Abstract Background: Many patients admitted to acute hospital services are underweight or harbour vitamin deficiencies. Objectives: To determine the effect on patient throughput of a policy of routine vitamin supplementation, and of early routine sipfeed supplementation in ‘thin’ patients (5–10% weight loss or body mass index 18–22). Design: Factorial randomized placebo controlled trial of oral multivitamins from the first day of admission, and, after nutritional screening, of a nutritionally complete sipfeed from the second day in ‘thin’ patients. Setting: Acute medical, surgical and orthopaedic hospital services of a London teaching hospital. Participants: 1561 patients admitted as emergencies were included in the vitamin study of which 549 were included in the sipfeed study. Main Outcome Measure: Length of hospital stay (LOS). Results: Offering multivitamins to acute admissions resulted in a mean change (reduction) in LOS of −0.4 days 95% CI (−2–1.2days). The results suggest greater reductions for those discharged after 10 days: mean change=−2.3 days 95% CI (−5.7 to 1.2). Sipfeed supplementation was associated with an increased mean length of stay 2.8 days 95% CI (−0.8–6.3). 18% of acute admissions were classified undernourished on the basis of BMI, MUAC or percent weight loss combined. Conclusions: No benefit was observed for sipfeed intervention although a small benefit of less than one day is not excluded. Vitamin supplementation may have slight but economically important benefit.