Objective. To determine the factors associated with disease activity in patients with recent-onset (≤5 years) systemic lupus erythematosus (SLE) who were of Hispanic, African-American, or Caucasian ethnicity. Methods. Incident and prevalent cases of SLE, as defined by the American College of Rheumatology criteria for SLE, among the 3 ethnic groups were identified in Alabama (The University of Alabama at Birmingham) and Texas (The University of Texas-Houston Health Science Center and The University of Texas Medical Branch at Galveston). Variables from the socio-demographic, clinical, immunologic, immunogenetic, behavioral, and psychological domains were obtained using validated instruments. Disease activity was ascertained with the Systemic Lupus Activity Measure (SLAM). Stepwise domain regressions with SLAM score as the dependent variable were performed. Final ethnic-specific and overall regression models were obtained by entering variables that were retained in the domain regressions. Results. SLAM scores at study entry were higher in the African Americans (mean ± SD 12.6 ± 6.9) and Hispanics (11.0 ± 6.2) than in the Caucasians (8.5 ± 3.7) (P ≤ 0.001). The final overall regression model (R2 = 28%) for higher SLAM score included the following variables: African-American ethnicity, lack of private health insurance, abrupt disease onset, presence of anti-Ro antibodies, absence of HLA- DRB1*0301, higher levels of helplessness, and abnormal illness-related behaviors. Conclusion. Socioeconomic, immunologic, immunogenetic, behavioral, and psychological variables were all predictive of disease activity early in the course of SLE, irrespective of ethnic group. However, there remain ethnic group differences in disease activity that were not explained by these factors.