OBJECT: The authors investigated return-to-play duration for initial and recurrent concussion in the same season in 2 teams of junior (16-21-year-old) ice hockey players during a regular season. METHODS: The authors conducted a prospective cohort study during 1 junior regular season (2009-2010) of 67 male fourth-tier ice hockey players (mean age 18.2 +/- 1.2 years [SD], range 16-21 years) from 2 teams. Prior to the start of the season, every player underwent baseline assessments that were determined using the Sideline Concussion Assessment Tool 2 (SCAT2) and the Immediate Post-Concussion Assessment and Cognitive Test (ImPACT). The study protocol also required players who entered the study during the season to complete a baseline SCAT2 and ImPACT. If the protocol was not followed, the postinjury test results of a player without true baseline test results were compared with previously established age- and sex-matched group normative levels. Each game was directly observed by a physician and at least 1 neutral nonphysician observer. Players suspected of suffering a concussion were evaluated by the physician during the game. If a concussion was diagnosed, the player underwent clinical evaluation at the physician’s office within 24 hours. The return-to-play decision was based on clinical evaluation guided by the Zurich return-to-play protocol (contained in the consensus statement of international expert opinion at the 3rd International Conference on Concussion in Sport held in Zurich, November 2008). This clinical evaluation and return-to-play protocol was augmented by the 2 tests (SCAT2 and ImPACT) also recommended by the Zurich consensus statement, for which baseline values had been obtained. RESULTS: Seventeen players sustained a physician-observed or self-reported, physician-diagnosed concussion during a physician-observed ice hockey game. The mean clinical return-to-play duration (in 15 cases) was 12.8 +/- 7.02 days (median 10 days, range 7-29 days); the mean number of physician office visits by players who suffered a concussion (15 cases) was 2.1 +/- 1.29 (median 1.5 visits). Five of the 17 players who sustained a concussion also suffered a recurrent or second concussion. One of the 5 individuals who suffered a repeat concussion sustained his initial concussion in a regular season game that was not observed by a physician, and as a result this single case was not included in the total of 21 concussions. This initial concussion of the player was identified during baseline testing 2 days after the injury and was subsequently medically diagnosed and treated. The mean interval between the first and second concussions in these 5 players was 78.6 +/- 39.8 days (median 82 days), and the mean time between the return-to-play date of the first and second concussions was 61.8 +/- 39.7 days (median 60 days). CONCLUSIONS: The mean rates of return to play for single and recurrent concussions were higher than rates cited in recent studies involving sport concussions. The time interval between the first and second concussions was also greater than previously cited. This difference may be the result of the methodology of direct independent physician observation, diagnosis, and adherence to the Zurich return-to-play protocol.