Improve Random Measurement Error and Practice Effect of the Tablet-based Symbol Digit Modalities Test in Patients with Schizophrenia

Background: Slow information processing speed is a common deficit of patients with schizophrenia, severely affecting their functions of daily life. The Tablet-based Symbol Digit Modalities Test (T-SDMT) is a quick and easy-to use measure assessing processing speed in patients with psychosis. However, the practice effect and substantial random measurement error have limited the utility of the T-SDMT. The purposes of this study were to explore the effect of different measuring methods on reducing the T- SDMT’s practice effect and the random measurement error. Methods: The author proposed three kinds of measuring methods on three groups of patients, respectively. The first group was measured by repeated measurements for 2 formal tests, the second group was to increase practice time for three minutes and then taking one formal testing, and the last group was the combination of increased practice times for three minutes and taking repeated measurements for 2 formal tests. The author recruited 87 patients with chronic schizophrenia. These patients were randomly assigned to 3 groups mentioned above, repeated-measurements group, increasing- practice-time group, combination of repeated measurements and increasing practice time group. Three groups of patients received test and re-test testing. The test and retest were administered 2 weeks apart. Results: The results of this study found that the three methods had good test-retest consistency. The increasing-practice-time group had the highest test-retest consistency. The combination group could effectively reduce random measurement errors. However, practice effect existed in all three groups. The increasing-practice-time group and combination group could reduce the practice effects effectively than the repeated- measurements group. Conclusion: The combination of repeated measurements and increasing practice time can reduce the random measurement errors, which was better than the other 2 groups. However, the random measurement errors needed to be further decreased for clinical utility. In addition to the practice effect, the future studies should further increase practice time and the number of testing (e.g., 3 formal tests), in order to enhance the utility of the T-SDMT.