Advertisement

Cognitive-motor interference in people with mild to moderate multiple sclerosis, in comparison with healthy controls

  • Andreas Wallin
    Correspondence
    Corresponding author at: Division of Physiotherapy 23100, Department of Neurobiology Care Sciences and Society, Karolinska Institutet, Alfred Nobels Allé 23, Huddinge 14183, Sweden.
    Affiliations
    Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, Sweden

    Rehab Station Stockholm, Research and Development Unit, Solna, Sweden
    Search for articles by this author
  • Erika Franzén
    Affiliations
    Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, Sweden

    Women's Health and Allied Health Professionals Theme, Medical Unit Occupational Therapy and Physiotherapy, Karolinska University Hospital, Stockholm, Sweden

    Stockholm Sjukhem Foundation, R&D Unit, Stockholm, Sweden
    Search for articles by this author
  • Lucian Bezuidenhout
    Affiliations
    Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, Sweden

    Faculty of Community and Health Sciences, University of Western Cape, Cape Town 7535, South Africa
    Search for articles by this author
  • Urban Ekman
    Affiliations
    Department of Neurobiology, Care Sciences and Society, Division of Clinical Geriatrics, Karolinska Institutet, Stockholm, Sweden

    Women's Health and Allied Health Professionals Theme, Medical Unit Medical Psychology, Karolinska University Hospital, Stockholm, Sweden
    Search for articles by this author
  • Fredrik Piehl
    Affiliations
    Department of Clinical Neuroscience, Center for Molecular Medicine, Karolinska Institutet, Stockholm, Sweden

    Department of Neurology, Karolinska University Hospital and Neuroimmunology Unit, Stockholm, Sweden
    Search for articles by this author
  • Sverker Johansson
    Affiliations
    Department of Neurobiology, Care Sciences and Society, Division of Physiotherapy, Karolinska Institutet, Huddinge, Sweden

    Women's Health and Allied Health Professionals Theme, Medical Unit Occupational Therapy and Physiotherapy, Karolinska University Hospital, Stockholm, Sweden
    Search for articles by this author
Open AccessPublished:September 12, 2022DOI:https://doi.org/10.1016/j.msard.2022.104181

      Highlights

      • People with MS as well as healthy controls show walking interference in dual-tasking.
      • Only mild MS show cognitive interference in auditory-Stroop during walking.
      • Moderate MS perform worse in dual-tasking compared to mild MS and healthy controls.
      • No cognitive interference (auditory-Stroop) in healthy controls during motor tasks.

      Abstract

      Background

      Reduced motor and cognitive dual-task capacity is found to be more common among people with multiple sclerosis (MS), than among healthy populations. However, studies in larger samples of MS conducted using a more stringent methodology, which includes comparisons to healthy controls, are needed. Thus, the primary aim of this study was to explore the effects on motor and cognitive dual-tasking in people with mild to moderate overall MS-disability, in comparison to healthy controls. A second aim was to explore the differences in dual-task performance on a cognitive task between two motor tasks in people with mild to moderate MS and healthy controls.

      Methods

      This case-control study evaluated dual-task performance of the motor tasks standing with eyes closed (hereafter standing) and walking and a cognitive task assessing selective executive functions (auditory-Stroop test). Fifty-five people with MS (mild MS, n = 28; moderate MS, n = 27), and 30 healthy controls participated. Standing and walking were assessed using wireless inertial measurement unit sensors (APDM). Standing (three 30 s trials) was measured using sway area and root mean square sway, while walking (2 min) was measured using speed, stride length, and step time. Auditory-Stroop was measured using accuracy and response time. During dual-task assessments, each subject was instructed to pay equal attention to both tasks. Statistical significance was considered if p < .05.

      Results

      In standing no significant within-group differences in the standing measures were found between single-task and dual-task performance. However, dual-task performance differed significantly between all groups (moderate MS > mild MS > healthy controls), except between mild and moderate MS in sway area. In walking, all groups slowed down speed and shortened stride length during dual-task condition compared to single-task condition. Moderate MS performed significantly poorer than mild MS and healthy controls in dual-task walking, but mild MS did not differ from healthy controls. In the cognitive task only mild MS increased significantly in auditory-Stroop response time during walking. In healthy controls, the performance of auditory-Stroop was not affected by dual-tasking. Moderate MS had significantly longer response time in dual-task auditory-Stroop compared to the other groups, but no differences were observed between mild MS and healthy controls. Only mild MS had significantly longer response time during walking than during standing.

      Conclusion

      This study showed that cognitive-motor interference in people with MS is present also in the early phases of the disease. This was shown during dual-tasking with slower walking and a longer response time in the cognitive task compared to healthy controls. Moderate MS performed poorer in almost every aspect of the motor and cognitive assessments in dual-task condition, compared to mild MS and healthy controls. Furthermore, during standing, people with MS performed poorer in standing measures compared to healthy controls. Additionally, healthy controls showed no cognitive interference during motor tasks.
      The results suggest that standardized regular assessment of dual-tasking in MS care might increase the individual's knowledge of dual-task capacity and contribute to understanding of possible related consequences. However, feasible assessment equipment and specific motor-cognitive dual-task training interventions for people with MS need to be developed.

