Traumatic brain injury is a global health priority. The burden is highest in the western pacific region, and it is estimated that two-thirds of patients in rural Papua New Guinea die before hospital admission. Managing traumatic brain injury is further compounded by limited investigations and neurosurgery services. The aim of this study was to investigate the potential factors of mortality amongst patients with moderate and severe head injuries. A retrospective cohort study was conducted from two provincial hospitals in Papua New Guinea. Potential factors of mortality were investigated by using logistic regression analysis. There was a significant odds of reduced mortality in patients with vomiting and headache (OR 0.16, 95% CI 0.04–0.69, p = 0.0132), reactive pupils (OR 0.02, 95% CI 0.00–0.17, p = 0.0005), a higher GCS (OR 0.77 for every 1 point increase in GCS, 95% CI 0.63–0.95, p = 0.0147), and length of hospital stay (OR 0.84 per 1 day increase, 95% CI 0.72–0.98, p = 0.0258). In contrast, mortality was increased with the use of mannitol (OR 9.17, 95% CI 1.34–62.71, p = 0.0239), hypoxia (OR 20.91, 95% CI 4.00–109.37, p = 0.0003), presence of complications (OR 5.25, 95% CI 1.41–19.51, p = 0.0133), and admission to KPH compared with APH (OR 4.71, 95% CI 1.25–17.75, p = 0.0222). This study highlights potential factors associated with traumatic brain injury mortality in rural Papua New Guinea. The findings can help direct policy makers, assist in public health awareness, and improve surgical research, care, and management for patients with traumatic brain injury in rural Papua New Guinea.