Related Experiment Video
Updated: Jun 6, 2025
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
Association of Social Determinants of Health With Brain MRI Outcomes in Individuals With Pediatric Onset Multiple
Ruby Ross1, Kimberly A O'Neill1, Rebecca A Betensky1
1From the Departments of Neurology (R.R., K.A.O.N., R.K., A.S., L.B.K.), Population Health (R.K.), and Radiology (J.T.L.), New York University Grossman School of Medicine, New York; Department of Neurology (R.R.), Weill Cornell Medical Center, New York; Department of Biostatistics (R.A.B.), New York University School of Global Public Health, New York; icometrix (T.B.), Leuven, Belgium; Department of Pediatrics (M.M.-S.), Baylor College of Medicine, Houston, TX; Jan and Dan Duncan Neurological Research Institute at Texas Children's Hospital (M.M.-S.), Houston; and Department of Pediatrics (H.D.M., M.W.), University of Utah, Salt Lake City.
Background And Objectives:
Accumulating evidence points to worse clinical outcomes among adults with multiple sclerosis (MS) belonging to minority or poverty-affected groups. By contrast, little is known about the outcomes of these populations with pediatric-onset MS (POMS). Individuals with POMS represent 5% of the MS population and are more racially diverse yet have been understudied regarding socioeconomic environment or characteristics. In this study, we investigated the association between childhood social determinants of health (SDOH) and brain MRI outcomes in patients with POMS.
Methods:
This is a retrospective single-site cohort study of patients with POMS with brain MRI quantitatively analyzed using icobrain software to yield total white matter lesion, black hole, whole brain, white matter, and gray matter volumes. All patients with POMS evaluated at New York University Langone MS Center and who underwent high-quality volumetric MRI scans were included in this study. SDOH indicators of race, ethnicity, health insurance type, parental education, and childhood neighborhood social vulnerability index (SVI) were examined for association with MRI outcomes using linear least absolute shrinkage selection operator penalized regression modeling. Disease-modifying therapy (DMT) timing and DMT efficacy were compared for each SDOH category.
Results:
A total of 138 patients with POMS (70% female) were included with a mean age of 19.86 years and median disease duration of 4 years at time of scan. Public health insurance, Black race, Hispanic ethnicity, low parental education, and high SVI (greater neighborhood disadvantage) were each associated with white matter lesion and black hole volume. SVI was the strongest individual predictor of total white matter lesion (β = 4.63, p = 0.002) and black hole volume (β = 2.91, p = 0.003). In models incorporating all SDOH variables, public health insurance was the strongest predictor of total lesion (β = 2.48, p = 0.01) and black hole volume (β = 1.50, p = 0.02), attenuating the effect of SVI (β = 1.66, p = 0.33 and β = 1.00, p = 0.39). There were no differences in DMT timing or efficacy between categories of social disadvantage.
Discussion:
Individual-level and neighborhood-level indicators of social disadvantage are associated with worse brain MRI outcomes in POMS. Further investigation of race, ethnicity, and childhood disadvantage as risk factors of MS susceptibility and severity is needed to reduce MS health disparities.
