A new device - shoulder active movement extent discrimination apparatus (Shoulder AMEDA) - was developed to access the shoulder proprioception using backstroke and freestyle patterns, and the reliability and validity of the Shoulder AMEDA were investigated.
Fifteen recreational swimmers without shoulder pain volunteered (8 males, 7 females; age=23.7±2.3 years; height=1.66±0.69 m; weight=59.5±8.9 kg). All participants completed the shoulder proprioception tests on Shoulder AMEDA by mimicking the backstroke and freestyle arm swing. During the test, participants were instructed to stand upstraight with barefoot and wore a sleeveless shirt to eliminate tactile input from the cloth. The Shoulder AMEDA was adjusted to the height of one arm above the head of the participant. When facing towards the apparatus, participants mimicked the freestyle swimming by swinging their testing arm from hanging naturally to overhead until the hand touching the Shoulder AMEDA; similarly, when facing backward the apparatus, participants mimicked the backstroke swimming. The Shoulder AMEDA was manually controlled to generate 4 shoulder flexion extents of 168, 166, 164, and 162 degrees. During the testing, the 4 different positions were presented 10 times for each in a random order. Participants were asked to make an absolute judgment on each position presented, based on proprioceptive information. After each test, participants were asked to rate the difficulty levels of the task, by giving a number between 0 (hardest) to 10 (easiest). Participants were retested within a 7-day interval to investigate test-retest reliability. Shoulder proprioceptive discrimination sensitivity scores as mean pairwise Area Under the ROC Curve(AUC) values were calculated by using SPSS software. Test-retest reliability was assessed via the inter-class correlation coefficient (ICC). The Pearson correlation coefficient was used to correlate the AUC value and the difficulty levels of the task.
The ICC(3,1) was 0.76 with 95% CI=0.40-0.91 for backstroke and 0.74 with 95% CI=0.38-0.90 for freestyle. The paired-sample t-test showed that the proprioceptive AUC discrimination scores for freestyle were significantly better than that of backstroke (95% CI=0.03-0.08, p0.01), and the perceived difficulty level of discriminating shoulder movement extents when mimicking freestyle was significantly lower than mimicking backstroke (95%CI=0.13-2.13, p=0.03) The Pearson correlation analysis showed no significant correlation between difficulty leaves and AUC values in freestyle and backstroke, respectively (r=-0.93, p=0.74; r=-0.25, p=0.37).
The Shoulder-AMEDA showed good reliability and validity for assessing shoulder proprioception in backstroke and freestyle patterns, and thus can be used for testing shoulder proprioception in swimmers.
Given that shoulder proprioception tested in the backstroke pattern was significantly worse, which was consistent with participants’ perceived difficulty level, task-specific physiotherapy training program should be developed to improve shoulder proprioception in backstroke
motor control
injury prevention
