Project:
Understanding the Relationships Among Child Food Insecurity, Diet Quality, and Physical Activity in Family, School, and Geographic Context

Year: 2013

Research Center: Southern Rural Development Center, Mississippi State University

Investigator: Fram, Maryah, Edward Frongillo, Lorrene Ritchie, and Nila Rosen

Institution: University of South Carolina

Project Contact:
Maryah Fram
University of South Carolina
College of Social Work
Columbia, SC 29208
Phone: 803-777-5945
E-mail: fram@mailbox.sc.edu

Summary:

Although there is substantial evidence that food insecurity influences children’s experiences and their development, little is known about the pathways through which this occurs. Conceptually, food insecurity is posited to affect individual well-being through both nutritional and non-nutritional pathways, and in ways that are shaped by attributes of individuals as well as by the family, social, and resource contexts in which food choices are negotiated. There is little empirical evidence, however, regarding the relationships between child food insecurity and child health and the pathways or contexts that ultimately impact the health of a food-insecure child.

Clarifying how children’s experiences of food insecurity impact their health is critical for developing policies and programs to improve child health. This is particularly salient now, given the childhood obesity epidemic and the potentially greater risk for obesity among children who are food insecure. In particular, to inform policies and programs that can improve child food security without promoting unhealthy weight gain, we need to more fully understand how food insecurity shapes what children eat and how much they engage in physical activity, and how this may change for children in different geographic, social, and cultural contexts.

Toward that end, data from an evaluation of the Network for a Healthy California–Children’s PowerPlay! Campaign were analyzed. PowerPlay! is a social marketing initiative that targets ethnically diverse, low-income 4th and 5th grade children and their families in California, encouraging children to eat the recommended amount of fruits and vegetables and get at least 60 minutes of physical activity daily. The study sample consists of 4th and 5th grade children, with 50 percent self-identified as Latino, in 44 schools in San Diego County, randomly assigned to intervention (n=22) or control (n=22) conditions. All schools had at least 50 percent of students eligible for free or reduced-price meals. Following the intervention, children completed a 24-hour diary-assisted recall on all foods and beverages consumed over one day; a survey that included items from the Child Food Security Assessment (CFSA); and questions about physical activity, including self-reported minutes of time spent in physical activity over 1 day, as well as preference for, and barriers to, physical activity. CFSA items indicate whether a child has experienced food insecurity during the last school year, asking if the child has had an experience never, one or two times, or many times.

CFSA assesses three domains of food insecurity: (1) emotional awareness (2 items); (2) cognitive awareness (1 item); and (3) physical awareness (2 items). Total CFSA scores ranged from 0 to 10 with a higher score indicative of a higher level of food insecurity. Food diary information was used to determine both quantities consumed of key dietary components (fat, sugar, fiber, fruits, vegetables, and whole grains) and the quality of diet adjusted for energy intake (0-100) as reflected in the Healthy Eating Index-2005 (HEI) and its subscales. Higher HEI overall and subscale score indicate a higher quality diet. Additional survey questions indicate the child’s gender, age, and race/ethnicity. Linked administrative data show the proportion of students receiving free or reduced-priced lunch in the child’s school and whether the school serves an urban or non-urban area. The analytic sample consisted of 3,605 children for whom post-test dietary, food insecurity, physical activity, and contextual data were available.

Child food insecurity was assessed using descriptive statistics. Relationships between child food insecurity and diet and physical activity were assessed using regression and logistic regression (with Stata 12 xtmixed and logistic commands), clustering by school. The influence of contextual factors, including race/ethnicity, school free/reduced lunch population, intervention participation, and geographic location, on the relationships were evaluated by including interaction terms in the regression equations.

Over 60 percent of sample children experienced at least some food insecurity during the academic year (average CFSA score of 2.2). Child food insecurity was associated with differences in both quantity and quality of diet. Increased CSFA score was associated with increased consumption of total energy, fat, sugar, and fiber. For instance, on average a child scoring 10 on the CFSA consumed about 117 calories and 8 grams of fat more in a day than did a child who scored 0. Increased CSFA score was also associated with a lower HEI total vegetable subscale score. These findings suggest that children with higher levels of food insecurity appear to eat more overall and their diets are less healthy in terms of vegetable consumption. Increased CSFA score was associated with marginally fewer daily minutes of self-reported physical activity, averaging about 17 minutes per day less physical activity for a child scoring 10 on the CFSA versus a child scoring 0. Observed measurement of child physical activity would reduce error and could clarify the nature of this relationship (the study recommends that future research take such an approach).

The study also revealed that each unit increase in CFSA score was significantly associated with 8-percent lower odds of liking physical activity, 20-percent higher odds of always feeling too tired for physical activity, and 26-percent higher odds of always feeling that weight makes physical activity hard. Context had some influence on the relationships between child food insecurity, diet, and physical activity. Although neither race/ethnicity nor geographic location (for example, urban or non-urban) moderated these relationships, children in higher poverty schools (as assessed by percent eligible for free or reduced-price meals) had increased fruit consumption as food insecurity increased. The PowerPlay! intervention was also a significant contextual factor. Being in an intervention rather than a control school reduced the negative effect of food insecurity on vegetable consumption and was associated with healthier overall diet (as assessed by total HEI-2005 score), although sugar consumption increased as food insecurity increased.

Taken together, these findings indicate that child food insecurity is associated with generally worse diet quality and challenges to getting adequate physical activity, regardless of children’s race/ethnicity or where they live (for example, urban or non-urban). These findings help to explain why children who experience food insecurity are at risk for negative developmental outcomes. Interventions may help to ameliorate some of this risk, providing both information and opportunities for healthier eating. Additional research should be conducted, using observed measures of child physical activity, and further examining the relationships among child food insecurity, diet, and physical activity among children of different ages and in different geographic and socioeconomic contexts.