A Behavioral Science-Informed Agentic Workflow for Personalized Nutrition Coaching: Development and Validation Study
Bibliographic record
Abstract
BACKGROUND: Effective management of cardiometabolic conditions requires sustained positive nutrition habits, often hindered by complex and individualized barriers. Direct human management is simply not scalable, while deterministic automated approaches to nutrition coaching may lack the personalization needed to address these diverse challenges. OBJECTIVE: We report the development and validation of a novel large language model (LLM)-powered agentic workflow designed to provide personalized nutrition coaching by directly identifying and mitigating patient-specific barriers. METHODS: We used behavioral science principles to create a comprehensive workflow that can map nutrition-related barriers to corresponding evidence-based strategies. First, a specialized LLM agent to intentionally probe for and identify root causes of a patient's dietary struggles. Subsequently, a separate LLM agent to deliver tailored tactics designed to overcome those specific barriers. We conducted a user study with individuals with cardiometabolic conditions (N=16) to inform our workflow design and then validated our approach through an additional user study (n=6). We also conducted a large-scale simulation study, grounding on real patient vignettes and expert-validated metrics, where human experts evaluated the system's performance across multiple scenarios and domains. RESULTS: In our user study, the system accurately identified barriers and provided personalized guidance. Five out of 6 participants agreed that the LLM agent helped them recognize obstacles preventing them from being healthier, and all participants strongly agreed that the advice felt personalized to their situation. In our simulation study, experts agreed that the LLM agent accurately identified primary barriers in more than 90% of cases. Additionally, experts determined that the workflow delivered personalized and actionable tactics empathetically, with average ratings of 4.17-4.79 on a 5-point Likert scale. CONCLUSIONS: Our findings demonstrate the potential of this LLM-powered agentic workflow to improve nutrition coaching by providing personalized, scalable, and behaviorally-informed interventions. CLINICALTRIAL: Not applicable.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.012 | 0.034 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".