A Model for Personalised Nutrition Literacy and Behavior Change Framework

Thesis Abstract

This study addresses the persistent gap between nutrition knowledge and sustained healthy eating behaviours by developing and validating a model of personalised nutrition literacy and behaviour change that integrates individual cognitive, affective, and environmental determinants. The aim is to construct a theoretically sound framework that explains how tailored nutrition literacy interventions translate into measurable diet quality improvements and sustained behavioural change. Specific objectives are (1) to identify core components of nutrition literacy relevant to personalised interventions; (2) to delineate the pathways linking literacy to dietary behaviour via self-efficacy, motivation, and perceived barriers; (3) to empirically test the proposed framework using multi-source data; (4) to refine the model through qualitative insights from end-users; and (5) to propose a practical assessment and intervention toolkit for dietitians and public health practitioners. A mixed-methods sequential explanatory design will be employed. In the quantitative phase, a cohort of 600 adults aged 18–65 will be recruited from primary care and community settings, stratified by age, sex, and socioeconomic status. Baseline data will be collected using a validated Nutrition Literacy Scale, a General Self-Efficacy scale, a Motivation for Healthy Eating index, a Perceived Barriers to Healthy Eating questionnaire, and a 24-hour dietary recall to derive the Healthy Eating Index-2015 (HEI-2015). An adaptive personalised intervention, grounded in Social Cognitive Theory and the Health Belief Model, will be delivered over 12 weeks via a digital platform that curates tailored educational content, goal setting, and feedback. Follow-up assessments will occur at 3, 6, and 12 months, with the primary outcome being change in HEI-2015 score and secondary outcomes including changes in nutrition literacy scores, self-efficacy, dietary self-monitoring frequency, and perceived barriers. Data will be analyzed using structural equation modelling to test mediation pathways from nutrition literacy to diet quality through self-efficacy and motivation, and latent growth modelling to examine trajectory changes over time. Regression analyses will identify predictors of successful behavioural adoption, and sensitivity analyses will address potential confounders. To enrich the quantitative findings, 40 in-depth interviews will be conducted with a purposive subsample representing diverse literacy levels and demographic backgrounds. Thematic analysis will extract experiential and contextual factors that facilitate or hinder the translation of nutrition literacy into action, informing iterative refinement of the theoretical model. Triangulation will integrate quantitative and qualitative results to elucidate the mechanisms by which personalised nutrition literacy exerts influence on dietary behaviours and to identify user-centric features essential for scalable interventions. Anticipated findings include (a) validation of a multi-construct model in which nutrition literacy directly and indirectly predicts diet quality through self-efficacy, goal commitment, and perceived barriers, with effect sizes indicating meaningful practical impact; (b) evidence that tailoring content to individual literacy profiles and motivational drivers enhances recruitment, engagement, and adherence to dietary recommendations; (c) identification of differential effects by age, sex, and socioeconomic status, highlighting the need for equity-focused design; and (d) qualitative insights revealing contextual facilitators such as social support, habit formation strategies, and perceived usefulness of personalised feedback. The study contributes to knowledge by operationalising a novel framework that explicitly links nutrition literacy to behavioural outcomes through validated psychological constructs, thereby informing theory advancement in health literacy, nutrition education, and behaviour change. Methodologically, it demonstrates the utility of integrating structural equation modelling with qualitative thematic insights to develop and validate a comprehensive, adaptable model for personalised nutrition interventions. Practically, the resulting toolkit—comprising a concise Nutrition Literacy Profile, an intervention tailoring engine, and a set of evidence-based behavioural strategies—will aid dietitians, clinicians, and public health practitioners in delivering scalable, personalised nutrition guidance. Limitations include potential self-report bias in dietary measures and generalisability constrained to similar healthcare contexts; these will be mitigated by objective dietary biomarkers in a subsample and by recruiting a diverse study population. Recommendations for future work include external validation in different cultural settings, exploration of integration with electronic health records, and assessment of long-term health outcomes beyond dietary quality.