Smartphone App for Personalised Dietetic Feedback and Metabolic Health Monitoring

Thesis Abstract

The increasing prevalence of diet-related non-communicable diseases and the complexity of individual dietary responses underscore the need for accessible, evidence-based nutrition guidance delivered through digital platforms. This study addresses the gap in scalable, personalized dietary feedback that integrates real-time metabolic health monitoring and behavior-change support within a single mobile application. The aim is to evaluate the effectiveness of a smartphone app that provides personalised dietetic feedback linked to metabolic health indicators, with objectives to (1) assess user engagement and adherence to personalised recommendations over 12 weeks, (2) determine the impact on dietary quality and metabolic markers (fasting glucose, HbA1c, lipid profile, BMI, and waist circumference), (3) examine the mediating role of self-efficacy and health literacy in behaviour change, and (4) identify user-perceived barriers and facilitators to sustained use of the app. A mixed-methods design was employed, integrating a quasi-experimental, controlled before-and-after component with a convergent qualitative strand. The population consisted of adults aged 25–60 years at risk of metabolic syndrome recruited from urban primary care clinics (n = 240). Participants were allocated to an intervention group (n = 120) receiving the smartphone app with personalised dietetic feedback and continuous metabolic monitoring, and a control group (n = 120) receiving standard care. Data collection employed validated instruments and device-generated metrics dietary intake via multiple 24-hour recalls and the Diet Quality Index; metabolic health markers including fasting plasma glucose, HbA1c, total cholesterol, HDL-C, triglycerides, systolic/diastolic blood pressure, BMI, and waist circumference; app usage analytics; and psychosocial measures using the Self-Efficacy for Diet Change Scale and the Newest Vital Sign for health literacy. Data were collected at baseline, 6 weeks, and 12 weeks. The primary analysis compared changes in dietary quality and metabolic markers between groups using repeated-measures ANOVA and ANCOVA, adjusting for baseline values and potential covariates. Mediation analyses tested whether self-efficacy and health literacy accounted for observed behavioural changes. A per-protocol sub-analysis examined the dose–response relationship between app engagement (session frequency, feature utilization) and outcomes. Qualitative data were gathered through semi-structured interviews with a purposive sample of 30 participants from the intervention group to explore user experiences, perceived usefulness, and barriers to sustained use. The qualitative data were analyzed thematically following the framework method, with triangulation to reinforce interpretation. Expected findings include greater improvements in dietary quality and metabolic markers in the intervention group relative to controls, with statistically significant reductions in fasting glucose, HbA1c, LDL cholesterol, and abdominal adiposity. It is anticipated that higher engagement with personalized feedback and real-time monitoring will be positively associated with adherence to dietary recommendations, and that self-efficacy and health literacy will mediate these behavioural changes. The study also expects qualitative insights to reveal that timely, actionable feedback, perceived credibility of dietetic guidance, and ease of app use support sustained engagement, while data privacy concerns and technological literacy may serve as barriers. The contribution to knowledge lies in demonstrating the feasibility and effectiveness of an ICT-driven, personalised nutrition intervention that couples dietetic feedback with objective metabolic monitoring, thereby bridging gaps between behavioural nutrition and clinical outcomes. The study advances understanding of how digital health tools can enhance self-management of diet-related risks and elucidates the mechanisms through which psychological determinants influence adoption and maintenance of healthy dietary behaviours. Limitations include potential selection bias from clinic-based recruitment, short duration for long-term outcome assessment, and reliance on self-reported dietary data. Practical implications emphasize integrating such apps into primary care workflows, informing guidelines for digital personalised nutrition interventions, and guiding future research on long-term maintenance and scalability. Recommendations include extending follow-up to 12–24 months, refining algorithms for more precise personalization, incorporating social support features, and evaluating cost-effectiveness across diverse populations.