Background & aims Older malnourished patients experience increased length of hospital stay and greater morbidity compared to their well nourished counterparts. This study aimed to assess whether nutritional status at hospital admission predicted clinical outcomes at 12 months follow-up. Methods Secondary data analysis of 2602 consecutive patient admissions to an acute tertiary hospital in New South Wales, Australia on or before 1st June 2009. Twelve-month data was analysed in a sub-sample of 774 patients. Nutritional status was determined within 72 h of admission using the Mini Nutritional Assessment (MNA). Outcomes, obtained from electronic patient records included hospital readmission rate, total length of stay (LOS), change in level of care at discharge, and in-hospital mortality. Results A third (34%) of patients were malnourished and 55% at risk of malnutrition. Using a Cox proportional hazards regression model, controlling for underlying illness and age, patients at risk of malnutrition were 2.46 (95% CI: 1.36, 4.45; p = 0.003) times more likely to have a poor clinical outcome (mortality/discharge to higher level of care), while malnourished patients had a 3.57 (95% CI: 1.94, 6.59; p = 0.000) times higher risk. Conclusions A poor nutritional status carries a substantially greater risk of death and/or loss of dependency in older adults. Interventions to improve the nutritional status of patients during their hospital stay, and following discharge back to the community, are needed to lower the risk of adverse outcomes.