Nutrients

Ultra-processed foods (UPFs) have become a growing part of everyday diets worldwide and have been linked to obesity, non-communicable diseases, and adverse mental health outcomes. Yet their role in deepening nutritional and health inequalities across socioeconomic groups and different stages of life is still not fully addressed in policy discussions. Drawing on recent population-based studies, longitudinal surveys, and mechanistic evidence, this perspective examines how UPF consumption intersects with food insecurity, psychosocial distress, and neurodevelopmental health from pregnancy through adulthood. Evidence from the United States shows that adults experiencing food insecurity obtain a larger share of their daily energy from UPFs. At the same time, surveys from low- and middle-income countries (LMICs) show that many infants and young children are already consuming sugar-sweetened beverages. Other observational studies have linked higher UPF intake with addictive-like eating behaviours and symptoms of depression, anxiety, and stress, although these studies cannot yet prove causality. In many LMICs, ongoing dietary transitions are also contributing to a rapid increase in overweight and obesity among socioeconomically disadvantaged groups, even while childhood stunting declines only slowly. Together, these findings suggest that UPF exposure is not only a dietary issue, but also a marker of wider social and nutritional vulnerability across the life-course. Breaking this cycle will require coordinated policy action, including stronger restrictions on unhealthy food marketing, front-of-package warning labels, better regulation of foods marketed to infants and young children, reforms to food-assistance programmes, and fiscal measures that make minimally processed foods more affordable and accessible.

Background/Objectives: Vitamin D regulates both the adaptive and the innate immune system, while having anti-inflammatory, immunomodulatory and anti-fibrotic properties. In clinical studies, 25(OH)D₃ deficiency was associated with systemic inflammation, but it is unclear whether these associations are causal. In this study, we aimed to investigate the impact of Vitamin D receptor (VDR) polymorphisms (SNPs)—as surrogates for functional relationships—on 25(OH)D₃ status and systemic inflammation in decompensated liver cirrhosis. Methods: Patients with compensated and decompensated cirrhosis with or without ACLF were enrolled prospectively from a monocentric cohort study (N = 355). Pyrosequencing was performed for genotyping SNPs (rs7968585, rs731236 (TaqI), rs7975232 (ApaI), rs2239179 and rs2228570 (FokI)). VDR-SNPs were analysed for their association with 25(OH)D₃ concentrations and markers of systemic inflammation. Results: Severe 25(OH)D₃ deficiency was frequent in decompensated cirrhosis. 25(OH)D₃ supplementation was associated with significantly higher circulating 25(OH)D₃ in compensated and decompensated liver cirrhosis, whereas in ACLF, 25(OH)D₃ levels did not differ significantly between supplemented and unsupplemented patients. An inverse correlation of 25(OH)D₃ and IL-6 levels was identified (R = −0.2286, 95% CI −0.328 to −0.124, p < 0.0001). In exploratory analyses among the analyzed VDR-SNPs, the GG-variant of rs2239179 was associated with lower 25(OH)D₃ levels (β = 0.705, 95% CI 0.498-0.998, p = 0.049). Moreover, the functionally important VDR-SNP rs2228570 TC-variant showed a nominal independent association with higher IL-6 levels (ß = 1.685, 95% CI 1.102–2.576, p = 0.016) after adjustment. Conclusions: These exploratory findings suggest that VDR genetics may partially contribute to 25(OH)D₃ status and systemic inflammation in liver cirrhosis and ACLF and warrant further replication in independent cohorts.

Background/Objectives: Large language model (LLM) pipelines increasingly convert free-text meal descriptions into nutrient values, yet reference-standard validation is rarely feasible in production. We asked whether the controls such pipelines conventionally pass suffice and built a reference-standard-free battery any pipeline can run on its output. Methods: We audited 13,073 meals logged by 513 adults through a WhatsApp conversational agent in rural Colombia (9 February 2026 to 6 August 2026). The pipeline is a two-stage cascade of commercial LLMs from different providers: a vision model renders photographs as text, while a text model maps these to nutrients. Six pre-specified tests were applied, including test–retest ICC(1,1) over recurring strings, stratified by modality because 48.3% of records carried a photograph. Results: The pipeline passed every conventional control (Atwater-reconstructed energy r = 0.9951; median relative error 0.0000) but failed reproducibility. Across 191 recurring strings (944 meals), energy ICC point estimates were 0.7061 (95% CI 0.5419–0.8217) text only and 0.5258 (0.3778–0.6513) with photographs; none reached 0.75. Modality is a design control, not an established effect, and the coefficient depends on which inputs recur, so it is not intrinsic. Under the strictest identity available—byte-identical text, no photograph—ICC was 0.6944. Intermediate-text inspection exposed a failure invisible at the output: 1.44% of photograph-accompanied records carried a vision-stage artefact string versus 0.01% of text-only records (ICC 0.0911). Conclusions: Internal consistency does not validate an LLM-assisted nutrition pipeline. This design measures instability under recurring inputs without isolating its cause. Prospective test–retest reproducibility is a cheap, necessary acceptance criterion, and in a cascade, the intermediate representation must also be inspected.