The evaluation of preventable trauma deaths can serve as an audit on the quality of trauma care provided by a hospital. The trauma injury severity score (TRISS) and clinical audit method were used retrospectively in evaluating 126 trauma mortalities admitted at the surgery ward of the Tarlac Provincial Hospital from January 1, 1995 to December 7,1997. Majority of these deaths occurred in ages below 40 years (67.46 percent) with a male to female ratio of 3:1. Ninety one trauma deaths (72.22 percent) were classified as unexpected deaths by the TRISS method while 35 (27.78 percent) were considered expected deaths. The clinical audit method classified 89 mortalities (70.63 percent) as preventable and 37(29.36 percent) as non-preventable. The preventable death rate was 70.63 percent. The TRISS method has a sensitivity of 80.89 percent and specificity of 48.65 percent. Its positive predictive value was 79.12 percent. Vehicular accidents topped the leading causes of mortality accounting for 58 of the 126 mortalities (46.03 percent) followed by burns at 22 (17.46 percent) and falls at 18 (14.29 percent). As to the immediate cause of death, intracranial injuries accounted for 55 mortalities in both the preventable and non-preventable deaths (35.95 percent) and 62.16 percent respectively) followed by 23 hypovolemic deaths, and 19 deaths due to pulmonary insufficiency. Preventive measures in the community should focus on vehicular and road safety while in-hospital improvements should concentrate on efforts to upgrade our capability of handling severe craniocerebral injuries. (Author)