Shoulder proprioception: How is it measured and is it reliable? A systematic review

Study Design: Systematic review. Introduction: Constituents of proprioception include our awareness of the position (joint position sense ([JPS]) and motion (kinesthesia) of our limbs in space. Proprioceptive deficits are associated with musculoskeletal disorders but remain a challenge to quantify, particularly at the shoulder. Purpose of the Study: To report the psychometric values of validity, reliability, and responsiveness for shoulder JPS and/or kinesthesia protocols. Methods: A review of 5 databases was conducted from inception to July 2016 for studies reporting a psychometric property of a shoulder proprioception protocol. The included studies were evaluated using the QualSyst checklist and COSMIN 4-point scale. Results: Twenty-one studies were included, yielding 407 participants and 553 evaluated shoulders (n). The included studies support excellent methodological scores using the QualSyst checklist (88.1 +/- 9.9%) and good psychometric scores with the COSMIN for reliability (71.1%) and moderate-to-low quality score (50%) for criterion validity. Weighted average intraclass correlation coefficients (ICCs) for intrarater reliability were highest for passive JPS and kinesthesia, ICC = 0.92 +/- 0.07 (n = 214) and ICC = 0.92 +/- 0.04 (n = 74), respectively. The most reliable movement and tool are internal rotation at 90 of abduction, ICC = 0.88 +/- 0.01 (n = 53), and the dynamometer, ICC = 0.92 +/- 0.88 (n = 225). Only 2 studies quantify an aspect of validity and no responsiveness indices were reported among the included studies. Conclusion: Based on the results of the included studies, the evaluation of shoulder proprioception is most reliable when using a passive protocol with an isokinetic dynamometer for internal rotation at 90 of shoulder abduction. Standardized protocols addressing the psychometric properties of shoulder proprioception measures are needed. (C) 2017 Hanley & Belfus, an imprint of Elsevier Inc. All rights reserved.