      Keywords

      1. Introduction

      Multiple sclerosis (MS), a chronic inflammatory and neurodegenerative disease of the central nervous system (
      • Filippi M.
      • Bar-Or A.
      • Piehl F.
      • Preziosa P.
      • Solari A.
      • Vukusic S.
      • et al.
      Multiple sclerosis.
      ), commonly leads to balance and walking limitations (
      • Comber L.
      • Sosnoff J.J.
      • Galvin R.
      • Coote S.
      Postural control deficits in people with multiple sclerosis: a systematic review and meta-analysis.
      ,
      • Comber L.
      • Galvin R.
      • Coote S.
      Gait deficits in people with multiple sclerosis: a systematic review and meta-analysis.
      ) and cognitive impairment (
      • Chiaravalloti N.D.
      • DeLuca J.
      Cognitive impairment in multiple sclerosis.
      ;
      • Penner I.K.
      Cognition in multiple sclerosis.
      ). To maintain balance control while standing or walking, interactions between multiple underlying physiological systems are required, including systems for movement, sensory function, and cognitive processing (
      • Horak F.B.
      Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls?.
      ). In people with MS (PwMS) it has been shown that performing a simultaneous cognitive task while walking is associated with reduced walking performance (
      • Leone C.
      • Patti F.
      • Feys P.
      Measuring the cost of cognitive-motor dual tasking during walking in multiple sclerosis.
      ), and vice versa (
      • Wajda D.A.
      • Wood T.A.
      • Sosnoff J.J.
      The attentional cost of movement in multiple sclerosis.
      ). Further, there is a higher risk of falling in PwMS, even in the early phases of the disease, when performing a simultaneous cognitive task while walking or standing (dual tasking) (
      • Kalron A.
      • Dvir Z.
      • Achiron A.
      Walking while talking-difficulties incurred during the initial stages of multiple sclerosis disease process.
      ;
      • Etemadi Y.
      Dual task cost of cognition is related to fall risk in patients with multiple sclerosis: a prospective study.
      ). This highlights the need for a better understanding of how disabilities in balance control, ambulation and cognitive function interact, to then develop interventions aimed at reducing the disease burden in PwMS.
      How cognitive-motor interference (CMI) affects PwMS when performing motor and cognitive dual-tasks has recently begun to be explored (
      • Learmonth Y.C.
      • Ensari I.
      • Motl R.W.
      Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
      ;
      • Chamard Witkowski L.
      • Mallet M.
      • Belanger M.
      • Marrero A.
      • Handrigan G.
      Cognitive-postural interference in multiple sclerosis.
      ). Complex cognitive tasks that challenge executive functions, sustained attention, information processing speed, and those that involve conflicting stimuli are recommended for use when testing CMI in PwMS (
      • Prosperini L.
      • Castelli L.
      • Sellitto G.
      • De Luca F.
      • De Giglio L.
      • Gurreri F.
      • et al.
      Investigating the phenomenon of "cognitive-motor interference" in multiple sclerosis by means of dual-task posturography.
      ,
      • Prosperini L.
      • Castelli L.
      • De Luca F.
      • Fabiano F.
      • Ferrante I.
      • De Giglio L.
      Task-dependent deterioration of balance underpinning cognitive-postural interference in MS.
      ). When performing motor and cognitive tasks simultaneously, the effect of dual-tasking could either be negative (a dual-task cost) or positive (a dual-task benefit) for either of the two tasks or for both (
      • Kelly V.E.
      • Janke A.A.
      Effects of instructed focus and task difficulty on concurrent walking and cognitive task performance in healthy young adults.
      ).
      In PwMS, more severe overall MS-disability, involving physical and cognitive impairments, has been shown to be associated with poorer dual-task performance (i.e., CMI) (
      • Rooney S.
      • Ozkul C.
      • Paul L.
      Correlates of dual-task performance in people with multiple sclerosis: a systematic review.
      ). Furthermore, dual-task cost while standing and doing a cognitive task simultaneously is associated with worsened aspects of health-related quality of life (
      • Castelli L.
      • De Luca F.
      • Marchetti M.R.
      • Sellitto G.
      • Fanelli F.
      • Prosperini L.
      The dual task-cost of standing balance affects quality of life in mildly disabled MS people.
      ).
      Although the difference in CMI between people with mild overall MS-disability and healthy controls (HC) was shown to be small in a systematic review (
      • Learmonth Y.C.
      • Ensari I.
      • Motl R.W.
      Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
      ), other studies have shown that CMI is considerably disabling also in PwMS with mild disease (
      • Argento O.
      • Spanò B.
      • Pisani V.
      • Incerti C.C.
      • Bozzali M.
      • Foti C.
      • et al.
      Dual-task performance in multiple sclerosis' patients: cerebellum matters?.
      ;
      • Coghe G.
      • Pilloni G.
      • Zucca E.
      • Porta M.
      • Corona F.
      • Frau J.
      • et al.
      Exploring cognitive motor interference in multiple sclerosis by the visual Stroop test.
      ). In some studies, HC did not show CMI on the cognitive task during walking, while CMI was present among PwMS, suggesting that the performance of a cognitive task during walking could serve as a marker of cognitive status in PwMS (
      • Downer M.B.
      • Kirkland M.C.
      • Wallack E.M.
      • Ploughman M.
      Walking impairs cognitive performance among people with multiple sclerosis but not controls.
      ;
      • Postigo-Alonso B.
      • Galvao-Carmona A.
      • Conde-Gavilán C.
      • Jover A.
      • Molina S.
      • Peña-Toledo M.A.
      • et al.
      The effect of prioritization over cognitive-motor interference in people with relapsing-remitting multiple sclerosis and healthy controls.
      ).
      However, studies on CMI in larger samples of PwMS with a wider range of overall MS-disability, including comparisons with HC, are lacking (
      • Leone C.
      • Patti F.
      • Feys P.
      Measuring the cost of cognitive-motor dual tasking during walking in multiple sclerosis.
      ). Also, studies using a more stringent methodology for the cognitive task in test procedures and in presentation of results are warranted (
      • Leone C.
      • Patti F.
      • Feys P.
      Measuring the cost of cognitive-motor dual tasking during walking in multiple sclerosis.
      ;
      • Chamard Witkowski L.
      • Mallet M.
      • Belanger M.
      • Marrero A.
      • Handrigan G.
      Cognitive-postural interference in multiple sclerosis.
      ), specifically that the design of the dual-task paradigm and how results on the cognitive task should be reported. Therefore, in the present study, the primary aim was to explore the effects on motor and cognitive dual tasking in people with mild to moderate MS, in comparison to HC. A second aim was to explore the difference in dual-task performance on the cognitive task between two motor tasks in people with mild to moderate MS and HC.

      2. Method

      2.1 Study design

      A case-control study with a cross-sectional design.

      2.2 Subjects

      Inclusion criteria were PwMS: diagnosed according to the McDonald criteria (
      • McDonald W.I.
      • Compston A.
      • Edan G.
      • Goodkin D.
      • Hartung H.P.
      • Lublin F.D.
      • et al.
      Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis.
      ;
      • Thompson A.J.
      • Banwell B.L.
      • Barkhof F.
      • Carroll W.M.
      • Coetzee T.
      • Comi G.
      • et al.
      Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria.
      ) with an overall MS-disability score of 2.0 to 5.5 according to the Expanded Disability Status Scale (EDSS) (
      • Kurtzke J.F.
      Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS).
      ); 18 to 65 years of age; able to walk 100 meters without aid. Exclusion criteria were cognitive impairment indicated by a score <21 in the Montreal Cognitive Assessment (MoCA) (
      • Nasreddine Z.S.
      • Phillips N.A.
      • Bedirian V.
      • Charbonneau S.
      • Whitehead V.
      • Collin I.
      • et al.
      The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment.
      ); other conditions that would substantially influence balance; an MS relapse or change of disease-modifying treatment the last eight weeks; alcoholism; or pregnancy. Criterion for inclusion of HC was self-reported good health specifically related to balance and walking performance.
      Eligible subjects with MS were recruited from MS specialist centers and clinical rehabilitation units in Stockholm, Sweden. PwMS who fulfilled the criteria for inclusion and exclusion, were included and categorized into two groups: mild MS (EDSS 2.0 to 3.5) and moderate MS (EDSS 4.0 to 5.5). Sex-and-age-matched HC were purposefully recruited via local press advertisement. The targeted sample size was set to 60 PwMS and 30 HC.
      The Stockholm ethical review board approved the study, Nos. 2018/374-31 and 2019-01562. Procedures were conducted in accordance with the Declaration of Helsinki.

      2.2.1 Data collection

      Data collection was conducted from March 2019 to September 2020 at the movement laboratory at Karolinska Institutet, Stockholm, Sweden. Information on demography, fall-frequency, use of walking aid, education, years since MS diagnosis, and disease course, was collected through a structured interview. Balance control was assessed with the Mini-BESTest (
      • Franchignoni F.
      • Horak F.
      • Godi M.
      • Nardone A.
      • Giordano A.
      Using psychometric techniques to improve the balance evaluation systems test: the mini-BESTest.
      ).

      2.3 Motor and cognitive tasks, and procedures for assessment

      The two studied motor tasks were standing with eyes closed (hereafter standing) and walking. In standing, the subject was instructed to remain standing for 30 s: feet together (wearing shoes), eyes closed, hands on hips, not talking or moving. If the subject failed to remain with the feet together, standing was performed with feet apart. Feet together position was standardized with a 2 cm block placed between the feet, and feet apart with a 20 cm block. The subject performed three 30 s trials per condition, i.e., six trials in total.
      In walking, the subject was instructed to walk for two minutes at a self-selected comfortable pace on a straight 25-meter indoor walkway with 180° turns at each end. For the motor task assessments, the subject was equipped with six wireless inertial measurement unit sensors (Opal sensors, APDM, Portland, OR, USA). Sensors were attached on the trunk, lumbar, wrists and feet, according to APDM Mobility Lab system manual (

      APDM Inc. Mobility Lab. Copyright 2017 ed: APDM, Inc; 2017.

