Longitudinal IRT Detects Tremor, Nontremor Changes Over Time in Parkinson Trial

Using a unique application of longitudinal item response modeling (IRT), researchers were able to separate nontremor and tremor changes that occurred over time in an interventional Parkinson disease (PD) study. They discussed their analytic approach in a study published in the journal Movement Disorders Clinical Practice.

“Although nontremor and tremor Part 3 Movement Disorder Society–Unified Parkinson's Disease Rating Scale items measure different impairment domains, their distinct progression and drug responsivity remain unstudied longitudinally,” researchers explained in the study background. “The total score may obscure important time-based and treatment-based changes occurring in the individual domains.”

The study tested novel longitudinal unidimensional and multidimensional models on participants with early PD from the STEADY-PD III (Safety, Tolerability, and Efficacy Assessment of Isradipine for Parkinson’s Disease, placebo vs isradipine) study. Unidimensional longitudinal IRT was used to assess Part 3 data for 22 nontremor items and 10 tremor items. Multidimensional longitudinal IRT assessed progression rates over time and in response to treatment in overall motor severity, nontremor, and tremor domains.

According to the study, the models identified significant but different deterioration rates in patients’ total motor, nontremor, and tremor scores over time regardless of treatment. In total score and nontremor score, but not tremor score, isradipine was associated with significant deterioration compared with placebo. Nontremor and tremor severity changes were positively but weakly correlated (researchers reported a correlation coefficient of 0.108), further highlighting the two separate domains.

Researchers consider longitudinal IRT modeling well suited for clinical trials involving patients with PD, particularly because of its ability to separate nontremor and tremor changes over time and in response to treatments.

“In our view,” they wrote, “this type of multidimensional longitudinal IRT analysis provides a new and highly focused statistical approach to detect changes that could have clinical implications obscured by traditional analytic methods.”

Reference:
Luo S, Zou H, Goetz CG, et al. Novel approach to Movement Disorder Society-Unified Parkinson's Disease Rating Scale monitoring in clinical trials: longitudinal item response theory models. Mov Disord Clin Pract.2021 Aug 3;8(7):1083-1091. doi:10.1002/mdc3.13311