Thresholds for defining cognitive impairment in people with multiple sclerosis using an electronic version of a symbol substitution task

Abstract

Introduction: Up to 70% of people with Multiple Sclerosis (MS) experience cognitive problems which can be debilitating and impact on day-to-day life. A digital Symbol Substitution Task (SST), a variant of the Symbol Digit Modalities Test (SDMT), was developed as part of a multi-centre project aiming to develop a neuropsychological pathway to routinely assess people with MS attending UK MS clinics for cognitive problems (NEuRoMS; www.neuroms.org). This work presents thresholds for determining cognitive impairment using the newly developed SST. Method(s): This is a cross-sectional study. Data were collected for control participants who do not have MS. The levels of cognitive impairment in people with MS were defined as follows, based on previously published work on a similar digital variant of SDMT known as Multiple Screener: 1) None evident: Score of less than 1.5 standard deviations (SD) below the mean; 2) Possibly mildly impaired: Scores of 1.5 SD or more, and less than 2 SD below the mean; 3) Probably impaired: Score of 2 SD or more below the mean. Data for the SST were also collected for people with MS as part of routine clinical care in three National Health Service (NHS) Trusts. The thresholds defined above were applied to data for people with MS to establish the prevalence of cognitive impairment. This abstract presents findings from a planned interim analysis. Data collection is ongoing. Result(s): People with MS (n=632) were mostly female and of mean age 49.1. Most people with MS had relapsing-remitting MS, and had been diagnosed an average of 18.7 years. Control participants (n=41) were mostly female with a mean age of 44.3 years. From control participants who completed the task, the mean score was 47.9, SD 8.6. The thresholds for defining that a patient is possibly mildly impaired and probably impaired are 35.1 and 30.8 respectively. Applying these thresholds to the MS cohort to establish cognitive impairment, there were 329 (52.1%) who were classified as 'None evident', 104 (16.5%) who were classified as 'Possibly mildly impaired', and 199 (31.5%) who were classified as 'Probably impaired'. Discussion(s): Pervious research has mainly categorised people into 'impaired' or 'not impaired'. This work however has generated thresholds for defining the severity of cognitive impairment in people with MS using an electronic version of the SST, enabling clinicians to interpret test results more accurately. Further data will be collected to refine these thresholds.