      ). Measures for standing were sway area (degrees2) and root mean square sway (RMS-sway) (degrees). Measures for walking were speed (meter/second), stride length (meter), and step time (second).
      The cognitive task auditory-Stroop (
      • Morgan A.L.
      • Brandt J.F.
      An auditory stroop effect for pitch, loudness, and time.
      ), which challenges selective executive functions, e.g., response inhibition, has been suggested for assessment of CMI in PwMS (
      • Learmonth Y.C.
      • Ensari I.
      • Motl R.W.
      Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
      ). Auditory-Stroop consists of two stimulus words “high” and “low”, verbally presented congruent or incongruent in a high or low pitch level through wireless headphones (RazerTM ManO'War) using Audacity software (version 3.0.2). The high and low pitch frequencies used were approximately 310Hz and 90Hz, respectively. The subject was instructed to verbally respond to the stimulus pitch with the words “high” or “low”, irrespective of the actual word presented, as quickly and accurately as possible. These responses were recorded through wireless headphones using Audacity. An interstimulus interval of 1.5 to 2.0 s was randomly delivered to avoid the stimuli becoming a metronome for walking. Auditory-Stroop accuracy and response time of incongruent stimuli were analyzed.

      2.4 Procedures for assessment

      First, the single-task assessments of walking and thereafter standing were performed. Before the recorded assessments the subject performed one or two practice trials of each motor task. Thereafter auditory-Stroop was presented. Two practice trials were performed in seated position. Then, one or two practice trials with auditory-Stroop dual-task conditions of each motor task were performed. Assessments of the single-task auditory-Stroop and the dual-task standing and walking while performing the auditory-Stroop, were then conducted in a randomized order created by a computerized random sequence generator (http://www.randomization.com). During dual-task assessments the subject was instructed to pay equal attention to both the motor and the cognitive tasks.

      2.4.1 Data processing and statistical analysis

      Summarized motor task data was exported from the Mobility Lab software to Excel (Microsoft), where data processing was performed. Mean values of three standing trials were calculated for each measure (sway area and RMS-sway). Likewise, mean values of left and right legs in walking were calculated for each measure (speed, stride length, and step time). The calculated values were used in the statistical analyses.
      Recorded auditory-Stroop responses were analyzed using MATLAB version R2017b (

      MathWorks Inc. MATLAB version R2017b [computer application]. 2017 [Available from: https://matlab.mathworks.com/.

      ). Response time was measured from the beginning of a stimulus to the beginning of the corresponding response. The auditory-Stroop audio files consisted of both congruent and incongruent stimuli, equally represented but presented in a randomized order. Unique audio files were consistently used in a standardized way in each trial. Response time was calculated as the mean of responses to all incongruent stimuli, irrespective of whether they were correct or incorrect. The first auditory-Stroop stimulus in each trial was excluded from the analysis to eliminate any bias of being surprised at the beginning of a trial. Mean standard deviation (SD) of response time was used as a measure of intraindividual variability. Accuracy was calculated as the percentage of correct responses of the total number of stimuli, where missing responses were counted as incorrect responses.
      For each task, the dual-task effect (DTE) was calculated as described by
      • Kelly V.E.
      • Janke A.A.
      Effects of instructed focus and task difficulty on concurrent walking and cognitive task performance in healthy young adults.
      . For outcomes where a higher value indicated improved performance the following equation was used:
      DTE(%)=(DualtaskSingletask)Singletask*100


      For outcomes where a lower value indicated improved performance, a negative sign was inserted in the equation, thereby the presence of dual-task cost or benefit is indicated by either a negative or a positive DTE-value, respectively.
      DTE(%)=(DualtaskSingletask)Singletask*100


      Descriptive statistics were used to summarize the data. Normality was assessed using the Shapiro-Wilk test, and by visual inspection of figures. Normally distributed data are presented with mean and SD. Non-normally distributed data are presented with median and interquartile range (IQR). Differences in characteristics between the groups were assessed using the independent t-test and a chi-square test.
      For within group analyses between single-task and dual-task performances a repeated measures analysis of variance was used for normally distributed data and the Wilcoxon signed rank test for non-normally distributed data. For between groups analyses in dual-task performance an analysis of variance was used for normally distributed data and the independent samples sign test for non-normally distributed data.
      To explore possible impact of cognitive function on the results, an exploratory analysis was performed after the main analysis was completed, on alternative groups of PwMS created based on the presence of mild cognitive impairment (MoCA score < 26) or normal cognitive function (MoCA score ≥26) (
      • Nasreddine Z.S.
      • Phillips N.A.
      • Bedirian V.
      • Charbonneau S.
      • Whitehead V.
      • Collin I.
      • et al.
      The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment.
      ).
      A p-value <.05 was considered statistically significant. Within group and between groups analyses were adjusted by the Bonferroni correction for multiple tests. IBM® SPSS® Statistics version 28 was used in the statistical analyses.

      3. Results

      3.1 Description of sample

      Of 90 PwMS interested in participation in the study, a total of 55 PwMS were included, 28 with mild and 27 with moderate MS. For comparison, a total of 30 sex-and-age-matched HC were included. Demographic characteristics of the groups are presented in Table 1.
      Table 1Demographic characteristics with descriptions of statistically significant differences (p < .05) between groups.
      CharacteristicsMild MS

      (n = 28)
      Moderate MS

      (n = 27)
      Healthy controls

      (n = 30)
      Sex, no. (%)
      - Woman19 (68)20 (74)22 (73)
      - Men9 (32)7 (26)8 (27)
      Age, years, mean (SD)
      Statistical significant difference between the Mild and the Moderate MS groups;
      ,
      Statistical significant difference between the Moderate MS group and Healthy controls;
      45.5 (9.4)54.3 (8.1)49.1 (10.9)
      Height, m, mean (SD)1.7 (0.1)1.7 (0.1)1.7 (0.1)
      Weight, kg, mean (SD)72.4 (12.7)77.2 (16.3)74.3 (11.7)
      Body mass index, kg/m2, mean (SD)24.6 (3.6)26.2 (5.5)25.1 (3.6)
      Education, years, mean (SD)15.2 (2.2)14.1 (2.4)14.8 (1.7)
      Use of walking aid, no. (%)
      Statistical significant difference between the Mild and the Moderate MS groups;
      ,
      Statistical significant difference between the Moderate MS group and Healthy controls;
      ,
      Statistical significant difference between the Mild MS group and Healthy controls;
      - No22 (79)11 (41)30 (100)
      - Yes, outdoors6 (21)10 (37)0 (0)
      - Yes, in- and outdoors0 (0)6 (22)0 (0)
      Have fallen last six months, no. (%)
      Statistical significant difference between the Moderate MS group and Healthy controls;
      ,
      Statistical significant difference between the Mild MS group and Healthy controls;
      - No19 (68)14 (52)30 (100)
      - Yes9 (32)13 (48)0 (0)
      Mini-BESTest
      The Mini-BESTest score ranges from 0 to 28;
      , mean (SD)
      Statistical significant difference between the Mild and the Moderate MS groups;
      ,
      Statistical significant difference between the Moderate MS group and Healthy controls;
      ,
      Statistical significant difference between the Mild MS group and Healthy controls;
      22.5 (3.1)17.7 (3.7)25.9 (1.3)
      Cognitive function, MoCA
      MoCA = Montreal Cognitive Assessment, the MoCA score ranges from 0 to 30;
      , mean (SD)
      Statistical significant difference between the Moderate MS group and Healthy controls;
      ,
      Statistical significant difference between the Mild MS group and Healthy controls;
      26.9 (1.8)26.3 (2.0)28.1 (1.4)
      Mild Cognitive Impairment according to MoCA
      MoCA = Montreal Cognitive Assessment, the MoCA score ranges from 0 to 30;
      (<26), no. (%)
      Statistical significant difference between the Moderate MS group and Healthy controls;
      ,
      Statistical significant difference between the Mild MS group and Healthy controls;
      - No22 (79)18 (67)30 (100)
      - Yes6 (21)9 (33)0 (0)
      Overall MS disability, EDSS
      The EDSS score ranges from 0 to 10.
      , mean (SD)
      Statistical significant difference between the Mild and the Moderate MS groups;
      3.0 (0.5)4.5 (0.5)-
      Course, no (%)
      Statistical significant difference between the Mild and the Moderate MS groups;
      - Relapsing remitting26 (93)16 (59)-
      - Progressive2 (7)11 (41)-
      Years since diagnosis, mean (SD)11.5 (8.1)12.9 (9.2)-
      a Statistical significant difference between the Mild and the Moderate MS groups;
      b Statistical significant difference between the Moderate MS group and Healthy controls;
      c Statistical significant difference between the Mild MS group and Healthy controls;
      d The Mini-BESTest score ranges from 0 to 28;
      e MoCA = Montreal Cognitive Assessment, the MoCA score ranges from 0 to 30;
      f The EDSS score ranges from 0 to 10.

