Objective To understand the application of nutrition risk screening 2002 (NRS 2002), patient-generated subjective global assessment (PG-SGA), body mass index (BMI), serum transferrin (TF) in nutritional screening and assessment of the hospitalized patients with gastric cancer and analyze the correlation between them. Methods 136 patients with gastric cancer hospitalized in our hospital from January 2015 to September 2016 were consecutively enrolled in this study. Nutritional risk screening by NRS 2002 and assessment of nutritional status by PG-SGA were performed in their first 24 hours in hospital. Their BMI, TF were measured at the same time. Results There were 64 patients (47%) of non nutritional risk (NRS 2002<3), while 72 patients (53%) of nutritional risk (NRS 2002≥3). There were 5 patients (4%) of eutrophy (PG-SGA 0~1), 21 patients (15%) of suspected malnutrition (PG-SGA 2~3) and 49 patients (36%) of moderate malnutrition (PG-SGA 4~8) while there were 61 patients (45%) of severe malnutrition (PG-SGA≥9). The age was positively correlated with NRS 2002 and PG-SGA (P<0.01). The TF, BMI, body weight were negatively correlated with NRS 2002 and PG-SGA (P<0.05), and the body height was not correlated with NRS 2002 and PGSGA (P>0.05). There was a significant correlation between NRS 2002 and PG-SGA score (r= 0.565, P<0.01). Conclusions There is high incidence of preoperative malnutrition and nutritional risk in patients with gastric cancer. NRS 2002, PG-SGA and BMI, TF can be beneficial for the overall assessment of nutritional status and the analysis of nutritional risk for patients with gastric cancer. The combination of them may be more helpful in perioperative nutritional treatment in patients with gastric cancer.