Objectives: The aim of the study was to validate a diet history interview (DHI) method and a 3-day activity registration (AR) with biological markers. Subjects and study design: The reported dietary intake of 33 rheumatoid arthritis patients (17 patients on a Mediterranean-type diet and 16 patients on a control diet) participating in a dietary intervention study was assessed using the DHI method. The total energy expenditure (TEE), estimated by a 3-day AR, was used to validate the energy intake (EI). For nine subjects the activity registration was also validated by means of the doubly labelled water (DLW) method. The excretion of nitrogen, sodium and potassium in 24-h urine samples was used to validate the intake of protein, sodium and potassium. Results: There was no significant difference between the EI and the TEE estimated by the activity registration or between the intake of protein, sodium and potassium and their respective biological markers. However, in general, the AR underestimated the TEE compared to the DLW method. No significant differences were found between the subjects in the Mediterranean diet group and the control diet group regarding the relationship between the reported intakes and the biological markers. Conclusion: The DHI could capture the dietary intake fairly well, and the dietary assessment was not biased by the dietary intervention. The AR showed a bias towards underestimation when compared to the DLW method. This illustrates the importance of valid biological markers. © 2005 Nature Publishing Group. All rights reserved.