Objectives The purpose of this paper is to provide reliable change metrics and assess the test–retest psychometrics of the Sport Concussion Assessment Tool fifth Edition (SCAT5) in a multilingual sample of professional ice hockey players.

Methods The SCAT5 was administered to National Hockey League and American Hockey League male professional ice hockey players as part of the 2018 and 2019 season medical evaluations. We extracted data from an existing clinical database of players who were administered two consecutive baseline evaluations and had no intervening concussions to assess psychometric properties (eg, test–retest reliability) and develop reliable change metrics.

Results Overall, 1007 players met inclusion criteria. Players were reassessed on average 344.52 (±62.52) days following their first baseline. Test–retest reliability ranged from r=0.3 to r=0.5. Reliable change metrics for Immediate Memory varied considerably dependent on form order. Significant differences were found between English versus non-English language preference on Immediate Memory and Concentration but not Delayed Memory or the modified Balance Error Scoring System (mBESS). Reliable change indices at the 90% CI were: Symptom Severity (5), Total Symptoms (3), Immediate Memory (3-8), Concentration (2), mBESS (5) and Delayed Memory (2-4).

Conclusions Our findings highlight the importance of considering form differences and language preference when determining reliable change from baseline on the SCAT5 in male professional hockey players. The relatively modest test–retest reliability of the measures leaves room for improved psychometric properties and increased sensitivity to concussion in future iterations of the SCAT.