RESUMEN
BACKGROUND: Information processing speed (IPS) is a marker for cognitive function. It is associated with neural maturation and increases during development. Traditionally, IPS is measured using paper and pencil tasks requiring fine motor skills. Such skills are often impaired in patients with neurological conditions. Therefore, an alternative that does not need motor dexterity is desirable. One option is the computerized symbol digit modalities test (c-SDMT), which requires the patient to verbally associate numbers with symbols. METHODS: Eighty-six participants (8-16 years old; 45 male; 48 inpatients) were examined, 38 healthy and 48 hospitalized for a non-neurological disease. All participants performed the written SDMT, c-SDMT, and the Test of Non-verbal Intelligence Fourth Edition (TONI-4). Statistical analyses included a multivariate analysis of covariance (MANCOVA) for the effects of intelligence (IQ) and hospitalization on the performance of the SDMT and c-SDMT. A repeated measures analysis of variance (repeated measures ANOVA) was used to compare performance across c-SDMT trials between inpatients and outpatients. RESULTS: The MANCOVA showed that hospitalization had a significant effect on IPS when measured with the SDMT (p = 0.04) but not with the c-SDMT (p = 0.68), while IQ (p = 0.92) had no effect on IPS. Age (p < 0.001) was the best predictor of performance of both tests. The repeated measures ANOVA revealed no significant difference in within-test performance (p = 0.06) between outpatient and inpatient participants in the c-SDMT. CONCLUSION: Performance of the c-SDMT is not confounded by hospitalization and gives within-test information. As a valid and reliable measure of IPS for children and adolescents, it is suitable for use in both inpatient and outpatient populations.
RESUMEN
OBJECTIVE: The objective of this study was to validate the computerized Symbol Digit Modalities Test (c-SDMT) in a Swiss pediatric cohort, in comparing the Swiss sample to the Canadian norms. Secondly, we evaluated sex effects, age-effects, and test-retest reliability of the c-SDMT in comparison to values obtained for the paper and pencil version of the Symbol Digit Modalities Test (SDMT). METHODS: This longitudinal observational study was conducted in a single-center setting at the University Children's Hospital of Bern. Our cohort consisted of 86 children (45 male and 41 female) aged from 8 to 16 years. The cohort included both healthy participants (n = 38) and patients (n = 48) hospitalized for a non-neurological disease. Forty eight participants were assessed during two testing sessions with the SDMT and the c-SDMT. RESULTS: Test-retest reliability was high in both tests (SDMT: ICC = 0.89, c-SDMT: ICC = 0.90). A reliable change index was calculated for the SDMT (RCIp = -3.18, 14.01) and the c-SDMT (RCIp = -5.45, 1.46) corrected for practice effects. While a significant age effect on information processing speed was observed, no such effect was found for sex. When data on the c-SDMT performance of the Swiss cohort was compared with that from a Canadian cohort, no significant difference was found for the mean time per trial in any age group. Norm values for age groups between 8 and 16 years in the Swiss cohort were established. CONCLUSION: Norms for the c-SDMT between the Swiss and the Canadian cohort were comparable. The c-SDMT is a valid alternative to the SDMT. It is a feasible and easy to administer bedside tool due to high reliability and the lack of motor demands.