Statement of purpose Evaluate a mentored intervention program designed to teach safe play techniques and thereby reduce head impact frequency and severity in high school football players. The behavioural modification program is based on analysing each player’s head impact biomechanics.

Methods and Approach In a clustered quasi-experimental study design, the intervention was provided to two out of three teams in year one, and two out of four teams in year two. All four teams will receive the intervention in year three. Impacts were measured using in-helmet accelerometers worn by all players. Impact data and game video were used to identify ‘high risk’ players on intervention teams to receive individualised mentoring. Mentoring effects were evaluated using head impact data. Coach and player satisfaction were assessed to evaluate the program’s acceptability and impact.

Results In total, 142 athletes (63 intervention; 69 control) have participated to date. Of these, 18 (25.8%) were classified ‘high risk’, with playing techniques amenable to intervention. On average, the proportion of head impacts classified as high-magnitude (>60 g) dropped 3.9% from pre to post-intervention in mentored players. The proportion of top-of-head impacts (considered a high-risk impact zone) dropped 4.7%. There was minimal (<1%) change on these measures in non-mentored players. In post-season exit surveys, all mentored players either ‘strongly agreed’ (56%) or ‘agreed’ (44%) that his playing behaviours improved because of the study. All intervention team coaches thought the study was ‘very effective’ (n=6, 40%) or ‘effective’ (n=9, 60%) in changing playing behaviour.

Conclusions To a modest degree, mentoring decreased impacts associated with increased concussion risk. Athletes and coaches viewed the program as effective.

Significance and contributions to Injury and Violence Prevention Science Behavioural interventions have potential as a means of reducing head impact frequency and severity in contact sports.