Who is at high risk of poor diet quality? A precision public health nutrition approach using machine learning.

Dietary patterns, diet quality scores and the risk of diabetes: a multi-country study

Dietary patterns are key determinants of non-communicable disease (NCD) risk. Different dietary assessment approaches may capture diet quality differently. Methods include data-driven derived patterns and predefined diet quality scores such as the NCD-Protect and NCD-Risk indices. Evidence comparing agreement between these approaches and their associations with glycemic outcomes in low- and middle-income countries (LMICs) remains limited. This cross-sectional analysis used data from the Severe Acute Malnutrition: Role of the Pancreas (SAMPA) study which included 2,251 participants (1,821 adults and 430 children) from Tanzania, Zambia, the Philippines, and India. Dietary patterns were derived using principal component analysis (PCA) of food frequency questionnaire (FFQ) data, NCD-Protect and NCD-Risk scores were generated by pragmatically recoding FFQ data to approximate 24-hour dietary recall-based indicators. Agreement between PCA-derived healthy/unhealthy patterns and NCD-Protect/NCD-Risk scores was assessed using Kappa-statistics. Associations between diet and glycemic markers, plasma glucose at 120 min during an oral glucose tolerance test (glucose120) and HbA1c were examined using multivariable linear regression adjusted for age, sex, socioeconomic status, and HIV status. Fair to very good agreement was observed between PCA-derived dietary patterns and NCD-Protect/NCD-Risk scores across cohorts (kappa 0.21–0.57), indicating overlap in constructs of diet quality. Associations with glycemic markers were inconsistent and cohort-specific. There were also paradoxical associations, including higher glucose120 among Zambian adults classified as having healthier diets. Additional analyses showed that prior wasting malnutrition did not modify associations between participant characteristics and diabetes, and that demographic, socioeconomic, and clinical factors alone did not explain the observed heterogeneity. While PCA-derived dietary patterns and global diet quality scores show reasonable agreement, their associations with glycemic outcomes vary across populations and diagnostic measures. These highlight the need for context-specific dietary assessment tools and caution against the universal application of diet quality metrics when evaluating metabolic risk in diverse LMIC settings.

Artificial intelligence in nutritional health: sustainable consumers’ right to quality of life with special reference to youth

Background Maintaining optimal youth nutritional health is an urgent socio-economic imperative that underpins long-term human productivity and rights-based development. However, modern youth cohorts face unique dietary threats caused by the widespread availability of ultra-processed foods, targeted digital marketing, and complex food labeling protocols. Although Artificial Intelligence (AI) presents innovative avenues for personalized dietary profiling, existing systems remain largely technocentric and detached from statutory frameworks or behavioral realities. Objective This study bridges this interdisciplinary divide by evaluating a rights-based, technology-driven framework to improve youth nutritional health. It aims to: (1) empirically evaluate the “Knowledge-Attitude-Practice” (KAP) gap linking statutory consumer rights to real-world eating habits; (2) present the engineering design of a non-commercial Progressive Web Application (PWA) built to translate legal safeguards into daily behavioral changes; and (3) triangulate these findings using data from youth surveys and expert legal and nutritional panels. Methods Using a cross-sectional approach based on non-parametric power constraints, a validated survey instrument was completed by a target sample of Indian youth (n = 354, aged 15–25 years). Concurrently, data matrices were compiled from regional legal experts (n = 12) and public nutrition professionals (n = 12) to cross-verify structural bottlenecks. Group variances, demographic dependencies, and rank associations were analyzed using robust non-parametric tests, including One-Way ANOVA, Kruskal-Wallis (H), Welch’s t-test, and Kendall’s Tau (τ) correlation coefficients. Results Inferential analysis revealed unexpected demographic trends: undergraduate status predicted significantly higher FSSAI safety awareness than post-graduate status (p = 0.0037), while subjective health ratings exhibited a non-linear relationship with household income (p = 0.0004), peaking in the lower-middle financial tier. Crucially, rank correlation testing revealed that the relationship between statutory knowledge and actual dietary actions is functionally non-existent (τ = −0.001). This near-zero correlation provides clear empirical proof of a pronounced Knowledge-Action Gap, confirming that passive legal literacy fails to influence food selection in modern environments. Conclusion By framing automated behavioral interventions within the constitutional protections of Article 21 of the Constitution of India and the Consumer Protection Act, 2019, this study shows how the open-access PWA (nutrition-zb.pages.dev) can bridge this behavioral gap. This shifts the focus of consumer health informatics from basic self-tracking to a rights-based, systemic public health intervention.

Toddler diet quality: a factor associated with childhood growth?

Association between the planetary health diet and sleep disorders: results from the UK Biobank.