      3.2 Performance during single-task and dual-task conditions

      During standing within each group (mild MS, moderate MS, and HC), no significant differences between the performance in single-task and dual-task conditions were shown in sway area and RMS-sway (Table 2). However, all groups tended to decrease in sway area and RMS-sway during the dual-task condition, compared with the single-task condition.
      Table 2Performance in single-task and dual-task conditions, differences, and dual-task effect on motor (standing with eyes closed and walking) and cognitive (auditory-Stroop) tasks, in people with mild (n = 28) and moderate (n = 27) multiple sclerosis and in healthy controls (n = 30). Negative dual-task effect values indicate a dual-task cost, and positive dual-task effect values indicate a dual-task benefit.
      TaskMeasureGroupSingle-task conditionDual-task conditionP value
      Bonferroni adjustment for multiple comparisons. RMS = root mean square.
      Dual-task effect %
      StandingStanding+CognitiveMotor measures(Standing)
      Motor

      (Standing with eyes closed)
      Sway area (°2)
      Values presented in median and interquartile range.
      Mild MS4.80 (6.59)3.92 (5.72).60018.3 (-13.2)
      Moderate MS9.24 (9.52)6.46 (6.68)1.0030.1 (-29.8)
      Healthy controls2.13 (1.97)1.85 (1.49).75613.0 (-24.2)
      RMS Sway
      Values presented in median and interquartile range.
      Mild MS.73 (.46).67 (.43).7808.6 (-7.5)
      Moderate MS1.12 (.82).99 (.73)1.0011.8 (-10.8)
      Healthy controls.51 (.24).51 (.20).768.1 (-17.1)
      WalkingWalking+CognitiveMotor measures(Walking)
      Motor

      (Walking)
      Speed (m/s)
      Values presented in mean and standard deviation.
      Mild MS1.35 (.17)1.32 (.18).033-2.3 (7.0)
      Moderate MS1.19 (.21)1.13 (.22)<.001-5.3 (4.7)
      Healthy controls1.48 (.15)1.44 (.17).006-2.6 (12.3)
      Stride length (m)
      Values presented in mean and standard deviation.
      Mild MS1.37 (.13)1.35 (.14).002-1.9 (6.5)
      Moderate MS1.26 (.15)1.21 (.16)<.001-3.8 (7.3)
      Healthy controls1.47 (.11)1.43 (.11)<.001-2.3 (5.2)
      Step time (s)
      Values presented in mean and standard deviation.
      Mild MS.51 (.03).51 (.04).426-.5 (5.8)
      Moderate MS.53 (.05).54 (.06).041-2.0 (6.5)
      Healthy controls.50 (.04).50 (.04).326-.6 (8.7)
      CognitiveA.Cognitive+StandingCognitive measures
      B.Cognitive+Walking
      Cognitive

      (auditory-Stroop
      Incongruent stimulus in auditory-Stroop.
      )
      Accuracy (%)
      Values presented in median and interquartile range.
      Mild MS100 (.0)A.100 (.0)1.00.0 (.0)
      B.100 (.0)1.00.0 (.0)
      Moderate MS100 (5.0)A.100 (5.9).552.0 (17.6)
      B.100 (11.5)1.00.0 (130.8)
      Healthy controls100 (.0)A.100 (.0)1.00.0 (.0)
      B.100 (.0)1.00.0 (.0)
      Response time (s)
      Values presented in mean and standard deviation.
      Mild MS.86 (.17)A..86 (.16).978-.1 (-4.9)
      B..92 (.17).003-7.5 (2.5)
      Moderate MS1.05 (.20)A.1.00 (.17).0804.6 (-14.9)
      B.1.08 (.33).454-3.1 70.3)
      Healthy controls.85 (.13)A..82 (.14).1632.7 (6.2)
      B..85 (.14).967.1 (6.3)
      a Incongruent stimulus in auditory-Stroop.
      b Values presented in median and interquartile range.
      c Values presented in mean and standard deviation.
      d Bonferroni adjustment for multiple comparisons. RMS = root mean square.
      During walking, there were significant differences between single-task and dual-task performance in speed and stride length across all groups (Table 2). For moderate MS also step time was significantly prolonged. All groups showed a significant dual-task cost in speed and stride length; the effect size for both measures were about twice as large in moderate MS as compared to mild MS and HC.
      The median auditory-Stroop accuracy was 100% in both single-task and dual-task conditions of both motor tasks across the three groups (Table 2). However, although median accuracy in single-task was also 100% in moderate MS, the variance showed that some had difficulties with the task. The mean auditory-Stroop response time during single-task conditions in mild MS and HC were similar, while it was higher in moderate MS. In mild MS, response time during standing remained unchanged, while moderate MS and HC tended to have a dual-task benefit. Mild MS showed a significant difference in response time between single-task and dual-task walking, the dual-task cost was 7.5%.
      The dual-task performance of walking speed and auditory-Stroop response time showed an increased variance with increased disability (see Table 2). In Fig. 1, the mean differences, and the scattered, linear distribution in the dual-task performance of these measures are shown for the three groups.
      Fig 1
      Fig. 1Scatter plot on performance of dual-task walking speed and auditory-Stroop response time in people with mild (n = 28, gray circles) and moderate (n = 27, white circles) multiple sclerosis and in healthy controls (n = 30, black circles). Mean values on speed and response time are marked for mild MS with dotted lines, for moderate MS with dashed lines, and for healthy controls with dash-dotted lines.

