Background/Purpose:
Health disparities in SLE are well established, with African American women being disproportionately impacted. Despite the known associations between sociodemographic and genetic factors with SLE, there is limited research into epidemiologic factors and research integrating molecular and epidemiologic data to understand the combined role of these factors in SLE disparities. The All of Us Research Program is collecting both genetic and epidemiologic data, thus promises to be a valuable resource to close this research gap. However, its reliance on volunteer participants may introduce biases that compromise the external validity of epidemiological studies. The goal of this study is to compare sociodemographic data from African American women with SLE between the nation-wide All of Us (AoU) and our research population from coastal South Carolina, to assess whether sociodemographic differences exist between these studies.

Methods:
Our study population consists of 138 African American women meeting either the 1997 ACR revised criteria or the SLICC criteria for SLE who have been recruited in our ongoing Social Factors, Epigenomics, and Lupus in African American women (SELA) study. The All of Us (AoU) dataset includes self-report questionnaire data from 633,540 participants, of which 309 are African American women diagnosed with SLE (identified by ICD-9 710.0 or ICD-10 M32 codes (excluding M32.0)) who have completed “The Basics,” “Lifestyle,” and “Social Factors of Health” surveys. We subset these individuals and compared their survey responses to those asked in the SELA study, including sociodemographic data and questions aligning with the Medical Outcomes Study Social Support Survey (MOS-SS). MOS-SS scores between the two groups were analyzed using a Wilcoxon rank-sum test. Sociodemographic data was analyzed using a series of Chi-squared tests.

Results:
The African American women with SLE in the AoU dataset demonstrate less social support than those of the SELA dataset (p = 0.0022). They are more likely to have been divorced, widowed, or separated (p = 0.0034) or single (p = 0.0013) as opposed to married. The women in the SELA cohort are more likely to be employed (p = 0.0005) than the AoU cohort, and the AoU cohort has a higher proportion of participants that are under the federal poverty line (p = 0.0033). However, the proportion of total participants defined as low income (under 200% of the federal poverty line) does not differ significantly between the two cohorts. The women in the AoU cohort are more likely to have a history of smoking (p = 0.0025) but show no significant difference in binge drinking behavior with the SELA cohort.

Conclusion:
Our study may suggest key differences between these two datasets in terms of social support and sociodemographics that researchers should be aware of before using these datasets for further genomic and epidemiological research. These results reflect the social, economic, and behavioral diversity of populations with SLE. Awareness and inclusion of these diverse populations in research is paramount to the development of precision public health strategies to close the health disparities gap.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/comparison-and-assessment-of-the-all-of-us-dataset-for-epidemiologic-studies-of-systemic-lupus-erythematosus-sle-among-african-american-women/