Prevalence of and risk factors for cognitive problems in people with Multiple Sclerosis: A multi-centre observational study

Abstract

Introduction: Cognitive problems are common among people with Multiple Sclerosis (pwMS), which can be debilitating and impact daily life. An online self-administered Symbol Substitution Task (SST), measuring information processing and working memory, was developed and deployed in three UK MS clinics to screen pwMS for cognitive problems as part of the NEuRoMS programme (www.neuroms.org). Objectives/Aims: This study aimed to explore the prevalence of cognitive problems and related risk factors in pwMS. Method(s): We used a multi-centre observational study design. The levels of cognitive problems using SST normative data were defined as follows: 1) within normal range (-1.49 standard deviations (SD) or higher); 2) mild (-2.49 SD to -1.5 SD); 3) moderate (-2.99 SD to -2.5 SD); 4) severe (-3 SD or lower). The prevalence was calculated from 1376 pwMS, and we performed ordinal logistic regression with cognitive impairment level as a dependent variable. Three models evaluated three outcome measures: 1) Patient Health Questionnaire-8 (PHQ-8), 2) Generalised Anxiety Disorder-7 (GAD-7), and 3) Multiple Sclerosis Self Efficacy Scale (MSSE). Covariates were age and duration of MS; factors were gender, education, MS type, and ethnicity; site as a random effect. Result(s): The SST threshold calculated based on SD from normative data as defined above, and prevalence from 1376 pwMS suggested that 56% were within normal range (SST score: >=35), 25% in mild (25-34), 8% in moderate (20-24) and 12% in severe (0-19) ranges. Covariates found to be significant predictors of cognitive impairment were: higher depression (PHQ-8; 1.08, p<0.001), higher anxiety (GAD-7; 1.06, p=0.01), and lower self-efficacy (MSSE; 0.96, p<0.001). For all three models age, duration of MS, gender (woman vs man), education (higher degree vs below GCSE) and MS type (relapsing-remitting vs primary progressive) had significant odds ratios, which means an increased odds of being in a higher cognitive impairment group. Conclusion(s): Cognitive problems affect 45% of pwMS attending routine clinics, with 33% having mild-moderate cognitive problems who are often left unidentified or treated. Therefore, routine monitoring for cognitive problems is crucial for identification of those who may benefit from timely support. Older, male pwMS with long disease duration, primary progressive MS, lower levels of education, low mood and low self-efficacy appear at higher risk of cognitive impairment. .