Abstract TU253: Predictors of Language Strength in State Legislation Related to School Nutrition

Abstract

Introduction: Dietary habits are a major contribution to the risk of developing chronic disease. School-based nutrition programs play a vital role in supporting children’s nutrition, and state-level policy can shape how these programs are implemented. The specific wording used in legislation can influence how policies are interpreted/enforced. Literature suggests that the use of strong language (e.g. must, shall, required) in child nutrition bills is associated with improved dietary quality. No studies to date have examined predictors of strong language in state-level child nutrition bills; this study addresses that gap by identifying factors associated with its use. Hypothesis: We hypothesized that there would be characteristics and content areas associated with the likelihood of including strong language. Methods: A legislative database (Legiscan) was used to construct a state-level, child nutrition legislation database from 2010 and 2020. Bills were coded for general characteristics (status, year, type, branch introduced, state), content areas (e.g. meal participation, stigma, unpaid meals), and additional factors (funding, enforcement, strength of bill language). Multivariable models were used to identify predictors of strong bill language. Results: Of the 582 bills introduced, 80.1% (n=466) used strong language. In a multivariable model, open/closed campus policies, school garden, unpaid meals, funding, and enforcement were all significantly related to the use of strong language. Bills addressing open/closed campus policies had 9.5 times higher odds of using strong language (CI:1.12, 80.21) than those that did not. Bills addressing school gardens were associated with 4.4 times higher odds of strong language (CI:1.78, 10.69) compared with bills that did not. Bills addressing unpaid meals had significantly lower odds of using strong language (OR = 0.16, CI:0.05,0.54) compared with bills that did not address unpaid meals. Bills including funding provisions had 60% lower odds of using strong language than those without funding (CI:0.24, 0.96). Bills including enforcement provisions had significantly lower odds of using strong language (OR = 0.27, CI:0.12, 0.64) compared with bills without enforcement provisions. Conclusion: Both supportive and limiting predictors of strong language were identified. Understanding how bill topics are associated with bill language strength can inform the drafting of child nutrition policy and guide future advocacy efforts.