Multiple Sclerosis Journal - October 2017 - 1555

RE Ventura, AO Antezana et al.
Table 1. Demographics and P-MSSS for CA, AA, and HA patients in combined cohort, as well as insurance information
for NYU cohort.
Caucasian

Hispanic

African American

N (% of total cohort)
% Female
Age (mean ± SD), years

1255 (66)
71
48.0 ± 12.0

391 (21)
81; p = 0.01*
44.4 ± 12.4; p = 0.008*

Disease duration (mean ± SD), years

13.5 ± 9.8

% Disease-modifying therapy
% RRMS
P-MSSS, total cohort, mean (SD),
ANCOVA controlled for age and gender
NYU cohort, n (%)
% Private insurance
% Medicaid
% Medicare
% Other
P-MSSS, NYU cohort, mean (SD),
ANCOVA controlled for age, gender,
and insurance

80
74
3.4 (2.6)

258 (14)
78; p < 0.0001*, p = 0.05**
41.6 ± 10.9; p < 0.0001*,
p = 0.002**
10.7 ± 7.1; p < 0.0001*,
p = 0.26**
81; p = 0.40*, p = 0.36**
82; p = 0.01*, p = 0.001**
3.9 (2.6); p < 0.0001*,
p = 0.11**
190 (18)
66
26
7
<1
4.0 ± 2.6; p < 0.0001*,
p = 0.17**

640 (59)
77
14
8
1
3.4 ± 2.6

11.3 ± 8.1; p < 0.0001*
79; p = 0.80*
78; p = 0.11*
4.5 (3.0); p < 0.0001*
250 (23)
57
33
9
<1
4.5 ± 3.0; p < 0.0001*

P-MSSS: Patient-Derived Multiple Sclerosis Severity Score; CA: Caucasian Americans; AA: African Americans; HA: Hispanic
Americans; SD: standard deviation; RRMS: relapsing-remitting multiple sclerosis; ANCOVA: analysis of covariance.
Significant p-values in bold.
*AA and HA compared to CA.
**HA compared to AA.

Insurance information was obtained from the NYU
cohort and was grouped into Private, Medicaid,
Medicare, and Other. P-MSSS values were calculated
using the published reference table.11
P-MSSS was compared between the CA, HA, and
AA groups within the NYU and Barnabas datasets
using SPSS version 23.0. We used analysis of covariance (ANCOVA) adjusted for age, gender, and
insurance type when applicable. For all comparisons, we used the two-tailed t-test; p < 0.05 was considered significant.
The study was approved by the institutional review
boards (IRBs) of the respective institutions.
Results
The combined NYU and Barnabas cohorts enrolled
1255 CA (66%), 391 AA (21%), and 258 HA (14%)
patients during study period. The demographics of
the three groups are presented in Table 1. There
were significantly more females in the AA and HA
cohorts than the CA; the HA and AA patients were
significantly younger than the CA, and the HA and
AA had a significantly shorter mean disease

journals.sagepub.com/home/msj

duration than the CA. The proportion of patients on
disease-modifying therapies was similar across the
groups. In terms of disease subtype, the proportion
of patients with relapsing-remitting multiple sclerosis (RRMS) was significantly higher in HA as
compared to CA or AA.
We ran an ANCOVA with race (CA, AA, and HA) as
the independent variable and P-MSSS as the dependent variable while controlling for age and gender. We
found that the covariate of age (F = 31.0, p < 0.0001)
had a direct impact on P-MSSS, but gender (F = 2.3,
p = 0.13) did not. Race had a significant impact on
P-MSSS after controlling for age and gender
(F = 23.4, p < 0.0001). The P-MSSS for HA (3.9 ± 2.6)
adjusted for age and gender was significantly higher
than for CA (P-MSSS = 3.4 ± 2.6; p < 0.0001), but
was not significantly different than for AA
(P-MSSS = 4.5 ± 3.0; p = 0.11). AA also had a significantly higher P-MSSS than CA (p < 0.0001). Given
that there was a significantly different proportion of
RRMS patients in each ethnic subtype, we analyzed
the RRMS patients in each ethnic group separately
and found that there was still a significantly higher
P-MSSS for HA (p < 0.0001) and AA (p = 0.002) than
CA when controlling for age and gender.

1555
http://journals.sagepub.com/home/msj
Table of Contents for the Digital Edition of Multiple Sclerosis Journal - October 2017
