To evaluate the antiemetic eŠects of 5-HT 3 receptor antagonists, we investigated the relationship between condition of food intake and occurrence of nausea and vomiting. We collected data such as sex, age, disease, combination of steroids and central antiemetic agents, eating condition, and vomiting condition from medical records in 33 hematologic cancer patients receiving chemotherapy; combination with 5-HT 3 receptor antagonists. The conditions of food intake and nausea/vomiting were checked at 4 mealtime points (lunch, supper, breakfast, and next lunch) after chemotherapy, and were recorded as 1, 3, or 5 as each condition score. To calculate eating scores and nausea/vomiting scores, the sum of scores from 4 mealtime points was used. We found a signiˆcant negative correlation between eating scores and nausea/ vomiting scores (n=62, p<0.01). At eating points in which combination therapy with steroids and central antiemetic agents was not given, antiemetic eŠects of granisetron, azasetron and ramosetron were compared and revealed that azaseton was the most eŠective antiemetic agent. This result is inagreement with our previous study predicting antiemetic eŠects of 5-HT 3 receptor antagonists based on the receptor occupancy theory. This study suggests that eithes receptor occupancy or eating score is a useful indicator for assessment of the e‹cacy of 5-HT 3 receptor antagonists.