      3.3 Comparisons of dual-task performance between groups

      During dual-task standing, mild and moderate MS had significantly larger sway area compared to HC, but there was no significant sway area difference between mild and moderate MS (Table 3). Regarding dual-task RMS-sway there were significant differences between all groups, (i.e., moderate MS > mild MS > HC).
      Table 3Between group comparison of dual-task performance and dual-task effect, in dual-task conditions of motor (standing with eyes closed and walking) and cognitive (auditory-Stroop) tasks, in people with mild (n = 28) and moderate (n = 27) multiple sclerosis and in healthy controls (n = 30).
      TaskMeasureDual-taskGroup comparisonDual-task performanceDual-task effect %
      Mean-/Median differenceP value
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Mean-/Median differenceP value
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Motor

      (Standing with eyes closed)
      Sway area (°2)
      Median values used.
      auditory-StroopModerate MS–Mild MS2.54.130f-
      Moderate MS–Healthy controls4.61.000f-
      Mild MS–Healthy controls2.07.001f-
      RMS sway
      Median values used.
      auditory-StroopModerate MS–Mild MS.26.031f-
      Moderate MS–Healthy controls.48.000f-
      Mild MS–Healthy controls.16.001f-
      Motor(Walking)Speed (m/s)
      Mean values used.
      auditory-StroopModerate MS–Mild MS-.19.001-3.2.126
      Moderate MS–Healthy controls-.31<.001-2.8.217
      Mild MS–Healthy controls-.12.058.41.00
      Stride length (m)
      Mean values used.
      auditory-StroopModerate MS–Mild MS-.14.001-2.0.047
      Moderate MS–Healthy controls-.23<.001-1.6.163
      Mild MS–Healthy controls-.09.065.51.00
      Step time (s)
      Mean values used.
      auditory-StroopModerate MS–Mild MS.03.0351.5.412
      Moderate MS–Healthy controls.04.0011.5.439
      Mild MS–Healthy controls.01.899-.11.00
      Cognitive (auditory- Stroop
      Incongruent stimulus in auditory-Stroop.
      )
      Accuracy (%)
      Median values used.
      StandingModerate MS–Mild MS
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      Moderate MS–Healthy controls
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      Mild MS–Healthy controls
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      WalkingModerate MS–Mild MS
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      Moderate MS–Healthy controls
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      Mild MS–Healthy controls
      Unable to compute: All test field values were less than or equal to the median.
      -
      Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      -
      Response time (s)
      Mean values used.
      StandingModerate MS–Mild MS.15.002-4.2.419
      Moderate MS–Healthy controls.18<.001-1.01.00
      Mild MS–Healthy controls.031.003.2.729
      WalkingModerate MS–Mild MS.16.031-6.3.301
      Moderate MS–Healthy controls.24<.0012.01.00
      Mild MS–Healthy controls.07.6508.3.080
      a Incongruent stimulus in auditory-Stroop.
      b Median values used.
      c Mean values used.
      d Unable to compute: All test field values were less than or equal to the median.
      e Significance values were adjusted by the Bonferroni correction for multiple tests.
      f Multiple comparisons were not performed because the overall test did not show significant differences across samples. RMS = root mean square.
      During dual-task walking, moderate MS walked significantly slower, with shorter stride length and longer step time compared to mild MS and HC, but there were no significant differences in walking between mild MS and HC (Table 3). Moderate MS had a larger dual-task cost in stride length than mild MS.
      The between group comparisons of auditory-Stroop accuracy could not be computed since all values across the groups were less than, or equal to, the median (Table 3). However, there was significantly longer response time in moderate MS compared to the other two groups during standing and during walking.

      3.4 Comparison of performance of the cognitive task during standing and walking

      Mild MS increased by 7.5% in auditory-Stroop response time from dual-task standing to dual-task walking (Table 4). Moderate MS and HC also increased in response time from standing to walking, albeit not significantly. In Fig. 2, the differences, and the linear distribution in dual-task response time during standing and during walking for the three groups, implying increased response time and variance with increased disability.
      Table 4Descriptive results of performance in dual-task condition, differences, and change in auditory-Stroop response time during standing with eyes closed and during walking in people with mild (n = 28) and moderate (n = 27) multiple sclerosis and in healthy controls (n = 30).
      TaskMeasureGroupPerformance in dual-task conditionsP valuePercentage change between A and B
      A. Cognitive+StandingB. Cognitive+Walking
      Cognitive (auditory-Stroop
      Incongruent stimulus in auditory-Stroop.
      )
      Response time (s)
      Values presented in mean and standard deviation.
      Mild MS.86 (.16).92 (.17)<.0017.5 (7.8)
      Moderate MS1.00 (.17)1.08 (.33).1728.0 (100.1)
      Healthy controls.82 (.14).85 (.14).2222.7 (.1)
      a Incongruent stimulus in auditory-Stroop.
      b Values presented in mean and standard deviation.
      Fig 2
      Fig. 2Scatter plot on dual-task performance of auditory-Stroop response time during standing with eyes closed and during walking in people with mild (n = 28, gray circles) and moderate (n = 27, white circles) multiple sclerosis and in healthy controls (n = 30, black circles). Mean values on response times are marked for mild MS with dotted lines, for moderate MS with dashed lines, and for healthy controls with dash-dotted lines.

      3.5 Exploratory analysis of the influence of cognitive function on dual-task performance in PwMS

      The group with mild cognitive impairment performed significantly worse in auditory-Stroop response time both during standing and during walking compared with the group with normal cognitive function (Table 5). There were no other significant differences in dual-task performance between the groups. However, the group with mild cognitive impairment had a larger dual-task cost in walking step time.
      Table 5Exploratory analysis of the influence of cognitive function on dual-task performance. Between group comparison of dual-task performance and dual-task effect, in dual-task conditions of motor (standing with eyes closed and walking) and cognitive (auditory Stroop) tasks, in people with MS with mild cognitive impairment
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 15) and people with MS with normal cognitive function
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 40). Negative dual-task effect values indicate a dual-task cost, and positive dual-task effect values indicate a dual-task benefit.
      TaskMeasureDual-taskDual-task performanceDual-task effect %
      Mild cognitive impairment
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 15)
      Normal cognitive function
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 40)
      Mild cognitive impairment
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 15)
      Normal cognitive function
      According to the Montreal Cognitive Assessment (MoCA) (Nasreddine et al., 2005).
      (n = 40)
      Mean-/MedianMean-/MedianP valueMean-/MedianMean-/MedianP value
      Motor

      (Standing with eyes closed)
      Sway area (°2)
      Median values used.
      auditory-Stroop5.246.02.601
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      16.213.4.491
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      RMS sway
      Median values used.
      auditory-Stroop.75.85.934
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      10.95.9.491
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      Motor

      (Walking)
      Speed (m/s)
      Mean values used.
      auditory-Stroop1.171.25.242
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      -6.3-2.8.064
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Stride length (m)
      Mean values used.
      auditory-Stroop1.261.29.589
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      -3.7-2.6.268
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Step time (s)
      Mean values used.
      auditory-Stroop.54.52.166
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      -3.1-.6.041
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Cognitive

      (auditory-Stroop
      Incongruent stimulus in auditory Stroop.
      )
      Accuracy (%)
      Median values used.
      Standing100100
      Unable to compute: All test field values were less than or equal to the median.
      .0.0.258
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      Walking100100
      Unable to compute: All test field values were less than or equal to the median.
      .0.0.256
      Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.
      Response time (s)
      Mean values used.
      Standing1.01.90.029
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      .41.7.687
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      Walking1.14.95.016
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      -8.8-4.3.345
      Significance values were adjusted by the Bonferroni correction for multiple tests.
      a According to the Montreal Cognitive Assessment (MoCA) (
      • Nasreddine Z.S.
      • Phillips N.A.
      • Bedirian V.
      • Charbonneau S.
      • Whitehead V.
      • Collin I.
      • et al.
      The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment.
      ).
      b Incongruent stimulus in auditory Stroop.
      c Median values used.
      d Mean values used.
      e Unable to compute: All test field values were less than or equal to the median.
      f Significance values were adjusted by the Bonferroni correction for multiple tests.
      g Yates's Continuity Corrected Asymptotic Sig. RMS = root mean square.

