Aim. The aim of the study was to determine whether a relationship exists between nutritional status, assessed by three recommended instruments, and body composition for older persons with Chronic Obstructive Pulmonary Disease (COPD). Background. Commonly reported symptoms of COPD include impaired nutritional status. This is difficult to assess accurately, because currently recommended instruments for risk identification do not evaluate body composition reliably. Design. A descriptive and correlational study; data were collected in 2007. Method. Study participants were 47 women and 34 men with a mean age of 65 years and 48% with very severe COPD. Data collection included body mass index (BMI), waist–hip ratio (WHR), fat-free mass index (FFMI), lung functions and assessment of nutritional status by the mini nutritional assessment (MNA), the malnutrition universal screening tool (MUST) and a national disease-specific instrument (ENS). Results. Participants' BMIs ranged from 15·9 to 43·8. Out of 15 with very low FFMIs five were classified as overweight. One of those participants was identified as malnourished by the MNA, two were identified by the MUST as at high risk for malnutrition and none was identified by the ENS as in need of qualified help with nutrition. Six participants were assessed using each instrument as being in need of special attention for their nutritional status. Correlations between each of the instrument and FFMIs were significant at the 0·01 level for each instrument. Conclusion. A combination of methods gives richer information as single measurements may give misleading results. Assessment of nutritional status cannot rely only on instrument values; there is a need for an astute 'clinical eye' as well. Relevance to clinical practice. The methods are suitable for skilled professional use in clinical practice with people with COPD, but the short form of the MNA and the ENS should be used with caution