Background: 5-HT3 antagonists are effective in reducing the acute nausea and vomiting caused by cancer chemotherapy. However, it is not clear whether continuing these agents beyond twenty four hours is useful in controlling emesis on days two to seven after chemotherapy. Patients and methods: Four hundred seven patients receiving moderately emetogenic chemotherapy who had been given dexamethasone 8 mg i.v. and either ondansetron 32 mg i.v. or dolasetron 2.4 mg/kg i.v. were randomized to continue either an oral form of their 5-HT3 antagonist (ondansetron 8 mg b.i.d. or dolasetron 200 mg daily) plus dexamethasone 8 mg p.o. daily or dexamethasone alone for days two to seven. Endpoints assessed by self-report were: 1) complete control (no vomiting, no rescue medications, no missing data) of emesis; 2) nausea severity; and 3) quality-of-life as measured by the EORTC QLQ-C30. Results: Continuation of 5-HT3 antagonists improved slightly, but not significantly, the complete control rate (47% vs. 41%; P = 0.24 one-sided) after chemotherapy. However, mean nausea severity was significantly (P = 0.015 one sided) reduced (by 3 mm on a 10 cm scale) on the combined arm. Minimal differences in quality of life were observed. Conclusion: The benefit of continuing 5-HT3 antagonists beyond 24 hours is modest and the merits of routine use in these circumstances debatable.