P95 Reliability of the Food Literacy Assessment Tool (FLAT) in Low-Income Adults

Background Food literacy is defined as the knowledge, skills, and behaviors required to plan and manage, select, prepare, and eat in support of health within cultural, socioeconomic, and environmental context. Low-income populations show disproportionately higher prevalence of diet-related chronic diseases compared to their counterparts. Nutrition plays a key role in prevention of these chronic diseases. There is no valid and reliable tool to assess food literacy in low-income populations. The Food Literacy Assessment Tool was developed to assess knowledge, self-efficacy, and behavior related to food literacy, and was previously tested for face validity and content validity. Objective To evaluate the reliability of the Food Literacy Assessment Tool (FLAT). Study Design, Setting, Participants A convenience sample of low-income adults were recruited at a non-profit organization providing emergency aid through a food pantry, charitable pharmacy, and social services to clients in need. Ninety-eight adults completed the FLAT, and 75 of the 98 adults returned and completed the same survey a second time 1-3 weeks after their initial survey. Measurable Outcome/Analysis Internal consistency reliability was evaluated using Kuder-Richardson Formula 20 (KR-20) for knowledge (dichotomous variable) and Cronbach's α for self-efficacy and behavior (measured using Likert scales). Test-retest reliability was assessed using intraclass correlation coefficient (ICC). Results The FLAT was found to have acceptable internal consistency in self-efficacy (Cronbach's α = .921) and behavior (Cronbach's α = .899), but not in knowledge (KR-20 = .510). The FLAT also showed acceptable test-retest reliability for knowledge (ICC = .839), self-efficacy (ICC = .703), and behavior (ICC = .930). Conclusions Overall, the FLAT is considered to be appropriate to assess food literacy in low-income adults in the United States.