Background The tackle is the most injurious game event in Rugby Union. We recently found that tackles account for 76% of head injury events in professional rugby. Specific analysis of the tackle is required to understand how interventions such as law change or educational programmes can minimize head-injury risk from tackles.

Objective Analyze tackles to identify high risk situations and behaviors.

Design Prospective cohort study.

Setting 1,516 Rugby matches in six professional tournaments.

Participants Head injury cases resulting from tackles.

Assessment of Risk Factors A professional video analyst evaluated every tackle-induced head injury event. Propensity was calculated for various tackle characteristics.

Main Outcome Measurements Head-injury propensity for tackles, focused on tackle type, player's speeds and body positions.

Results 464 Head Injury Events (HIEs) occurred during tackles, with a propensity of 1.94 HIEs/1000 tackles. For body position, the greatest risk was for falling ball carriers (9.8 HIEs/1000 tackles), or for upright tacklers (2.44 HIEs/1000 tackles). In combination, risk was highest when both players were upright, particularly for active shoulder tackles (17.4 HIEs/1000 tackles). Regardless of the tackler position, the lowest risk was for a ball carrier bent at the waist (1.15 HIEs/1000 tackles). Regarding speed, for static or low-speed tacklers, HIE risk increased with ball-carrier speed. However, when the tackler was at high speed, HIE risk decreased from 6.1 HIEs/1000 tackles for a static ball-carrier to 3.3 HIEs/1000 tackles for a high-speed tackler.

Conclusions Complex interplay exists between tackle type, body position and speeds. The ability to anticipate and protect oneself in a tackle is crucial for reduced risk, as is the correct selection of tackle type and direction based on relative speeds. The execution of higher force tackles to avoid high risk head impacts is also critical. Decision-making and tackle execution offer potential areas for intervention to reduce HIE risk.