The aim of this study was to investigate the factor structure and to conduct an item analysis of the Center for Epidemiological Studies–Depression Scale (CES–D) among people with arthritis from the UK. This 20-item self-report scale was designed by Radloff in 1977 to measure depressive symptomatology in the general population. Data were drawn from a national study and collected through self-administered questionnaires mailed to participants, each of whom had a confirmed diagnosis of arthritis. Reliability and structure of the scale were examined using standard item analysis, internal consistency (Cronbach alpha), and principal components analysis. A four-dimensional structure was identified: Self-worth, Depressed Affect, Positive Affect, and Somatic Disturbance; three items loaded on two factors and were excluded from consideration. Comparisons with previous studies indicated some differences in the subscales and showed that, in this sample, two subscales might be artifacts of measurement. In samples of people with arthritis from the UK, use of the total score may not reflect accurate depressive symptomatology and the subscale scores should be used with caution.