intake significantly differed. For fruit, 78.5% of Farmers’ Market Nutrition Program participants were actively trying to increase intake compared to 58.9% for WIC. For vegetable stages of change, 83.1% of Farmers’ Market Nutrition Program participants fell within the action category, compared to 74.1% for WIC. There were no significant differences in the other two constructs.

For the entire sample, food insecurity was negatively associated with perceived diet quality (r=−0.248, P<0.001). Women from these households had a greater perceived diet quality (χ2=7.219, P=0.027), with 30.8% of Farmers’ Market Nutrition Program respondents indicating “Very Good or Excellent” diet, compared to 15.9% for nonparticipants.

Discussion 

Studies show that low-income, rural, women with children are at increased risk for experiencing food insecurity (2, 4, 27). WIC households might also be at increased risk for experiencing food insecurity (2, 27, 28). Yet, no studies have been conducted regarding the household food security status of Farmers’ Market Nutrition Program participants.

Studies on Farmers’ Market Nutrition Program’s effectiveness to increase fruit and vegetable consumption behaviors have been somewhat inconclusive. Several studies suggest positive effects on attitudes, beliefs, consumption, and shopping habits with regard to fruits and vegetables (21, 29, 30, 31). However, at least one study has reported no evident effect on fruit and vegetable consumption (32).

In our study, differences between groups were found in several psychosocial indicators and behaviors, with women living in Farmers’ Market Nutrition Program households demonstrating positive increases in all substantially different categories. However, household food security status or perceived health status did not differ based on Farmers’ Market Nutrition Program participation. Yet, among the entire sample, food insecurity was negatively associated with perceived health and perceived diet quality.

Despite all being eligible to participate in Farmers’ Market Nutrition Program, women with higher levels of education were more likely to be from households participating in Farmers’ Market Nutrition Program. Low education attainment has long been associated with poor diet quality (33). Results reinforce the need for continued nutrition education to food assistance program recipients.

The 2005 US food security estimates highlighted that 11.0% of all households, and 39.5% of those receiving WIC benefits, experienced some level of food insecurity (2). In this study, 52.6% (n=122) of respondents were from households experiencing food insecurity within the previous 12 months. An additional 20.6% (n=47) were from households “at risk” for food insecurity. Several reasons might account for these levels of food insecurity. First, nationally, food insecurity is more prevalent in households with children under the age of 6 years (16.7%) than for those without children (8.5%). Food insecurity further increases among low-income households (<1.30 of the income-to-poverty ratio) with children (40.6%) (2). In addition, households with WIC-eligible incomes (<1.85 of the income-to-poverty ratio) experience food insecurity more than those with higher incomes (28.5% vs 5.2%) (2).

Household food security status did not differ between groups. The small scale of food benefits received by Farmers’ Market Nutrition Program participants ($18 per WIC recipient, per season) does not likely represent an adequate amount of relief to alleviate or substantially improve household food insecurity. Furthermore, this study’s design does not permit cause-and-effect conclusions.

Mean daily servings of vegetables for the Farmers’ Market Nutrition Program participants were greater than for nonparticipants. Mean fruit intake did not differ. This trend is inconsistent with the higher stages of change scores for fruit intake for the Farmers’ Market Nutrition Program group. A higher stages of change score suggests that either steps are being taken to increase consumption of fruits and vegetables, or that an optimal intake has already been achieved.

In this study, women who participated in the Farmers’ Market Nutrition Program had a greater perceived diet quality, greater perceived benefit of fruit and vegetable intake, and were at an advanced stages of change with regard to fruit and vegetable intake. Yet, there were no substantial differences in the other two psychosocial construct measurements (self-efficacy and perceived control). Women who were more informed about the benefits of fruit and vegetable intake may have been more likely to participate in the Farmers’ Market Nutrition Program. This study was conducted at the end of the Farmers’ Market Nutrition Program season, therefore, it is unclear whether the increased perceived diet quality was a result of participating in Farmers’ Market Nutrition Program, or a reason for participating. It is possible that women who initially perceived they had higher quality diets were more likely to participate.

