Objective: To evaluate diagnostic instruments for assessment of nutritional status in patients with rheumatoid arthritis (RA) in relation to objective body composition data. Subjects and methods: Study subjects include 60 in-ward patients (83% women, median age 65 years). Anthropometric measures and the nutritional tools Mini Nutritional Assessment (MNA), Subjective Global Assessment (SGA), Malnutrition Universal Screening Tool (MUST) and Nutritional Risk Screening tool 2002 (NRS-2002. Body composition was determined by dual-energy X-ray absorptiometry and fat-free mass index (FFMI; kg/m 2 ) and fat mass index (FMI; kg/m 2 ) were calculated. Results: Mean body mass index (BMI) for RA women and men were 24.4 and 26.9 kg/m 2 , respectively. Twelve per cent of the women and none of the few men had BMIo18.5 kg/m 2 , that is, the cutoff value for malnutrition. FFMI indicated 52% of the women and 30% of the men to be malnourished. The sensitivity and specificity for BMI to detect malnutrition according to FFMI were 27 and 100%, whereas for arm muscle circumference the sensitivity was 36% and the specificity 89% and for triceps skin fold 43 and 93%, respectively. For MNA, sensitivity was 85% and specificity 39% and for SGA 46 and 82%. Both MUST and NRS-2002 had sensitivity of 45% and specificity of 19%. Conclusion: A large proportion of in-ward RA patients had reduced FFMI. Concurrent elevation of fat mass made BMI a nonreliable tool to detect malnutrition. Of the tested clinical evaluation tools, MNA might be used as a screening instrument, but because of its low specificity it should be followed by body composition determination.