Combined Statistical Analysis Method Assessing Fast Versus Slow Movement Training in a Patient With Cerebellar Stroke: A Single-Case Study

Abstract

Gold standards of data analysis for single-case research do not currently exist. The purpose of this study was to determine whether a combined statistical analysis method is more effective in assessing movement training effects in a patient with cerebellar stroke. A crossover single-case research design was conducted. The patient was a 69-year-old man with a chronic cerebellar infarct who received two 5-week phases of finger tracking training at different movement rates. Changes were measured with the Box and Block Test, the Jebsen-Taylor test, the finger extension force test, and the corticospinal excitability test. Both visual analysis and statistical tests (including split-middle line method, t test, confidence interval, and effect size) were used to assess potential intervention effects. The results of the t tests were highly consistent with the confidence interval tests, but less consistent with the split-middle line method. Most results produced medium to large effect sizes. The possibility of an incomplete washout effect was a confounding factor in the current analyses. The combined statistical analysis method may assist researchers in assessing intervention effects in single-case stroke rehabilitation studies.