      4. Discussion

      This study explored dual-task effects on two motor tasks (standing and walking) and a cognitive task (auditory-Stroop) in groups of people with mild and moderate MS, and HC. In motor measures, walking deteriorated during dual-task in all groups when performing the auditory-Stroop simultaneously, but no dual-task effects were found in standing. Furthermore, only mild MS showed dual-task cost in response time during walking. During standing no cognitive or motor dual-task effects were found, although the dual-task standing performance deteriorated with the level of overall MS-disability (moderate MS > mild MS > HC), apart from a non-significant difference between mild and moderate MS in sway area. In all dual-task conditions, moderate MS performed worse compared to mild MS and HC. Mild MS also showed worsened performance in dual tasking compared with HC. The results imply that an increased level of overall MS-disability is associated with a greater deterioration of motor-cognitive dual-task capacity (
      • Rooney S.
      • Ozkul C.
      • Paul L.
      Correlates of dual-task performance in people with multiple sclerosis: a systematic review.
      ).
      Most previous studies addressing CMI in PwMS during standing have been conducted with eyes open and have consistently demonstrated a dual-task cost on postural sway (
      • Etemadi Y.
      Dual task cost of cognition is related to fall risk in patients with multiple sclerosis: a prospective study.
      ;
      • Prosperini L.
      • Castelli L.
      • De Luca F.
      • Fabiano F.
      • Ferrante I.
      • De Giglio L.
      Task-dependent deterioration of balance underpinning cognitive-postural interference in MS.
      ;
      • Castelli L.
      • De Luca F.
      • Marchetti M.R.
      • Sellitto G.
      • Fanelli F.
      • Prosperini L.
      The dual task-cost of standing balance affects quality of life in mildly disabled MS people.
      ;
      • Wajda D.A.
      • Motl R.W.
      • Sosnoff J.J.
      Correlates of dual task cost of standing balance in individuals with multiple sclerosis.
      ). The advantage of assessing standing in eyes-closed condition in PwMS, as in our study, is that the frequently occurring impact of impaired somatosensory and vestibular functions on balance control (
      • Gelfand J.M.
      Multiple sclerosis: diagnosis, differential diagnosis, and clinical presentation.
      ) becomes central since the options for compensation by visual input are eliminated. Interestingly, we found a trend of a dual-task benefit in sway area and RMS-sway across all groups. This has also been shown in two other studies using eyes-closed condition (
      • Negahban H.
      • Mofateh R.
      • Arastoo A.A.
      • Mazaheri M.
      • Yazdi M.J.
      • Salavati M.
      • et al.
      The effects of cognitive loading on balance control in patients with multiple sclerosis.
      ,
      • Negahban H.
      • Monjezi S.
      • Mehravar M.
      • Mostafaee N.
      • Shoeibi A.
      Responsiveness of postural performance measures following balance rehabilitation in multiple sclerosis patients.
      ). Another study has suggested that mild stress induced by dual tasking might enhance attention (
      • Shields G.S.
      • Rivers A.M.
      • Ramey M.M.
      • Trainor B.C.
      • Yonelinas A.P.
      Mild acute stress improves response speed without impairing accuracy or interference control in two selective attention tasks: Implications for theories of stress and cognition.
      ). Transferred to our study, stress could be a likely explanation of the results, with a dual-task benefit of the balance performance during standing with eyes closed. No benefit was shown in eyes-open condition which might be because when standing with eyes open, it is less of a challenge to maintain balance. Thus, increased attention might not be needed when visual input is allowed. Hence, the difference in performance in standing with eyes closed between PwMS and HC regardless of variation in overall MS-disability might be explained by the impairment in somatosensory and vestibular functions commonly occurring in PwMS (
      • Gelfand J.M.
      Multiple sclerosis: diagnosis, differential diagnosis, and clinical presentation.
      ).
      All groups showed dual-task costs on walking speed and stride length in line with a meta-analysis of CMI in people with mild MS and HC (
      • Learmonth Y.C.
      • Ensari I.
      • Motl R.W.
      Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
      ). This indicates a need for feasible methods to assess in the clinic the dual-task capacity in mild MS as well. Moderate MS differed from the other groups in dual-task walking performance, while mild MS and HC performed similarly. The differences shown may be linked to the group categorization based on the EDSS, where increased limitations in walking correspond to a higher EDSS score (
      • Kurtzke J.F.
      Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS).
      ). However, the EDSS is frequently used in MS research, which facilitates the transferability of results.
      Most studies conducted in PwMS have investigated CMI during walking intermittently on a ten-meter distance (
      • Etemadi Y.
      Dual task cost of cognition is related to fall risk in patients with multiple sclerosis: a prospective study.
      ;
      • Postigo-Alonso B.
      • Galvao-Carmona A.
      • Conde-Gavilán C.
      • Jover A.
      • Molina S.
      • Peña-Toledo M.A.
      • et al.
      The effect of prioritization over cognitive-motor interference in people with relapsing-remitting multiple sclerosis and healthy controls.
      ;
      • Veldkamp R.
      • Kalron A.
      • Baert I.
      • Hämäläinen P.
      • Tacchino A.
      • D'Hooge M.
      • et al.
      Differential effects and discriminative validity of motor and cognitive tasks varying in difficulty on cognitive-motor interference in persons with multiple sclerosis.
      ;
      • Leone C.
      • Moumdjian L.
      • Patti F.
      • Vanzeir E.
      • Baert I.
      • Veldkamp R.
      • et al.
      Comparing 16 different dual-tasking paradigms in individuals with multiple sclerosis and healthy controls: working memory tasks indicate cognitive-motor interference.
      ). Our study investigated CMI during a 2 min continuous walking trial corresponding to approximately 150 meters, which to our knowledge has been done only once previously (
      • Argento O.
      • Spanò B.
      • Pisani V.
      • Incerti C.C.
      • Bozzali M.
      • Foti C.
      • et al.
      Dual-task performance in multiple sclerosis' patients: cerebellum matters?.
      ). The advantage of using longer continuous walking trials is that the CMI phenomenon is studied in a setting more resembling real-life. The mobile sensor-based system (

      APDM Inc. Mobility Lab. Copyright 2017 ed: APDM, Inc; 2017.

