Background: Multiple sclerosis (MS) is more prevalent in White Americans (WAs), yet incidence is now highest in Black Americans (BAs), who are prone to more aggressive disease, increased lesion burden, and greater ambulatory impairments. Factors driving racial disparities in MS, such as social or health factors, are not known and have not been comprehensively explored.
Objectives: To identify mediators of racial disparities in MS ambulatory impairment between WAs and BAs with MS.
Methods: The source population was relapsing remitting (RR) MS patients engaging in long-term standard care at a tertiary MS referral center between 2008 to 2015. There were 1,795 RRMS patients with at least two encounters and at least one recorded Timed 25 Foot Walk (T25FW) measure. Age, sex, race (WA or BA only), MS disease duration, disease modifying therapy (DMT) status (yes/no) BMI, hypertension (HTN) (yes/no), smoking status (ever/never), and ZIP code of residence were abstracted from the medical records; zip code was used to estimate median household income. We conducted longitudinal mediation analysis using a latent growth modeling framework and cross-sectional mediation analysis using a Bayes estimator. Mediators of interest were median household income, BMI, HTN, and smoking status; covariates included age, sex, disease duration, and DMT status, and were examined for baseline and change-over-time disparities.
Results: Sex and age did not differ between BA (n=175) and WA (n=1620) RRMS patients. BAs had significantly slower T25FW speeds (4.6 vs 5.0 ft/s; p=4×10-7), shorter disease duration (7.6 vs 9.6 years; p=0.003), lower DMT usage (62% vs 71%; p=0.007), higher HTN burden (30% vs 17%; p=0.00002), were more likely to live in lower income neighborhoods ($44100 vs $56491, p=3×10-19), and were less likely to have ever smoked (36% vs 51%; p=0.0004) than WAs. There was no significant growth in T25FW after accounting for baseline mediation relationships when evaluating longitudinal mediation; thus, we reduced our model to a multivariable cross-sectional model. At baseline and adjusting all covariates, racial disparity was mediated by BMI (13%), neighborhood income (12%), and HTN (10%, p<0.05). In total, 35% of the racial disparity in ambulatory impairment experienced by BA RRMS patients was mediated by these social and health inequities. Conclusions: We add novel insights to drivers of racial disparities in ambulatory impairment in RRMS patients. Social inequities (i.e., neighborhood SES) and adverse health states underlie for a significant portion of the observed ambulatory impairment in BAs with RRMS; however, the majority of the racial disparity remains unexplained. Additional research is necessary to further our understanding of the root causes of racial disparities in MS with the long-term objective of health equity.