BACKGROUND Working long duty hours has often been associated with increased risk of incidents and accidents in transport industries. Despite this, information regarding the intermediate relationship between duty hours and incident risk is limited. This study aimed to test a work hours/incident model to identify the interplay of factors contributing to incidents within the aviation industry. METHODS Nine hundred and fifty-four European-registered commercial airline pilots completed a 30-item survey investigating self-report attitudes and experiences of fatigue. Path analysis was used to test the proposed model. RESULTS The fit indices indicated this to be a good fit model (χ(2) = 11.066, df = 5, p = 0.05; Comparative Fit Index = 0.991; Normed Fit Index = 0.984; Tucker-Lewis Index = 0.962; Root Mean Square of Approximation = 0.036). Highly significant relationships were identified between duty hours and sleep disturbance (r = 0.18, p < 0.001), sleep disturbance and fatigue in the cockpit (r = 0.40, p < 0.001), and fatigue in the cockpit and microsleeps in the cockpit (r = 0.43, p < 0.001). CONCLUSIONS A critical pathway from duty hours through to self-reported incidents in flight was identified. Further investigation employing both objective and subjective measures of sleep and fatigue is needed.