Table 2. Results of the Fully Adjusted Linear Regression Mixed-Effects Modelsa.
| Analysis | Mean between-group difference (2-sided 80% CI) | Futility test P value |
|---|---|---|
| Primary analysis | ||
| 24-mo Change in MDS-UPDRS part III score in mITT population | 1.52 (−0.77 to 3.8) | .003 |
| Sensitivity analyses | ||
| 24-mo Change in MDS-UPDRS part III score | ||
| Excluding outliers in the mITT population | 0.82 (−1.08 to 2.72) | .006 |
| Visit location in the non–COVID-19 population | 1.62 (−0.70 to 3.93) | .006 |
| Unrestricted imputation | ||
| mITT population | 1.10 (−1.16 to 3.36) | .01 |
| COVID-19 population | 1.51 (−0.78 to 3.81) | .007 |
| CACE | ||
| Compliance (i) | 2.17 (−1.29 to 5.64) | .03 |
| Compliance (ii) | 2.08 (−1.26 to 5.42) | .03 |
| Treatment complianceb | 1.74 (−1.03 to 4.51) | .02 |
| Post hoc analysis | ||
| 12-mo Change in MDS-UPDRS part III score | ||
| mITT populationb | 2.74 (0.70 to 4.77) | <.001 |
| Excluding outliers | 2.55 (0.75 to 4.35) | <.001 |
Abbreviations: CACE, complier average causal effect; mITT, modified intention to treat.
a
Adjusted for baseline, age, sex, Parkinson disease duration, Hoehn and Yahr stage as fixed effects and study center as a random effect, with futility tests of the primary outcome for the primary, planned sensitivity, and post hoc analyses.
b
Post hoc analysis.