Taking the tester out of the SDMT: A proof of concept fully automated approach to assessing processing speed in people with MS. (October 2019)

• Record Type: Journal Article
• Authors: Patel, Viral Prakash; Shen, Lingkai; Rose, Jonathan; Feinstein, Anthony
• Is Part Of: Multiple sclerosis. Volume 25:Number 11(2019)
• Page Start: 1506
• Page End: 1513
• Publication Date: 2019-10
• Subjects: Multiple sclerosis -- cognitive assessment -- information processing speed -- computerized testing -- automated testing; Central nervous system -- Diseases -- Periodicals; Myelin sheath -- Diseases -- Periodicals; Inflammation -- Periodicals; Multiple sclerosis -- Periodicals; Central Nervous System Diseases -- Periodicals; Demyelinating Diseases -- Periodicals; Inflammation -- Periodicals; Multiple Sclerosis -- Periodicals; Système nerveux central -- Maladies -- Périodiques; Gaine de myéline -- Maladies -- Périodiques; Inflammation (Pathologie) -- Périodiques; Sclérose en plaques -- Périodiques; Electronic journals; 616.834005
• DOI: 10.1177/1352458518792772
• Languages: English
• ISSNs: 1352-4585

Abstract

Background: One factor hindering the widespread use of cognitive testing for people with multiple sclerosis (pwMS) is the need for a tester to administer tests. Objective: To undertake a proof of concept study assessing the feasibility of a fully automated speech recognition version of the Symbol Digit Modalities Test (auto-SDMT) in detecting abnormalities in processing speed in pwMS. Methods: A sample of 50 pwMS and 32 matched healthy control (HC) subjects was tested with the auto-SDMT and the Brief International Cognitive Assessment for MS (BICAMS). Results: The percentages of MS participants impaired on the auto-SDMT and the traditional oral SDMT were 34% and 32%, respectively. Excellent convergent validity was found between the two tests (MS: r = −0.806, p < 0.001 and HC: r = −0.629, p < 0.001). The auto-SDMT had a similar sensitivity and specificity to the traditional oral SDMT in predicting overall impairment on the BICAMS. Conclusion: The auto-SDMT is a sensitive measure for detecting processing speed deficits in pwMS. The test, the first entirely computer administrated oral response version of the SDMT, uses speech recognition technology, thereby eliminating the need for a human tester. Replication of the results is required in a larger representative sample of pwMS.