Diagnosing structural narrowing in public digital health information on healthy eating: a stratified search-entry analysis of Xiaohongshu posts
- 1Law School, Qingdao University of Science and Technology
- 2School of Law, Qingdao University
Abstract
Objective: User-generated platforms now function as everyday sources of dietary guidance, making their visible content part of the public digital health information environment. This study diagnoses how healthy-eating posts on Xiaohongshu are organized across health goals, food or method pathways, and applicability conditions, and compares these structures across search entries and levels of public interaction. Structural narrowing is treated as a multidimensional pattern rather than as simple concentration around a search topic.
Methods: We manually accessed publicly searchable image-text posts through the Xiaohongshu mobile application between 28 July 2026 at 17:56:43 and 30 July 2026 at 13:25:47. Six search-entry categories each contained four original Chinese keywords, and 25 qualifying posts were sampled per keyword, producing 600 search-occurrence records. Duplicate views within a keyword were excluded during collection. When the same post appeared under more than one keyword, its search occurrences were retained for entry-level comparison but collapsed into one record when the unique-post corpus was formed. After manual review and deduplication, 594 unique core posts remained. Content was coded for health goals, food and method pathways, templated presentation, advice or effect claims, and condition disclosure. Overall structure and interaction patterns were analyzed using unique posts, while entry differences were examined using occurrence records. Two paired coding rounds were used to calculate Cohen’s kappa for the core fields.
Results: The 594 unique core posts were concentrated around weight/body-shape goals and restriction/control pathways, although the range of food objects was not uniformly narrow. Different search entries showed different combinations of structural features. Weight/body-shape entries emphasized weight goals and restriction/control; bodily-indicator and disease-state entries combined state management with control; method entries organized eating around rhythm-based schemes and templates; product entries shifted attention toward productized objects and weaker condition disclosure; and templated entries foregrounded replicable arrangements. Highly interactive posts were more concentrated in weight, control, and templated structures, but not in insufficient condition disclosure. Among posts that made advice or effect claims, some lacked information on target population, dosage, frequency, duration, or risks.
Conclusion: On Xiaohongshu, narrowing was expressed less through the disappearance of food diversity than through repeated combinations of goals, methods, presentation formats, and applicability conditions. A keyword’s topical focus alone was not treated as narrowing. The diagnosis required multiple features, such as concentrated goals, compressed methods, object substitution, templating, or weakened condition boundaries, to occur together within an entry. The findings support entry-level monitoring of public digital health information while remaining distinct from clinical safety assessment, patient-level research, and claims about user adoption.
Related
- Pinhole‐Like Central Bronchial Narrowing: A Rare Reversible Manifestation of Eosinophilic Bronchial Inflammation
- Narrowing the research to practice gap for trauma-focused psychotherapies in the Veterans Health Administration: A narrative review
- Posterior Pulmonary Vein Isolation Lesion Spacing and Left Atrial Posterior Wall Narrowing During Pulsed Field Ablation.
- Adequately increased postoperative medial proximal tibial angle can prevent medial joint space narrowing after open-wedge high tibial osteotomy: a 3-year cohort study.
- Response to Andrea Truini: Diagnosing a Lesion or Diagnosing Neuropathic Pain?
BibTeX
@article{xu2026diagnosing,
title = {{Diagnosing structural narrowing in public digital health information on healthy eating: a stratified search-entry analysis of Xiaohongshu posts}},
author = {Qingyong Xu and Tong Qi and Hangyu Zhou and Chunchun Sun},
journal = {Frontiers in Public Health},
year = {2026},
month = oct,
doi = {10.3389/fpubh.2026.1977016},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1977016/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Diagnosing structural narrowing in public digital health information on healthy eating: a stratified search-entry analysis of Xiaohongshu posts
AU - Qingyong Xu
AU - Tong Qi
AU - Hangyu Zhou
AU - Chunchun Sun
PY - 2026
DA - 2026/10/20
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1977016
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1977016/full
AB - Objective: User-generated platforms now function as everyday sources of dietary guidance, making their visible content part of the public digital health information environment. This study diagnoses how healthy-eating posts on Xiaohongshu are organized across health goals, food or method pathways, and applicability conditions, and compares these structures across search entries and levels of public interaction. Structural narrowing is treated as a multidimensional pattern rather than as simple concentration around a search topic. Methods: We manually accessed publicly searchable image-text posts through the Xiaohongshu mobile application between 28 July 2026 at 17:56:43 and 30 July 2026 at 13:25:47. Six search-entry categories each contained four original Chinese keywords, and 25 qualifying posts were sampled per keyword, producing 600 search-occurrence records. Duplicate views within a keyword were excluded during collection. When the same post appeared under more than one keyword, its search occurrences were retained for entry-level comparison but collapsed into one record when the unique-post corpus was formed. After manual review and deduplication, 594 unique core posts remained. Content was coded for health goals, food and method pathways, templated presentation, advice or effect claims, and condition disclosure. Overall structure and interaction patterns were analyzed using unique posts, while entry differences were examined using occurrence records. Two paired coding rounds were used to calculate Cohen’s kappa for the core fields. Results: The 594 unique core posts were concentrated around weight/body-shape goals and restriction/control pathways, although the range of food objects was not uniformly narrow. Different search entries showed different combinations of structural features. Weight/body-shape entries emphasized weight goals and restriction/control; bodily-indicator and disease-state entries combined state management with control; method entries organized eating around rhythm-based schemes and templates; product entries shifted attention toward productized objects and weaker condition disclosure; and templated entries foregrounded replicable arrangements. Highly interactive posts were more concentrated in weight, control, and templated structures, but not in insufficient condition disclosure. Among posts that made advice or effect claims, some lacked information on target population, dosage, frequency, duration, or risks. Conclusion: On Xiaohongshu, narrowing was expressed less through the disappearance of food diversity than through repeated combinations of goals, methods, presentation formats, and applicability conditions. A keyword’s topical focus alone was not treated as narrowing. The diagnosis required multiple features, such as concentrated goals, compressed methods, object substitution, templating, or weakened condition boundaries, to occur together within an entry. The findings support entry-level monitoring of public digital health information while remaining distinct from clinical safety assessment, patient-level research, and claims about user adoption.
SN - 2296-2565
ER -