We studied the reliability and validity of the Chinese Short-Form 36 Health Survey (SF-36) in a cross sectional study of patients with systemic lupus erythematosus (SLE). Sixty-nine consecutive subjects completed a questionnaire containing the Chinese SF-36 twice within 14 d. Disease activity and damage were assessed using the British Isles Lupus Activity Group (BILAG) and SLICC/ACR Damage Index (DI) scales, respectively. Internal consistency was assessed using Cronbach's alpha, reliability using Spearman's correlation and repeatability coefficients, and relationships between SF-36, BILAG and DI scores using Spearman's correlation. The Chinese SF-36 showed high internal consistency (alpha = 0.72-0.91) and good reliability, with correlations exceeding 0.70 for 7 scales and mean scale score differences of < 2 points for 6 scales. SF-36 scores correlated weakly with BILAG scores (-0.27 to -0.41) and DI scores (-0.24 to -0.35), and subjects' mean SF-36 scores were 6-24 points lower than the general population, supporting construct validity of the SP-36. These data suggest that the Chinese SF-36 is a reliable and valid measure of quality of life in patients with SLE.