Table 2.
Effectiveness endpoints (modified intention-to-treat population)a
| Endpoints | Value | P value |
|---|---|---|
| Primary endpointb | ||
| Strength responder, n (%) | 52 (86.7) | |
| Function responder, n (%) | 45 (75.0) | |
| Responder, n (%) | 43 (71.7) | < 0.001 |
| Secondary responder analysis endpointc | ||
| ARCEX non-responder, n (%) | ||
| Rehabilitation responder | 4 (6.7) | |
| Rehabilitation non-responder | 6 (10.0) | |
| ARCEX responder, n (%) | ||
| Rehabilitation responder | 34 (56.7) | |
| Rehabilitation non-responder | 16 (26.7) | 0.012 |
| Change from end of rehabilitation alone to end of ARCEX Therapy | P value | |
| Secondary hierarchical endpointd | ||
| Pinch force | 4.8 ± 16.1 | 0.002* |
| GRASSP-Prehension Performance | 1.6 ± 2.9 | <0.001 |
| GRASSP-Strength | 2.8 ± 5.4 | <0.001* |
| UEMS | 2.2 ± 3.2 | <0.001* |
| TSS | 9.6 ± 15.1 | <0.001 |
| EQ-5D-5L | 1.7 ± 14.1 | 0.028* |
| SCIM III total score | 0.5 ± 4.2 | 0.101* |
| WHOQOL-Physical Domain | 1.6 ± 8.0 | − |
| WHOQOL-Psychological Domain | −0.4 ± 9.4 | − |
| WHOQOL-Social Relationships | 1.5 ± 10.3 | − |
| WHOQOL-Environment | −0.4 ± 7.5 | − |
a Plus–minus values are means ± s.d. The modified intention-to-treat population included all participants who underwent at least 24 sessions (minimum 12 sessions per month) during the rehabilitation-alone period and at least 24 sessions during the ARCEX Therapy period.
b The primary effectiveness endpoint tested the hypothesis that more than 50% of participants would meet responder criteria for both strength and function. This hypothesis was evaluated using a one-sided exact binomial test.
c Superiority of ARCEX Therapy compared to rehabilitation alone was tested by comparing the proportion of participants who converted into responders from enrollment to the end of the rehabilitation-alone period with the proportion of participants who converted from enrollment to the end of the ARCEX Therapy period. This secondary endpoint was tested using McNemar’s test.
d Changes in individual outcomes were assessed from enrollment to the end of ARCEX Therapy compared to changes from enrollment to the end of rehabilitation alone using a one-sided paired t-test or a paired Wilcoxon rank-sum test if data were not normal. The order of the secondary endpoints according to the hierarchical analysis plan is also provided in the Supplementary Information. The hierarchy failed at the seventh endpoint (Spinal Cord Independence Measure, Version III (SCIM III) total score). Dashes indicate P values that are not provided due to failure of the hierarchical testing of results. * indicates a parametric test.