Objective: In this longitudinal study, we monitored two large groups of multiple sclerosis
(MS) patients and healthy controls (HC) for 2 years, with the aim of comparing their
neuropsychological profile over time.
Method: Three hundred and twenty-two patients with MS and 303 HC were
administered the Brief Repeatable Battery of Neuropsychological tests (BRB-N);
neuropsychiatric measures were also administered. Two follow-ups were scheduled at
1 and 2 years, respectively.
Results: A linear mixed model (LMM) with random intercept was run by considering
participants' performance on each test of the BRB-N at the three assessment points
(baseline and follow-ups) as the within-subjects variable, and group (patients and
controls) as the between-subjects factor. The interaction term was statistically significant
for the tests: Symbol Digit Modalities test (SDMT) (p D 0.044), Paced Auditory Serial
Addition test (PASAT) (p D 0.011) and Word List Generation (WLG) (p < 0.001), whereas
for the PASAT-3 approached statistical significance (p D 0.05). In addition, a LMM
with random intercept was also run by identifying three groups (controls, relapsingremitting
course of MS (i.e. RR-MS), and prog-MS). The interaction term was statistically
significant for: PASAT-3 (p D 0.017), PASAT-2 (p D 0.0026), and WLG (p D 0.0022).
Conclusions: Our results corroborate on a very large scale evidence that the abilities
tapped by the tasks SDMT, PASAT and WLG are particularly sensitive to MS, and further
extend this issue by showing that these abilities are likely to be more sensitive than
others to the progression of the disease, as compared to HC.
