Food Security, Hunger-coping, and Hunger-Symptoms, and Their Relationship With Daily Fruit and Vegetable Intake Frequency in a Low-income Sample

Abstract

Purpose To examine daily fruit and vegetable (FV) intake frequency by household food security status (high food security, marginal food security, low food security, and very low food security), and scores on three hunger-coping behavior scales (i.e., “rationing food supplies”, “financial strategies”, and “trade-off strategies”), and one physiological hunger symptoms scale among a very low-income population in the Midwestern United States.

Methods Adult participants (aged 19 and older and caregivers to at least one child aged 0-18) were recruited from public libraries, food pantries, and other community locations to participate in a cross-sectional self-administered survey (n = 306). The primary outcome variable was daily FV intake frequency, measured from five items from the 16-item NYPANS dietary screener (fruit, green salad, carrots, other vegetables, and non-fried potatoes). These items were converted into daily frequencies and then summed. Daily FV intake frequency was also log transformed to meet statistical model assumptions. The main independent variables in this study were household food security (measured using the 6-item USDA Household Food Security Survey Module), scores on three 5-item hunger-coping behavior scales (max scores = 5), and one 5-item physiological hunger symptoms scale (max scores = 5). Potential sociodemographic and household characteristics (e.g., sex, age, race/ethnicity, education, income, marital status, food assistance, income assistance, food pantry use, household child to adult ratio, and automobile access) were assessed as potential covariates. Generalized linear models were used to assess the relationship between independent variables and daily FV intake frequency.

Results 278 participants provided complete data for the outcome variable and were included in the sample. The sample was 73% female, 45% African American, 14% Hispanic, and 28% Caucasian. A majority of participants (60%) reported annual household incomes under $15,000. A high proportion of participants (42%) had very lowfood security, as compared to the national average of 6%. Participants reported consuming FV 2.40 times per day (SD=1.76) on average. Compared the high food secure group (In adjusted models), being in the “very low” food security group was associated with a 26% decrease, on average, in daily FV intake frequency (exp(β)=0.74, 95%CI=0.61-0.88, p

Conclusions Compared to the high food secure group, those in the “very low” food secure group reported a significantly lower mean daily FV intake frequency, but not those in the “low” food secure group. This emphasizes the need to examine the “low” and “very low” food secure groups separately when studying factors associated with diet. Engaging in hunger-coping behaviors and experiencing physiological hunger symptoms was associated with decreased daily FV intake frequency. It is not clear if the hunger coping-behaviors themselves directly lead to decreased daily FV intake frequency, or if they were a marker for food insecurity and associated economic distress. These constructs and relationships should be examined in future studies. Programs and policies, which seek to promote fruit and vegetable intake among low-income populations, should target the “very low” food secure as the most at-risk for low intake. Also, researchers and community organizations working with food insecure groups could potentially measure hunger-coping behaviors and hunger symptoms in a complementary fashion to traditional measures of food insecurity to gain more contextual information about the experiences of this population.