Musculoskeletal injuries are common during military and other occupational physical training programs. Employers have a duty of care to reduce employees' injury risk, where females tend to be at greater risk than males. However, quantification of principle co-factors influencing the sex-injury association, and their relative importance, remain poorly defined. Injury risk co-factors were investigated during Royal Air Force (RAF) recruit training to inform the strategic prioritization of mitigation strategies. A cohort of 1,193 (males n = 990 (83%); females n = 203 (17%)) recruits, undertaking Phase-1 military training, were prospectively monitored for injury occurrence. The primary independent variable was sex, and potential confounders (fitness, smoking, anthropometric measures, education attainment) were assessed pre-training. Generalized linear models were used to assess associations between sex and injury. In total, 31% of recruits (28% males; 49% females) presented at least one injury during training. Females had a two-fold greater unadjusted risk of injury during training than males (RR = 1.77; 95% CI 1.49-2.10). After anthropometric, lifestyle and education measures were included in the model, the excess risk decreased by 34%, but the associations continued to be statistically significant. In contrast, when aerobic fitness was adjusted, an inverse association was identified; the injury risk was 40% lower in females compared with males (RR = 0.59; 95% CI: 0.42-0.83). Physical fitness was the most important confounder with respect to differences in males' and females' injury risk, rather than sex alone. Mitigation to reduce this risk should, therefore, focus upon physical training, complemented by healthy lifestyle interventions.