Table 2.
Comparisons of accelerometer data and clinical scales
| Uswatte et al. (2006)[16] | Lang et al. (2007)[10] | Thrane et al. (2011)[12] | Van der Pas et al. (2011)[8] | Wang et al. (2011)[13] | Rand et al. (2012)[16] | Liao et al. (2012)[17] | Taub et al. (2013)[15] | |
|---|---|---|---|---|---|---|---|---|
| Statistical approach | Type 3,1 intraclass correlations. | Spearman correlation | Spearman correlation | Spearman correlation | Pearson correlations | Paired t-test | ANCOVA | Paired t-test |
| Accelerometer data comparison | AMR | Duration of use | Duration of use Calculated AMR | Activity counts Calculated AMR | Calculated AMR | Activity counts, the upper extremity activity did not change. | Calculated AMR. The robot-assisted therapy group improved compared with the active control group: accelerometer F1,16=5.91, p=0.026, effect size r=0.26. | Calculated AMR. Improved from baseline to post-intervention: t=2.9, p=0.016, d’=1.2 |
| Clinical Measures of Impairment | ||||||||
| AROM | Shoulder flex r=0.30, p<0.05, elbow flex r=0.50, p=0.01, wrist ext r=0.63, p<0.01 | Changed from baseline to post-intervention: t=6.1, p=0.001, d’=2.6. | ||||||
| FMS | The duration of use of the affected arm: r=0.60, p<0.001. AMR: r=-0.85, p<0.001 | FMA improved: t=-2.9, p=0.005 | The robot-assisted therapy group improved compared with the active control group: FMA F1,16=14.32, p=0.002, effect size r=0.46 | Change in FMA from baseline to post-intervention t=4.0, p=0.005, d’=1.6 | ||||
| Modified Ashworth Scale | r=-0.31, n.s | |||||||
| Pain | Shoulder pain r=0.41, p<0.01 | |||||||
| Sensation | Composite light touch r=-0.15, n.s., joint position sense r=-0.03, n.s | |||||||
| Strength using a hand-held dynamometer | Shoulder flex r=0.34, p<0.01, elbow flex r=0.52, p<0.01, wrist ext r=0.37, p<0.01, grip r=0.42, p<0.01 | |||||||
| Gait speed | Improved: t=-4.8, p<0.001 | |||||||
| Clinical Measures of Function | ||||||||
| ARAT | r=0.40, p<0.01 | Improved: t=-4.7, p<0.001 | ||||||
| FIM | Motor r=0.67, p<0.01, UE r=0.58. p<0.01. | Improved: t=-7.6, p<0.001 | The robot-assisted therapy group improved compared with the control group: FIM F1,16=0.03, p=0.88, effect size r=0.002 | |||||
| WMFT | Function r=0.62, p<0.01; time r=-0.65, p<0.01 | |||||||
| BBS | BBS improved: t=-6.4, p<0.001 | |||||||
| 6MWT | 6MWT improved: t=-4.8, p<0.001 | |||||||
| Clinical Questionnaires | ||||||||
| ABILHAND | At baseline and post treatment: r=0.45–0.54, p<0.01. | The robot-assisted therapy group improved compared with the control group: F1,16=4.76, p=0.043, effect size r=0.22 | ||||||
| MAL | AMR was correlated with QOM r=0.52, p<0.01 and AOU r=0.47, p<0.01. Less-impaired arm accelerometry was not correlated with QOM r=0.14, n.s. and AOU r=0.14, n.s. | Bilateral arm activity (mean of 2 arms): MAL-26AOU Scale r=0.37, p<0.01. MAL-26AOU Scale r=0.37, p<0.01. | The robot-assisted therapy group improved compared with the control group: MAL AOU F1,16=9.39, p=0.007, effect size r=0.36, MAL QOM F1,16=13.48, p=0.002, effect size r=0.44 | Change in FL-MAL Arm Use scale from baseline to post-intervention: t=7.4, p=0.001, effect size(d’)=3.0 | ||||
| AMR: MAL-26AOU Scale r=0.60, p<0.001, MAL-26QOM Scale r=0.66, p<0.001. | ||||||||
| Affected arm activity: MAL-26AOU Scale r=0.58, p<0.001, MAL-26QOM Scale r=0.65, p<0.001. | ||||||||
Note: AMR – arm movement ratio, AROM - Active Range of Motion, ARAT – Action Research Arm Test, WMFT – Wolf Motor Function Test, FIM – Functional Independence Measure, FIM UE – FIM Upper Extremity, FMS – Fugl-Meyer Scale, MAL – Motor Activity Log, LF-MAL – lower functioning MAL, MAL-26 QOM – MAL-26 Quality of Movement, MAL-26 AOU – MAL-26 Amount of Use, CIMT – Constraint-Induced Movement Therapy, SIS – Stroke Impact Scale, NEADL – Nottingham Extended Activities of Daily Living, BBS – Berg Balance Scale, 6MWT – 6 Minute Walking Test.