Fatigue Level is Associated with Sleep Quality but not Vector Magnitude in Multiple Sclerosis

Abstract

INTRODUCTION: Persons with multiple sclerosis (MS) often experience elevated fatigue that may be associated with poor sleep quality and therefore decreased quality of life. The association between fatigue and sleep quality might additionally be related to physical activity (PA) level, which can be objectively assessed using vector magnitude (VM) data from tri-axial accelerometry. PURPOSE: To investigate if there is an association between sleep quality and fatigue level in MS, and to study if associations exist between fatigue level and PA in persons with MS who have good or poor sleep quality. METHODS: Forty-seven persons with MS (48 ± 2 years, length of diagnosis 10 ± 1 year, 9 males) completed the Fatigue Severity Scale (FSS) and Pittsburgh Sleep Quality Index (PSQI) as markers of quality of life, and one week of PA assessment via free living accelerometry (analyzable data was obtained from a subset of 40 subjects). Subjects were classified as Bad Sleepers (BAD) or Good Sleepers (GOOD) based on PSQI scores (>5 and ≤5, respectively), and High Fatigue (HIGH) or Low Fatigue (LOW) based on the FSS (≥4 and <4, respectively). A chi-square analysis was performed between sleep quality and fatigue level. Subsequently, correlations between fatigue level and PA using average VM for GOOD and BAD subjects were calculated for the subset of 40 subjects (N=18 and N=22 for GOOD and BAD, respectively). Statistical significance for all analyses was set at p<0.05. RESULTS: Subjects who reported poor sleeping habits were significantly more likely to have high fatigue than those with good sleeping habits: BAD-HIGH=22, BAD-LOW=4, GOOD-HIGH=12, GOOD-LOW=9 (p=0.036). PA as determined by average VM was not significantly correlated with fatigue level in either the GOOD or BAD group (p=0.564 and p=0.269, respectively). DISCUSSION: These data suggest that poor sleep quality in MS significantly increases the likelihood of elevated fatigue compared to those with good sleeping habits. Additionally, analysis of PA using VM does not appear to demonstrate an association with fatigue level regardless of sleeping status. Further investigation of this potential relationship using alternative objectively-obtained PA variables is warranted.