Structured proactive nutritional care delivered via a digital follow-up platform is associated with better postoperative outcomes in gastric cancer patients after radical gastrectomy

Objective: To evaluate the association of a structured, proactive nutritional care model delivered via a self-developed digital follow-up platform, compared with routine low-intensity post-discharge follow-up, with nutritional outcomes, clinical endpoints, and process indicators in patients after radical gastrectomy for gastric cancer, compared with routine minimal post-discharge management. Methods: A retrospective controlled study was conducted, including 150 patients who underwent radical resection for gastric cancer between January 2022 and December 2022. The intervention group (n = 100) received care via the digital follow-up platform combined with individualized nutrition prescriptions, while the control group (n = 50) received conventional discharge education, outpatient visits, and telephone follow-ups. Follow-up assessments were performed at 1 week, 1 month, 3 months, 6 months, and 12 months postoperatively. Outcomes included body composition (weight, BMI, FFM, SMM, phase angle), nutritional status (PG-SGA, GLIM criteria), laboratory parameters (albumin, prealbumin, PNI), quality of life (EORTC QLQ-C30, QLQ-STO22), psychological status (HADS), and nutritional prescription compliance rates. The primary outcome was the remission rate of GLIM-defined malnutrition at 12 months. Secondary outcomes included complications, readmission rate, process indicators, and the return to intended oncologic therapy (RIOT). To avoid distortion of median times and time-to-event comparisons, patients who did not receive chemotherapy were treated as right-censored data in survival analyses rather than being recorded as 0 days. Statistical analyses involved linear mixed models, multivariable logistic regression, and inverse probability of treatment weighting (IPTW). Results: After weighting, baseline characteristics were balanced across the 150 patients with complete follow-up data. At 12 months, the intervention group showed less reduction in body weight and BMI (-2.3 ± 1.9 kg vs -4.6 ± 2.5 kg, P < 0.01), better preservation of FFM and SMM (aMD ≈ +0.9 kg, P ≤ 0.004), and a net increase in phase angle (+0.4° vs -0.1°, P = 0.003). The GLIM-defined malnutrition remission rate was 58.0% vs 34.0% (aOR = 2.21, 95% CI: 1.14-4.29). The intervention group showed lower rates of complications (21.0% vs 40.0%) and readmissions (10.0% vs 24.0%), as well as a shorter median RIOT by 13 days (32 days vs 45 days, P = 0.02). Higher compliance rates for energy and protein targets were observed in the intervention group (71%/66% vs 53%/43%, P < 0.001). The timely data upload rate was 89%, and the median alert response time was 14 hours. A dose-response pattern was observed between platform engagement and nutritional improvement, with protein intake compliance identified as a potential partial mediator. Conclusion: A structured nutritional care model facilitated by the digital follow-up platform was associated with better nutritional and clinical outcomes in patients after radical gastrectomy for gastric cancer. These observed associations may be related to enhanced compliance and timely intervention. This model is low-cost, reproducible, and scalable, and may represent a promising approach to postoperative nutritional management.