Objective To determine the prevalence of malnutrition risk in older people across three settings. Methods Older people living in the community or newly admitted to hospital or residential care were assessed for malnutrition risk using the validated Mini‐Nutritional Assessment – Short Form and dysphagia risk using the Eating Assessment Tool‐10. Demographic, physical and health data were collected. Results Of 167 participants, 23% were malnourished and 35% were at high risk of malnutrition. Those recently admitted to residential care versus a hospital or living in the community had a higher prevalence of malnourishment (47% vs 23% and 2%) (P < 0.001). Risk of dysphagia differed with settings (P < 0.001) with highest risk in residential care. Hospitalised and residential care participants were significantly more likely to have ≥4 comorbidities, take ≥5 medications and have below normal cognition compared to community participants. Conclusion Choice of nutrition intervention is setting dependent.