BACKGROUND Building on evidence that diet may ameliorate sleep disorders, and following the introduction of the EAT-Lancet Planetary Health Diet in 2019 to promote human health and environmental sustainability, this study aimed to investigate the potential association between the planetary health diet and the risk of sleep disorders. METHODS This study utilized data from the UK Biobank and included 165866 participants (mean baseline age 55.8 years; 53.5% female). The planetary health diet index (PHDI) was calculated using a 24-hour dietary recall, with adherence quantified using three scoring methods. Sleep disorders were identified using ICD-10 code G47 from linked health records. Cox proportional hazards regression was used to estimate the association between PHDI and sleep disorder risk. RESULTS During a median follow-up of 13.9 years, 3200 participants developed sleep disorders, of whom 1282 (40.1%) were female. At baseline, 78113 (47.1%) participants scored below the Knuppel PHDI mean, 83258 (50.2%) below the Stubbendorff PHDI mean, and 89528 (54.0%) below the Kesse-Guyot PHDI mean. Compared with the lowest adherence group, participants in the highest adherence group for the Knuppel index had a lower risk of sleep disorders (HR = 0.827, 95% CI: 0.733-0.933). Similar associations were observed for Stubbendorff (HR = 0.796, 95% CI: 0.712-0.891) and Kesse-Guyot (HR = 0.838, 95% CI: 0.754-0.931). CONCLUSIONS Higher adherence to the planetary health diet was associated with a reduced risk of sleep disorders, supporting the potential role of dietary recommendations in sleep health.

A machine learning-derived dietary pattern for aging

Diet is a modifiable determinant of aging. We develop and validate the Machine-learning YouTHful (MYTH) Diet, a dietary pattern associated with reduced aging-related mortality. Using data from 191,689 participants in the UK Biobank, we conducted a food-wide association analysis and identified 18 food groups significantly associated with aging-related mortality. A Light Gradient Boosting Machine (LightGBM) model was used to rank food importance, leading to the construction of a 10-component MYTH Diet score (range: 0–10). Higher MYTH scores were consistently associated with reduced aging-related mortality in both internal (Quartile 4 vs. 1: hazard ratio [HR] = 0.79; 95% CI: 0.75–0.84) and external (Q4 vs. Q1: HR = 0.68; 95% CI: 0.58–0.80) validation cohorts. Multi-omics analyses revealed that the diet’s protective effects were partly mediated through proteomic, metabolic, and inflammatory pathways, with mediators including TNFRSF4, the proportion of polyunsaturated fatty acids (PUFA%), and lipid-related metabolites such as medium very-low-density lipoprotein phospholipids (M-VLDL-PL). Higher MYTH scores were also linked to slower biological aging in the lungs, liver, pancreas, as well as lower risks for 15 aging-related diseases. These findings suggest that the MYTH Diet may offer a biologically informed, scalable framework for developing personalized nutrition strategies aimed at supporting healthy aging and longevity.

Construct validity of the Planetary Health Diet Index relative to global food group-based measures of a healthy diet: A multi-country analysis of 24-hour dietary intake data on the FAO/WHO Global Individual Food Consumption Data Tool.

BACKGROUND The Planetary Health Diet Index (PHDI) was developed as a measure of adherence to the Planetary Health Diet proposed by the EAT-Lancet Commissions, however, its construct validity for accurately reflecting a healthy diet as compared to other global dietary measures remains to be evaluated. OBJECTIVE The construct validity of PHDI was assessed by comparing the strength of its associations with reference metrics of nutrient adequacy and moderation of foods associated greater risks of diet-related non-communicable diseases against food group diversity score (FGDS) and Global Diet Quality Score (GDQS) and its Positive (GDQS+) and Negative (GDQS-) sub-metrics. METHODS Cross-sectional quantitative 24-hour recall and food record data from 152,004 non-pregnant females and males (≥15 years) from 45 surveys in 26 countries accessed from the FAO/WHO Global Individual Food Consumption Data Tool were analysed. Multilevel linear and modified Poisson regression models quantified associations between PHDI, FGDS, GDQS+, and GDQS- with reference measures [e.g., mean adequacy ratio (MAR), % energy from ultra-processed food (UPF)] or indicators of a healthy diet [e.g., ≥400 g/day of fruit and vegetables (F&V)], respectively. RESULTS Adherence to PHDI was highest in low-income (LIC) and lower-middle income countries (LMIC), and lowest in upper-middle (UMIC) and high income countries (HIC). One-SD increments in PHDI were associated with higher MAR (β: 5.49 percentage points (pp); 95% confidence interval (CI): 3,65, 7.33 in LIC to 1.41 (1.23, 1.58) in UMIC), appropriate F&V intakes (relative risk (RR): 2.41 (1.85, 2.49) in UMIC to 1.19 (1.05, 1.34) in LMIC), lower unprocessed red meat (β: -10.0 g/day (-11.4, -8.73) in LIC to -6.86 (-7.54, -6.17) in UMIC), and lower proportions of energy from UPF (β: -2.47 pp (-3.77, -1.18) in LMIC to -1.21 (-2.32, 0.105) in LIC), however, one-SD increases in FGDS and GDQS+ outperformed PHDI in predicting measures of nutrient adequacy, particularly MAR (β for FGDS: 11.2 pp (9.13, 13.3) in LIC to 8.57 (8.42, 8.72) in UMIC; β for GDQS+: 9.82 (7.26, 12.4) in LMIC to 7.18 (6.00, 8.36) in HIC) and appropriate F&V intakes (RR for FGDS: 2.78 (2.26, 3.41) in LMIC to 1.63 (1.22, 2.19) in HIC; RR for GDQS+: 2.39 (1.89, 2.89) in LMIC to 1.55 (1.16, 1.96) in LIC). CONCLUSIONS PHDI may serve as a robust, dual purpose measure for monitoring adherence to diets that are both sustainable and healthy in contexts with regular quantitative dietary intake assessments and low burdens of diet-related undernutrition, however, simpler food group-based measures such as FGDS often show stronger associations with healthy diet sub-constructs such as nutrient adequacy, and are more feasible for large-scale, routine monitoring of diets.