Although practice effects (PE) on repeated cognitive testing have received growing interest in Alzheimer's disease, they have been understudied in Parkinson's disease (PD). The current paper examined PE across one week in a sample of patients with PD via traditional methods and regression-based change scores, as well as if these change scores relate to clinical variables in PD. Thirty-five patients with PD were administered a brief cognitive battery twice across approximately one week. Using both simple-difference and standardized regression-based change scores, a series of one-sample and independent t-tests were calculated to assess for PE across the test battery. Pearson correlations examined both types of change scores and measures of mood and severity of motor symptoms. Whereas traditional analyses (i.e., simple difference scores and dependent t-tests) did not reveal any changes on test scores over this interval, regression-based change scores did identify that these individuals showed significantly smaller-than-expected PE on three of the seven cognitive scores. Furthermore, when these regression-based change scores were trichotomized (decline/stable/improve), four of the seven tests showed significantly more decline than expected in this sample. Finally, these regression-based change scores significantly correlated with motor measures, with smaller PE being associated with worse motor functioning. Although these results are preliminary and need to be replicated in larger and more diverse samples, smaller-than-expected PE are seen in PD and they may signal more advanced disease.
Read full abstract