Farmers’ Market Nutrition Program participants were at an advanced stages of change regarding both fruit and vegetable intake, reporting more frequently being in the “action” phase of the stages-of-change continuum. Again, from these data, it is difficult to conclude whether this was a result of participation in the Farmers’ Market Nutrition Program, or a reason for participation. The high cost of fresh fruits and vegetables, items not typically offered in the current WIC food packages, might have motivated participants who were already trying to increase their fruit and vegetable consumption to participate. Conversely, the Farmers’ Market Nutrition Program might have increased participant awareness of the availability of fresh, local, and seasonal fruits available at farmers’ markets and, therefore, affected intake habits.

Self-efficacy and perceived control with regard to fruit and vegetable intake did not differ between groups. This suggests that the women did not differ in their confidence, ability, or knowledge of how to prepare fruits and vegetables, or their level of control over meal preparation (22). Therefore, differences observed in the other constructs (perceived benefits, perceived diet quality, fruit and vegetable stages of change continuums, and number of vegetable servings per day) might not have been a result of these factors, which are usually addressed through nutrition education efforts. All of the respondents were living in households receiving WIC benefits, which includes some type of nutrition education for participants. These results suggest that WIC nutrition education efforts should not only focus on fruit and vegetable preparation, but should also focus on the benefits of increased fruit and vegetable intake.

All WIC food packages include fruit juice. While it was not measured, in view of the study by Herman and others (21), it is possible that Farmers’ Market Nutrition Program participants’ fresh fruit intake was greater than that of nonparticipants. Alternately, lack of difference in fruit intake might have been a result of the small scale of food benefits received from the Farmers’ Market Nutrition Program. Overall, it is difficult to assess whether differences are a result of program participation or “self-selection” of women with greater previous vegetable consumption habits.

Previous studies show that food insecurity is negatively associated with diet quality, variety, and nutrient intake (7, 8, 9, 10, 11, 34). In our study, food insecurity was negatively associated with poorer perceived diet quality. Overall, the participants of this study might not represent the targeted sample, women living in WIC households from Athens County, OH, because of the nonprobability sampling strategy and the possibility that women with a greater interest in nutrition may have been more likely to complete the survey. In addition, although this study was conducted in a rural Appalachian county, Ohio University is located in Athens, possibly skewing education levels and not representing the entire region or state accurately. This may also have resulted in an overestimation of fruit and vegetable intake in both groups. Yet, these results support that food insecurity is negatively associated with perceived diet quality. In addition, our findings support the need to explore the impact of the Farmers’ Market Nutrition Program on food insecurity and fruit and vegetable intake behaviors. Limitations of the study include that some members of the sample may have had lower rates of literacy than average for the United States, which may have posed a limitation because of the self-administered nature of the survey. Finally, this survey was only conducted at the end of the Farmers’ Market Nutrition Program season rather than using a pre- and posttest design; therefore, results may not be attributed to participation in the program.

Conclusions 

Food insecurity is negatively associated with reported poorer perceived diet quality among women living in households receiving WIC benefits. Participation in Farmers’ Market Nutrition Program is positively associated with indicators of increased fruit and vegetable intake, as well as higher levels of formal education. Continued research on this topic is necessary to fully understand the factors that contribute to food security status, perceived diet quality, behaviors related to fruit and vegetable intake, and outcomes of women living in WIC households participating in Farmers’ Market Nutrition Program. Researchers should attempt to increase the sample size and limit response bias by administering the survey on location at the WIC office. A pretest/posttest methodology would strengthen a follow-up study and would facilitate measuring behaviors both before and after participation in the Farmers’ Market Nutrition Program. In-person interviews would also limit response bias due to limited literacy level. In addition, it would be helpful to note the type of nutrition education each participant received to determine which methods are the most influential on the outcomes measured.

Despite attempts to create a food and nutrition safety net, food insecurity and poor diet quality exist at unsettling levels throughout the United States, particularly among specific populations, including rural, low-income women with children. Providing nutrition education to all food assistance program participants, including the benefits associated with the recommended intake of fruits and vegetables, as well as the availability and affordability of fresh produce, must be a priority. Equal access to quality nutrition education and a safe, nutritionally adequate food supply by all Americans can be achieved through the continuation and improvement of programs, such as WIC and Farmers’ Market Nutrition Program.