Large Language Model Dietary Recommendations for Multimorbidity with Conflicting Guidelines in Turkish and English
4 affiliations
- 1Department of Nutrition and Dietetics, Faculty of Health Sciences, Istanbul Aydın University, Istanbul 34295, Türkiye
- 2Department of Computer Engineering, Faculty of Engineering, Kocaeli University, Kocaeli 41001, Türkiye
- 3Department of Artificial Intelligence Engineering, Bahçeşehir University, Istanbul 34353, Türkiye
- 4KOU Teknopark, Bidist, Kocaeli 41275, Türkiye
Abstract
Background/Objectives: Patients with multimorbidity often receive conflicting dietary guideline recommendations, and large language models (LLMs) increasingly draft dietary advice. In this study, we evaluated nine LLMs, including two Turkish-adapted models, on 76 synthetic patient vignettes in Turkish and English. Methods: The vignettes combined type 2 diabetes, chronic kidney disease, hypertension, obesity, gout and celiac disease as single conditions, conflicting pairs, concordant pairs and triples. An LLM judge panel graded the responses against key-feature checklists approved by a dietitian, whose blinded ratings of a subset agreed with the panel on 90% of items. Results: The proprietary models covered 81% to 92% of the weighted checklist and fully resolved 57% to 89% of the guideline conflicts. Their coverage did not differ significantly between Turkish and English in any of these models, whereas across all models Turkish responses covered 2 percentage points less, mainly in conflicting-pair vignettes and in the open-weight models. The GPT and Claude models had the fewest safety-critical flags, 5% to 8% in both languages, and the proprietary models’ Turkish advice was rated 4.9 to 5.0 out of 5 for cultural appropriateness. Conclusions: Expert review of LLM drafts can therefore focus on conflict resolution, and the checklist procedure provides a reusable audit.
Related
- Global perspectives on dietary management of small intestinal bacterial overgrowth (SIBO): a systematic review of clinical guidelines, consensus statements and expert recommendations
- Syndrome-oriented CNS infectious complications after hematopoietic stem-cell transplantation and CAR T-cell therapy, part 1—viral encephalitis, meningitis, and myelitis: recommendations from the Practice Harmonization & Guidelines Committee of the EBMT
- Delphi consensus guidelines for fungal sepsis in non-neutropenic critically ill patients in resource-limited settings: fungal infection study forum expert recommendations
- Towards global clinical practice guidelines for the management of non-specific low back pain in primary care: a review of current guideline recommendations and how they have changed over the last 30 years
- Challenges and recommendations for implementation of oncology clinical practice guidelines in China
BibTeX
@article{goz2026large,
title = {{Large Language Model Dietary Recommendations for Multimorbidity with Conflicting Guidelines in Turkish and English}},
author = {Büşra Göz and Yılmaz Dikilitaş and Furkan Göz and Ahmet Sayar},
journal = {Nutrients},
year = {2026},
month = nov,
doi = {10.3390/nu18213674},
url = {https://www.mdpi.com/2072-6643/18/21/3674},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Large Language Model Dietary Recommendations for Multimorbidity with Conflicting Guidelines in Turkish and English
AU - Büşra Göz
AU - Yılmaz Dikilitaş
AU - Furkan Göz
AU - Ahmet Sayar
PY - 2026
DA - 2026/11/05
JO - Nutrients
DO - 10.3390/nu18213674
UR - https://www.mdpi.com/2072-6643/18/21/3674
AB - Background/Objectives: Patients with multimorbidity often receive conflicting dietary guideline recommendations, and large language models (LLMs) increasingly draft dietary advice. In this study, we evaluated nine LLMs, including two Turkish-adapted models, on 76 synthetic patient vignettes in Turkish and English. Methods: The vignettes combined type 2 diabetes, chronic kidney disease, hypertension, obesity, gout and celiac disease as single conditions, conflicting pairs, concordant pairs and triples. An LLM judge panel graded the responses against key-feature checklists approved by a dietitian, whose blinded ratings of a subset agreed with the panel on 90% of items. Results: The proprietary models covered 81% to 92% of the weighted checklist and fully resolved 57% to 89% of the guideline conflicts. Their coverage did not differ significantly between Turkish and English in any of these models, whereas across all models Turkish responses covered 2 percentage points less, mainly in conflicting-pair vignettes and in the open-weight models. The GPT and Claude models had the fewest safety-critical flags, 5% to 8% in both languages, and the proprietary models’ Turkish advice was rated 4.9 to 5.0 out of 5 for cultural appropriateness. Conclusions: Expert review of LLM drafts can therefore focus on conflict resolution, and the checklist procedure provides a reusable audit.
SN - 2072-6643
ER -