Junxiao Yang, Jingya Tai, Chunli Dong
The available evidence is most suitable for mapping technologies and proximal signals, not for firm conclusions about independent clinical effectiveness. Multicenter comparative studies should link validated measurement to documented clinical responses and evaluate generalizability, economics, and equitable access.
BACKGROUND: Digital technologies increasingly support inpatient meal ordering, dietary intake recording, automated measurement, decision support, and foodservice workflows, but evidence across technical, service, nutritional, and clinical outcomes remains fragmented.
OBJECTIVE: To map digital technologies embedded in hospital foodservice and nutrition care, separate technical-validation, service process, and patient outcomes, and identify implementation and translational gaps.
METHODS: PubMed, Embase, Web of Science Core Collection, and Scopus were searched from inception to 30 June 2026. Two reviewers independently screened records and appraised included reports; one reviewer charted data and a second checked all entries. Findings were synthesized descriptively without meta-analysis or certainty grading.
RESULTS: Twenty-five reports published between 1996 and 2026 were included; most were single site, nonrandomized, or early stage validation studies. Technical-validation studies reported improved completeness or agreement for some digital measurement tools, but artificial-intelligence performance varied and no artificial-intelligence system had independent external validation. Service-process studies reported favorable signals for ordering flexibility, documentation, selected foodservice costs, and plate waste in some settings. Several nonrandomized comparative studies reported higher energy and/or protein intake, but digital functions were usually bundled with menu, staffing, or delivery redesign, so the independent effect of the digital component remains uncertain. One randomized trial improved nutrition documentation and care planning without changing weight or length of stay. Sustained nutritional recovery, downstream clinical outcomes, formal cost-effectiveness, and equity were rarely evaluated.
CONCLUSION: The available evidence is most suitable for mapping technologies and proximal signals, not for firm conclusions about independent clinical effectiveness. Multicenter comparative studies should link validated measurement to documented clinical responses and evaluate generalizability, economics, and equitable access.