To construct a nomogram prediction model for the risk of malnutrition in patients with esophageal cancer EC during periradiotherapy and to evaluate the predictive performance of the model. Method From March 2019 to March 2022 194 EC patients who underwent radiotherapy in our hospital were taken One day after radiotherapy scored Patient-generated Subjective Global Assessment PG-SGA and serum albumin level were used for malnutrition screening. According to the presence or absence of malnutrition they were grouped into a malnutrition group PG-SGA score ≥4 and serum albumin <35 g / L and a good nutrition group PG-SGA score <4 or serum albumin ≥35 g / L . Multivariate Logistic regression analysis was applied to determine the influencing factors of malnutrition risk in EC patients during periradiotherapy and a nomogram prediction model was constructed based on the predictive factors and then the discriminativeness and precision of the model were verified by the area under the ROC curve AUC and H - L goodness of fit. Result Malnutrition occurred in 83 EC patients and the incidence of malnutrition was 42. 78% univariate analysis showed that there were significant differences between the malnutrition group and the good nutrition group in 9 factors including age thin monthly household income per capita whether to undergo surgery tumor stage the number of accompanying gastrointestinal symptoms concurrent chemotherapy dysphagia and depression P < 0. 05 Logistic regression analysis showed that the risk factors of malnutrition in EC patients during periradiotherapy were age ≥ 60 years old weight loss tumor stage > stage Ⅱ number of accompanying gastrointestinal symptoms ≥ 2 concurrent chemotherapy and dysphagia P < 0. 05 regression formula of malnutrition assessment in EC patients during periradiotherapy Logit P = -11. 563+0. 958×age+1. 167×weight loss+ 0. 915 × tumor stage + 0. 903 × number of accompanying gastrointestinal symptoms + 1. 306 × concurrent chemotherapy + 0. 839 × dysphagia the above 6 risk factors were introduced into R software to establish a nomogram model the AUC was 0. 786 > 0. 75 95%CI = 0. 722-0. 850 the calibration curve and the ideal curve fit well and goodness-of-fit H-L test χ 2 = 7. 114 P = 0. 524 indicating that the nomogram model has good discrimination and accuracy in predicting malnutrition in EC patients during periradiotherap. Conclusion The nomogram prediction model constructed based on six factors including age ≥60 years weight loss tumor stage >Ⅱ number of accompanying gastrointestinal symptoms ≥2 concurrent chemotherapy and dysphagia has a good effect on the risk of malnutrition in EC patients during periradiotherapy The predictive role of esophageal cancer can provide reference for oncology medical staff to take preventive nursing intervention in the periradiotherapy period of esophageal cancer patients in a timely manner.