Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation





First Advisor

Dr. Ellen Hahn


This study was designed to explore the literature on competitive food policy implementation (CFPI); examines demographic and school factors associated with CFPI; and explores the experiences of school leaders and staff in CFPI using a proposed theoretical framework to guide the research. Competitive foods are those sold in vending machines, a la carte settings, fundraisers, class parties and other venues which compete with foods offered through the national school lunch and breakfast programs. Competitive foods have traditionally been of low nutritional value and high energy density. CFPI may be effective in reducing student calorie intake and BMI. However, evaluation of competitive food policy effectiveness is difficult due to variability in policy implementation. A theoretical framework is needed to guide research on CFPI.

This research was a mixed methods study including a review of the literature, quantitative secondary analysis, and a qualitative content analysis of transcripts from semi-structured interviews with school personnel to understand their experience with CFPI. First, a systematic review of the research literature on CFPI was conducted. Demographic and school factors, policy features, and school and parent/community-level factors that impact CFPI were identified. Second, the association of multiple demographic and school factors with CFPI scores was examined. CFPI scale (overall) and sub-scales (“inside” and “outside” school) were developed and validated to evaluate CFPI effectiveness in Kentucky middle and high schools (N=640, grades 5-12). The scales were based on responses to 8 questions on competitive food practices from a 2011 School Tobacco and Wellness Policy biannual survey conducted by the University Of Kentucky College Of Nursing Tobacco Policy Research Program. Student BMI tracking and presence of a written wellness policy predicted higher scores on the overall CFPI scale (BMI OR=2.06, p=0.001; Wellness OR=1.74, p=0.02), inside subscale scores (BMI OR=2.46, p<0.0001; Wellness OR=1.58, p=0.05), and outside subscale scores (BMI OR=2.27, p=0.03; Wellness OR=1.54, p=0.0005). Greater county-level adult obesity rates predicted lower overall CFPI scores (OR=0.93, p=0.02). Private school status predicted lower scores on inside CFPI subscale scores (OR=0.47, p=0.004). Third, semi-structured interviews were conducted with 23 school personnel to explore CFPI. Interviews were recorded, transcribed and content analysis was conducted. Kentucky schools were stratified into four groups based on school level (middle or high) and CFPI scores (high or low). Sixteen schools were randomly selected for each of the four groups. A total of eight schools, two from each group, agreed to participate. The interview guide was based on a proposed CFPI framework based on implementation science, educational and organizational theory research. Six key themes emerged: internal/external forces enabling CFPI; internal and external obstacles to CFPI; key organizational values; organization value of CFPI; methods that organizations use to communicate organizational values; and CFPI policies and procedures. Findings were discussed in the context of the proposed theoretical framework. Implications for policy, practice and future research are presented.

Digital Object Identifier (DOI)