Article Text

Download PDFPDF
Test-retest reliability of the kinarm end-point robot for assessment of sensory, motor and neurocognitive function in athletes
  1. Brian W Benson1,2,3,4,
  2. Madeline Cosh3,
  3. Cameron S Mang1,
  4. Stephen H Scott5,
  5. Chantel Debert1,6,
  6. Sean Dukelow1,2,6
  1. 1Department of Clinical Neurosciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
  2. 2Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
  3. 3WinSport Medicine Clinic, winter Sport Institute, Calgary, Alberta, Canada
  4. 4Canadian Sport Institute Calgary, Calgary, Alberta, Canada
  5. 5Department of Biomedical and Molecular Sciences, Queen’s University, Kingston, Ontario, Canada
  6. 6Hotchkiss Brain Institute, Calgary, Alberta, Canada


Objective To determine the reliability of the KINARM robot in assessing sensorimotor and neurocognitive function in athletes.

Design Test-retest reliability.

Setting Sport Medicine Clinic.

Participants Thirty-nine randomly selected athletes (mean age: 18±3 years) over 2011–2012 and 2014–2015 seasons.

Outcome measures Sixty-three parameters from five robotic tasks (Visually Guided Reaching, Position Matching, Object Hit, Object Hit and Avoid, and Trail Making B) characterising sensorimotor and neurocognitive function. Each athlete was tested three times, with an average of 43±27 days between baseline (BL) and first retest (R1), and 306±32 days between R1 and second retest (R2). Reliability was assessed by intra-class correlation coefficients (ICCs) between BL and R1, and R1 and R2. Change in task performance between assessments was also evaluated with multivariate analysis of variance (MANOVA) and post-hoc comparisons within specific task parameters.

Main results ICCs ranged from −0.03 to 0.94 and −0.52 to 0.93 for BL to R1 and R1 to R2, respectively. ICCs were moderate to high (≥0.5) for 46 of 63 parameters (68%) from BL to R1, and 48 of 63 parameters (76%) from R1 to R2. MANOVAs and post-hoc comparisons indicated improvements in performance on 12/63 parameters (19%) between BL and R1, and changes in performance on 11/63 parameters (17%) between BL and R2. No significant changes were revealed in parameter performance between R1 and R2 sessions.

Conclusions Overall, results suggest moderate to good test-retest reliability for the majority of KINARM robot task parameters in measuring baseline sensorimotor and neurocognitive function in athletes.

Competing interests Brian W Benson In the future may receive a small royalty from BKIN Technologies Ltd. in consideration for assisting with development and validation of the KINARM end-point robotic device for use in acute sport concussion assessment and management.


Stephen H Scott Co-founder and CSO of BKIN Technologies that commercialises the KINARM robot.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.