      ) used in this study is one of several newer systems available that enable this type of assessment.
      The response time during walking increased in mild MS, implying that walking requires greater processing resources, thus limiting cognitive processing (
      • Bayot M.
      • Dujardin K.
      • Tard C.
      • Defebvre L.
      • Bonnet C.T.
      • Allart E.
      • et al.
      The interaction between cognition and motor control: a theoretical framework for dual-task interference effects on posture, gait initiation, gait and turning.
      ) in mild MS, as well. For moderate MS, during standing, accuracy but not response time was significant. However, there was no dual-task effect, which might result from the non-normal distribution of data. In single-task auditory-Stroop, mild MS performed similarly to HC in both accuracy and response time and was faster than moderate MS. The results in our study indicate that the moderate MS population is associated with greater variation in CMI compared with the mild MS and the healthy populations, as illustrated by the increased variance as disability increases in Fig. 1. However, these difficulties are also present in mild MS (
      • Learmonth Y.C.
      • Ensari I.
      • Motl R.W.
      Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
      ;
      • Downer M.B.
      • Kirkland M.C.
      • Wallack E.M.
      • Ploughman M.
      Walking impairs cognitive performance among people with multiple sclerosis but not controls.
      ), although this is often a hidden impairment that needs to be managed consciously.
      For exploration of CMI, we used a cognitive task that challenges executive functions and involves conflicting stimuli, as suggested for increased methodological stringency when investigating CMI in PwMS (
      • Prosperini L.
      • Castelli L.
      • De Luca F.
      • Fabiano F.
      • Ferrante I.
      • De Giglio L.
      Task-dependent deterioration of balance underpinning cognitive-postural interference in MS.
      ). Furthermore, the conflicting stimuli were delivered with a larger difference in frequency between high and low pitches than previously described (
      • Leone C.
      • Moumdjian L.
      • Patti F.
      • Vanzeir E.
      • Baert I.
      • Veldkamp R.
      • et al.
      Comparing 16 different dual-tasking paradigms in individuals with multiple sclerosis and healthy controls: working memory tasks indicate cognitive-motor interference.
      ). Thereby, the risk of not perceiving the difference in frequency between pitches was reduced.
      Mild MS had longer auditory-Stroop response time during walking than during standing. A similar trend was seen in moderate MS. Further, the effect differences on response time between the dual-task standing and dual-task walking among PwMS, were three times larger than among HC. This indicates that the cognitive task is negatively affected when performing the dual-task in walking in comparison to standing within PwMS but not within the healthy population. This might result from automaticity, i.e., that motor tasks can be performed automatically within a healthy population (
      • Clark D.J.
      Automaticity of walking: functional significance, mechanisms, measurement and rehabilitation strategies.
      ), but that they require greater processing resources in PwMS (
      • Wajda D.A.
      • Wood T.A.
      • Sosnoff J.J.
      The attentional cost of movement in multiple sclerosis.
      ;
      • Rooney S.
      • Ozkul C.
      • Paul L.
      Correlates of dual-task performance in people with multiple sclerosis: a systematic review.
      ).
      The exploratory analysis of groups of PwMS based on cognitive function according to MoCA (
      • Nasreddine Z.S.
      • Phillips N.A.
      • Bedirian V.
      • Charbonneau S.
      • Whitehead V.
      • Collin I.
      • et al.
      The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment.
      ), paralleled the study's main analysis. Interestingly, no significant differences in dual-tasking were shown between the groups regarding the motor tasks, which may be an argument for using EDSS to categorize groups when exploring CMI, in order to also consider the aspect of impaired motor function. However, it should be noted that the sample was unevenly distributed between the groups and that the group with mild cognitive impairment was non-normally distributed. To control for impact of cognitive function a higher precision in its’ classification should be accomplished by usage of a more extensive cognitive assessment battery.
      Our findings substantiate the value of assessing dual-task performance as an estimate of disability in PwMS, in consistence with a recent systematic review (
      • Rooney S.
      • Ozkul C.
      • Paul L.
      Correlates of dual-task performance in people with multiple sclerosis: a systematic review.
      ). For the individual, dual-task assessment might increase awareness of dual-task capacity, understanding of one's limitations, and possible related consequences. Higher dual-task cost during standing has been associated with worse health-related quality of life (
      • Castelli L.
      • De Luca F.
      • Marchetti M.R.
      • Sellitto G.
      • Fanelli F.
      • Prosperini L.
      The dual task-cost of standing balance affects quality of life in mildly disabled MS people.
      ), which implies that balance training that include dual-tasking can positively impact both functioning and health-related quality of life.
      Strengths of this study contributing to justification of the results include the larger sample of PwMS and HC compared to other studies in the field, the similar demographics and sizes across the three groups, a sample distribution reflecting the prevalence of the underlying condition, advanced equipment enabling methodologically stringent test procedures, and that the results were presented for all tasks.
      Limitations in the study were that cognitive function in the sample was briefly assessed that the difference in foot placement between subjects in the standing task (i.e., performed either with feet placed together or apart) was not controlled for in the analysis, and that the cognitive task could have been more challenging. The cognitive task auditory-Stroop could be further developed by the inclusion of additional types of conflicting stimuli requiring cognitive function (e.g., combining the stimuli with different names or numbers or adding memory challenging components), to increase the usefulness of accuracy as an outcome measure. The motor task walking could also be further developed by assessment at maximum speed. A further limitation was that the data collection had to be closed before completion due to the COVID pandemic.
      Future CMI studies conducted with longer walking trials and during standing with eyes closed are needed. Furthermore, studies exploring brain activity during dual tasking are warranted to better understand the mechanisms of CMI. This could be performed with functional near-infrared spectroscopy, which enables the measurement of real-time brain activity during different motor tasks (
      • Gramigna V.
      • Pellegrino G.
      • Cerasa A.
      • Cutini S.
      • Vasta R.
      • Olivadese G.
      • et al.
      Near-infrared spectroscopy in gait disorders: is it time to begin?.
      ).

      5. Conclusions

      This study showed that CMI in PwMS is present also in the early phases of the disease, as shown during dual tasking with slower walking and the need of longer response time in the cognitive task compared to HC. Moderate MS performed worse in almost every aspect of motor and cognitive assessments in dual-task condition, compared to mild MS and HC. Furthermore, during standing, people with MS performed worse in standing measures compared to HC. Additionally, HC showed no cognitive interference during motor tasks.
      The results implicate that assessment of dual-tasking in the clinic might increase the individual's knowledge of dual-task capacity and contribute to an understanding of possible related consequences. To accomplish this, feasible equipment for the assessment in the clinic needs to be developed. Further, development of specific training interventions for improved motor and cognitive dual-task ability among PwMS are needed.

      Funding

      This trial was supported by grants from: the Doctoral School in Health Care Sciences, Karolinska Institutet; the NEURO Sweden (F2019-0056 and F2021-0071); the Promobilia Foundation ( A22085 ); the Center for Innovative Medicine , Region Stockholm (CIMED, SLL20190446 ); and the Region Stockholm (ALF, SLL20180182).

      CRediT authorship contribution statement

      Andreas Wallin: Conceptualization, Supervision, Visualization, Project administration, Data curation, Formal analysis, Writing – review & editing. Erika Franzén: Conceptualization, Supervision, Visualization, Writing – review & editing. Lucian Bezuidenhout: Conceptualization, Supervision, Visualization, Formal analysis, Writing – review & editing. Urban Ekman: Conceptualization, Supervision, Visualization, Writing – review & editing. Fredrik Piehl: Conceptualization, Supervision, Visualization, Writing – review & editing. Sverker Johansson: Conceptualization, Supervision, Visualization, Project administration, Writing – review & editing.

      Declaration of Competing Interest

      None.

      Acknowledgments

      The authors wish to express their deepest gratitude to all who participated in the study. The authors also wish to thank uMOVE, the core facility for human movement science at Karolinska Institutet.

