Table 1.

Overview of studies evaluating instruments for cognitive assessment for pediatric-onset multiple sclerosis (POMS).

Author (Year)	Test/Battery	Participants	Subtests and Targeted Cognitive Domains	Main Findings and/or Additional Notes
Portaccio et al. (2009) [13]	Brief Neuropsychological Battery for Children (BNBC)	61 POMS (age 8.8–17.9 years),58 matched HC	WISC vocabulary: language SDMT: processing speed, attention TMT: processing speed, attention, executive function SRT: verbal learning/memory	Sensitivity: 96%, specificity: 76% (cut-off: failure on at least one test) ~30 min
Smerbeck et al. (2011) [15]	Brief Visuospatial Memory Test–Revised (BVMTR)	51 POMS, 4 with ADEM, 92 HC (age 6–17)	visuospatial learning/memory	Regression-based pediatric norms Significantly poorer performance in pediatric patients with demyelinating disorders, with medium and large effect size for BVMTR (Cohen’s d −0.38~−0.71) and SDMT (Cohen’s d −1.30) respectively, between children with demyelinating disorders and HC.
	SDMT (oral version)	22 POMS, 3 with ADEM, 92 HC (age 6–17)	processing speed, attention
Smerbeck et al. (2011) [20]	Brief Visuospatial Memory Test–revised (BVMTR)	43 POMS (age 9–18), 43 HC (age 9–18)	visuospatial learning/memory	significant difference between groups (Cohen’s d 0.9)
	SDMT (oral version)	20 POMS (age 9–18), 20 HC (age 8–18)	processing speed, attention	significant difference between groups (Cohen’s d 0.69)
Smerbeck et al. (2012) [16]	National MS Society Consensus Neuropsychological Battery for Pediatric Multiple Sclerosis (NBPMS)	51 POMS (age 9–18), 102 HC (age 5–18)	WASI: intelligence Grooved Pegboard Test: sensorimotor EOWPVT: language DKEFS Verbal Fluency: language Beery–Buktenika Test of Visual-Motor Integration: visuospatial processing CVLT-C: verbal learning/memory CPT-II: executive function, attention WISC-IV Digit Span: working memory WISC-IV Coding B: processing speed, attention Contingency Naming Test: executive function DKEFS TMT: executive function	Manual-based and regression-based (demographically adjusted) pediatric norms correlated strongly (r > 0.7) for all 30 variables. In 19 out of 30 variables, regression-based norms more readily detected neuropsychological impairment in POMS.
Charvet et al. (2014) [18]	Symbol Digit Modalities Test (SDMT, oral version)	70 POMS (70 underwent SDMT, 31 underwent neuropsychological testing), 40 other pediatric neurological diagnoses, 32 HC (note: significant difference in racial distribution between MS and HC)	processing speed, attention	SDMT showed 77% sensitivity and 81% specificity for neuropsychological impairment when the latter was done within one year, 100% sensitivity when the latter was done within two months. SDMT z score was significantly correlated with neuropsychological evaluation aggregate z score (r = 0.62, p < 0.001). Impaired SDMT performance in 37% of POMS and 9% of HC.
Bigi et al. (2017) [17]	Computer-Based Symbol Digit Modalities Test (c-SDMT)	27 POMS (22 female, 81.5%; age 8–18 years), 478 HC (237 female, 49.5%)	processing speed, attention	Regression analysis showed that increasing age (in the range 8–17) was significantly associated with better performance in HC. High test–retest reliability (ICC = 0.91) in HC. Total time to complete the task did not differ between POMS and HC, but POMS patients were less likely to show successively better performance over latter part of the task. Over 85% of participants (HC and POMS) indicated that they liked the test.
Charvet et al. (2018) [19]	Brief International Cognitive Assessment for Multiple Sclerosis battery (BICAMS)	69 POMS (7–21 years), 66 HC (8–21 years)	SDMT: processing speed, attention BVMTR: visuospatial learning/memory RAVLT: verbal learning/memory	Specificity: 91% Detection rate of cognitive impairment: 26% ~15 min
	Cogstate Brief Battery	67 POMS, 48 HC	Three speeded processing tasks: Detection: processing speed Identification: attention One-Back: working memory	Specificity: 92% Detection rate of cognitive impairment: 27% Detection and identification tasks (but not one-back) significantly discriminated between POMS and HC. ~15 min BICAMS and Cogstate agreed in the classification of impairment in 74% of the full sample (69% and 85% agreement for POMS and HC, respectively).
Kapanci et al. (2019) [8]	See subtests column (the study examined the interrelationships of tests tapping processing speed, working memory, and intelligence)	21 POMS, 21 matched HC	Reaction time task: processing speed Working memory task: working memory Cattell’s Culture Fair Test: intelligence	Intelligence measured by Cattell’s Culture Fair Test was significantly lower in POMS compared to HC. 33% of the variance in psychometric intelligence between POMS and HC was explained by differences in RT task performance. No difference in WM task performance between POMS and HC.
Brenton et al. (2019) [7]	See subtests column	20 POMS, 40 matched HC	SDMT: processing speed, attention PASAT (as a component of Multiple Sclerosis Functional Composite): processing speed, attention, working memory	POMS patients performed significantly lower on SDMT (p = 0.0002) and PASAT (p = 0.004). No significant correlation between SDMT z score and EDSS.
Falco et al. (2019) [14]	Rao’s Brief Repeatable Battery (BRB)	76 HC (age 14–17)	SRT and SRT-D: verbal learning/memory SPART and SPART-D: visuospatial learning/memory SDMT: processing speed, attention PASAT: processing speed, attention, working memory WLG: language (verbal fluency), executive function	Regression analysis showed that gender, age, and education were important variables in adolescent population. Younger age, male gender, and educational attainment were individually associated with better performance on SPART and SPART-D. Male gender was also associated with better performance on PASAT.
Storm Van’s Gravesande et al. (2019) [21]	Multiple Sclerosis Inventory of Cognition for Adolescents (MUSICADO)	106 POMS (age 12–18 years), 210 HC	Phonemic verbal fluency task (RWT “s-words”): executive function, language (verbal fluency) TMT-A: processing speed, attention Digit Span Forward: working memory	The phonemic verbal fluency task (RWT “s-words”), TMT-A, and Digit Span Forward tasks discriminated significantly between POMS and HC (p < 0.001, respectively). Specificity of MUSICADO: 88.6% Failure rate in POMS: RWT “s-words” 24.5%; TMT-A 17.9%; Digit Span Forward 15.1%.

Abbreviations: ADEM, acute disseminated encephalomyelitis; BICAMS, brief international cognitive assessment for multiple sclerosis battery; BVMTR, brief visuospatial memory test–revised; CPT-II, Conner’s continuous performance test—second edition; CVLT-C, California verbal learning test for children; DKEFS, Delis–Kaplan executive function system; EDSS, expanded disability status scale; EOWPVT, expressive one word picture vocabulary test; HC, healthy controls; ICC, intraclass correlation coefficient; PASAT, paced auditory serial addition test; POMS, pediatric-onset multiple sclerosis; RAVLT, Rey auditory verbal learning test; RT, reaction time; RWT, Regensburger Wortflüssigkeitstest; SDMT, symbol digit modalities test; SPART, spatial recall test; SPART-D, SPART delayed recall; SRT, selective reminding test; SRT-D, SRT delayed recall; TMT, trail-making test; WASI, Wechsler abbreviated scales of intelligence; WISC-IV, Wechsler intelligence scale for children—fourth edition; WLG, word list generation; WM, working memory.