Development of a Tablet-based symbol digit modalities test for reliably assessing information processing speed in patients with stroke

Abstract

Purpose: To develop a Tablet-based Symbol Digit Modalities Test (T-SDMT) and to examine the test–retest reliability and concurrent validity of the T-SDMT in patients with stroke.

Methods: The study had two phases. In the first phase, six experts, nine college students and five outpatients participated in the development and testing of the T-SDMT. In the second phase, 52 outpatients were evaluated twice (2 weeks apart) with the T-SDMT and SDMT to examine the test–retest reliability and concurrent validity of the T-SDMT.

Results: The T-SDMT was developed via expert input and college student/patient feedback. Regarding test–retest reliability, the practise effects of the T-SDMT and SDMT were both trivial (d=0.12) but significant (p≦0.015). The improvement in the T-SDMT (4.7%) was smaller than that in the SDMT (5.6%). The minimal detectable changes (MDC%) of the T-SDMT and SDMT were 6.7 (22.8%) and 10.3 (32.8%), respectively. The T-SDMT and SDMT were highly correlated with each other at the two time points (Pearson’s r=0.90–0.91).

Conclusions: The T-SDMT demonstrated good concurrent validity with the SDMT. Because the T-SDMT had a smaller practise effect and less random measurement error (superior test–retest reliability), it is recommended over the SDMT for assessing information processing speed in patients with stroke.

Implications for Rehabilitation

• The Symbol Digit Modalities Test (SDMT), a common measure of information processing speed, showed a substantial practise effect and considerable random measurement error in patients with stroke.

• The Tablet-based SDMT (T-SDMT) has been developed to reduce the practise effect and random measurement error of the SDMT in patients with stroke.

• The T-SDMT had smaller practise effect and random measurement error than the SDMT, which can provide more reliable assessments of information processing speed.

Keywords:

• Information processing speed
• practise effect
• random measurement error
• stroke
• Symbol Digit Modalities Test

Acknowledgements

We would like to thank the participants and raters for their work during data collection.

Declaration of interest

The authors report no conflicts of interest. This study was supported by research grants from E-DA Hospital (EDAHT102004) and Chi Mei Medical Centre [CSMU-CMMC-102-05 (CMCSMU10206)] in Taiwan.