Racial Discrimination, Disease Activity, and Organ Damage: The Black Women's Experiences Living With Lupus (BeWELL) Study

David H. Chae; Connor D. Martz; Thomas E. Fuller-Rowell; Erica C. Spears; Tianqi Tenchi Gao Smith; Evelyn A. Hunter; Cristina Drenkard; S. SamLim


Am J Epidemiol. 2019;188(8):1434-1443. 

In This Article


The mean (standard deviation) SLE activity score in our sample was 15.11 (7.94). More than half of participants (61.8%; n = 264) had a damage score of 2 or more, 22.3% (n = 95) had damage to 1 organ or system, and 15.9% of participants (n = 68) had no major organ damage Participants were an average of 46.7 (SD, 12.3) years and the average time since diagnosis with SLE was 15.9 (SD, 10.3) years. The majority of participants (80.6%; n = 344) reported experiencing racial discrimination in at least 1 domain, with 40.1% (n = 171) reporting experiencing racial discrimination in 5 or more. The most commonly reported domain of racial discrimination was "getting service at a store or restaurant" (65.6%). Participants were least likely to report racial discrimination "getting medical care," although this still represented a relatively large percentage of participants (27.6%). Additional characteristics of our sample are listed in Table 1.

According to linear regression analyses, racial discrimination had a significant bivariate relationship with SLE activity (b = 1.89, 95% confidence interval (CI): 1.16, 2.62). Results from multivariable analyses are reported in Table 2. Adjusting for potential demographic confounders (model 1: age, years since diagnosis, and relationship status), racial discrimination continued to be associated with SLE activity (b = 1.92, 95% CI: 1.18, 2.66). Although socioeconomic characteristics may be considered possible mediators, additional adjustment for these factors (model 2: model 1 plus education, poverty ratio, work status, and insurance status) somewhat increased the magnitude of the association between racial discrimination and SLE activity (b = 2.20, 95% CI: 1.52, 2.89). Adjustment for health-related variables (model 3: model 2 plus body mass index, exercise, smoking status, and use of steroids, hydroxychloroquine, and other immunosuppressants) did not substantively change this relationship (b = 2.00, 95% CI: 1.32, 2.68).

A significant bivariate relationship with racial discrimination (b = 0.09, 95% CI: 0.03, 0.15) was determined from linear regression models in which log-transformed organ damage was examined. Results from multivariable analyses are listed in Table 3. Similar to results from models examining SLE activity, greater reported racial discrimination was associated with higher organ damage scores. This association was also robust to adjustment for demographic factors (model 1: b = 0.08, 95% CI: 0.02, 0.14), socioeconomic characteristics (model 2: b = 0.08, 95% CI: 0.02, 0.14), and health-related variables (model 3: b = 0.08, 95% CI: 0.02, 0.13). The estimate from our final model indicated that each unit increase in the frequency of racial discrimination was associated with an increase of 0.08 units in log-BILD score. Alternatively, the exponential of this estimate, 1.20, indicated that each unit increase in racial discrimination was associated with a 20% increase in the geometric mean of BILD score.