      References

      1. APDM Inc. Mobility Lab. Copyright 2017 ed: APDM, Inc; 2017.

        • Argento O.
        • Spanò B.
        • Pisani V.
        • Incerti C.C.
        • Bozzali M.
        • Foti C.
        • et al.
        Dual-task performance in multiple sclerosis' patients: cerebellum matters?.
        Arch. Clin. Neuropsychol. 2021; 36: 517-526
        • Bayot M.
        • Dujardin K.
        • Tard C.
        • Defebvre L.
        • Bonnet C.T.
        • Allart E.
        • et al.
        The interaction between cognition and motor control: a theoretical framework for dual-task interference effects on posture, gait initiation, gait and turning.
        Neurophysiol. Clin. 2018; 48: 361-375
        • Castelli L.
        • De Luca F.
        • Marchetti M.R.
        • Sellitto G.
        • Fanelli F.
        • Prosperini L.
        The dual task-cost of standing balance affects quality of life in mildly disabled MS people.
        Neurol. Sci. 2016; 37: 673-679
        • Chamard Witkowski L.
        • Mallet M.
        • Belanger M.
        • Marrero A.
        • Handrigan G.
        Cognitive-postural interference in multiple sclerosis.
        Front. Neurol. 2019; 10: 913
        • Chiaravalloti N.D.
        • DeLuca J.
        Cognitive impairment in multiple sclerosis.
        Lancet Neurol. 2008; 7: 1139-1151
        • Clark D.J.
        Automaticity of walking: functional significance, mechanisms, measurement and rehabilitation strategies.
        Front. Hum. Neurosci. 2015; 9: 246
        • Coghe G.
        • Pilloni G.
        • Zucca E.
        • Porta M.
        • Corona F.
        • Frau J.
        • et al.
        Exploring cognitive motor interference in multiple sclerosis by the visual Stroop test.
        Mult. Scler. Relat. Disord. 2018; 22: 8-11
        • Comber L.
        • Galvin R.
        • Coote S.
        Gait deficits in people with multiple sclerosis: a systematic review and meta-analysis.
        Gait Posture. 2017; 51: 25-35
        • Comber L.
        • Sosnoff J.J.
        • Galvin R.
        • Coote S.
        Postural control deficits in people with multiple sclerosis: a systematic review and meta-analysis.
        Gait Posture. 2018; 61: 445-452
        • Downer M.B.
        • Kirkland M.C.
        • Wallack E.M.
        • Ploughman M.
        Walking impairs cognitive performance among people with multiple sclerosis but not controls.
        Hum. Mov. Sci. 2016; 49: 124-131
        • Etemadi Y.
        Dual task cost of cognition is related to fall risk in patients with multiple sclerosis: a prospective study.
        Clin. Rehabil. 2017; 31: 278-284
        • Filippi M.
        • Bar-Or A.
        • Piehl F.
        • Preziosa P.
        • Solari A.
        • Vukusic S.
        • et al.
        Multiple sclerosis.
        Nat. Rev. Dis. Primers. 2018; 4: 43
        • Franchignoni F.
        • Horak F.
        • Godi M.
        • Nardone A.
        • Giordano A.
        Using psychometric techniques to improve the balance evaluation systems test: the mini-BESTest.
        J. Rehabil. Med. 2010; 42: 323-331
        • Gelfand J.M.
        Multiple sclerosis: diagnosis, differential diagnosis, and clinical presentation.
        Handb. Clin. Neurol. 2014; 122: 269-290
        • Gramigna V.
        • Pellegrino G.
        • Cerasa A.
        • Cutini S.
        • Vasta R.
        • Olivadese G.
        • et al.
        Near-infrared spectroscopy in gait disorders: is it time to begin?.
        Neurorehabil. Neural Repair. 2017; 31: 402-412
        • Horak F.B.
        Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls?.
        Age Ageing. 2006; 35: ii7-ii11
        • Kalron A.
        • Dvir Z.
        • Achiron A.
        Walking while talking-difficulties incurred during the initial stages of multiple sclerosis disease process.
        Gait Posture. 2010; 32: 332-335
        • Kelly V.E.
        • Janke A.A.
        Effects of instructed focus and task difficulty on concurrent walking and cognitive task performance in healthy young adults.
        Exp. Brain Res. 2010; 207: 65-73
        • Kurtzke J.F.
        Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS).
        Neurology. 1983; 33: 1444-1452
        • Learmonth Y.C.
        • Ensari I.
        • Motl R.W.
        Cognitive motor interference in multiple sclerosis: insights from a systematic quantitative review.
        Arch. Phys. Med. Rehabil. 2017; 98: 1229-1240
        • Leone C.
        • Patti F.
        • Feys P.
        Measuring the cost of cognitive-motor dual tasking during walking in multiple sclerosis.
        Mult. Scler. 2015; 21: 123-131
        • Leone C.
        • Moumdjian L.
        • Patti F.
        • Vanzeir E.
        • Baert I.
        • Veldkamp R.
        • et al.
        Comparing 16 different dual-tasking paradigms in individuals with multiple sclerosis and healthy controls: working memory tasks indicate cognitive-motor interference.
        Front. Neurol. 2020; 11: 918
      2. MathWorks Inc. MATLAB version R2017b [computer application]. 2017 [Available from: https://matlab.mathworks.com/.

        • McDonald W.I.
        • Compston A.
        • Edan G.
        • Goodkin D.
        • Hartung H.P.
        • Lublin F.D.
        • et al.
        Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis.
        Ann. Neurol. 2001; 50: 121-127
        • Morgan A.L.
        • Brandt J.F.
        An auditory stroop effect for pitch, loudness, and time.
        Brain Lang. 1989; 36: 592-603
        • Nasreddine Z.S.
        • Phillips N.A.
        • Bedirian V.
        • Charbonneau S.
        • Whitehead V.
        • Collin I.
        • et al.
        The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment.
        J. Am. Geriatr. Soc. 2005; 53: 695-699
        • Negahban H.
        • Mofateh R.
        • Arastoo A.A.
        • Mazaheri M.
        • Yazdi M.J.
        • Salavati M.
        • et al.
        The effects of cognitive loading on balance control in patients with multiple sclerosis.
        Gait Posture. 2011; 34: 479-484
        • Negahban H.
        • Monjezi S.
        • Mehravar M.
        • Mostafaee N.
        • Shoeibi A.
        Responsiveness of postural performance measures following balance rehabilitation in multiple sclerosis patients.
        J. Bodyw. Mov. Ther. 2018; 22: 502-510
        • Penner I.K.
        Cognition in multiple sclerosis.
        Neurodegener. Dis. Manag. 2017; 7: 19-21
        • Postigo-Alonso B.
        • Galvao-Carmona A.
        • Conde-Gavilán C.
        • Jover A.
        • Molina S.
        • Peña-Toledo M.A.
        • et al.
        The effect of prioritization over cognitive-motor interference in people with relapsing-remitting multiple sclerosis and healthy controls.
        PLoS One. 2019; 14e0226775
        • Prosperini L.
        • Castelli L.
        • Sellitto G.
        • De Luca F.
        • De Giglio L.
        • Gurreri F.
        • et al.
        Investigating the phenomenon of "cognitive-motor interference" in multiple sclerosis by means of dual-task posturography.
        Gait Posture. 2015; 41: 780-785
        • Prosperini L.
        • Castelli L.
        • De Luca F.
        • Fabiano F.
        • Ferrante I.
        • De Giglio L.
        Task-dependent deterioration of balance underpinning cognitive-postural interference in MS.
        Neurology. 2016; 87: 1085-1092
        • Rooney S.
        • Ozkul C.
        • Paul L.
        Correlates of dual-task performance in people with multiple sclerosis: a systematic review.
        Gait Posture. 2020; 81: 172-182
        • Shields G.S.
        • Rivers A.M.
        • Ramey M.M.
        • Trainor B.C.
        • Yonelinas A.P.
        Mild acute stress improves response speed without impairing accuracy or interference control in two selective attention tasks: Implications for theories of stress and cognition.
        Psychoneuroendocrinology. 2019; 108: 78-86
        • Thompson A.J.
        • Banwell B.L.
        • Barkhof F.
        • Carroll W.M.
        • Coetzee T.
        • Comi G.
        • et al.
        Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria.
        Lancet Neurol. 2018; 17: 162-173
        • Veldkamp R.
        • Kalron A.
        • Baert I.
        • Hämäläinen P.
        • Tacchino A.
        • D'Hooge M.
        • et al.
        Differential effects and discriminative validity of motor and cognitive tasks varying in difficulty on cognitive-motor interference in persons with multiple sclerosis.
        Mult. Scler. 2021; (1352458520986960)
        • Wajda D.A.
        • Motl R.W.
        • Sosnoff J.J.
        Correlates of dual task cost of standing balance in individuals with multiple sclerosis.
        Gait Posture. 2014; 40: 352-356
        • Wajda D.A.
        • Wood T.A.
        • Sosnoff J.J.
        The attentional cost of movement in multiple sclerosis.
        J. Neural Transm. 2019; 126: 577-583