Predicting Prognosis of Isolated Head Injury: A Computer-Based Model with Simple Variables

Abstract

Objective: Over one hundred predictive models were defined in the past for head injury (HI) prognosis, but none of them have been widely used up to the present. The aim of this study is to predict the prognosis of isolated HI patients by simply using data from the first day after injury.Materials and Methods: Data of head injury patients in Trakya University Hospital between January 1996 and December 2006 were obtained from records. The age, gender, causes of HI, basic neurologic examination findings, radiologic findings and discharge status are examined.Results: Most of the data were simplified as absent (0) and present (1), and mortality rates for each groups were accepted weighted values. All data were processed statistically and two models were created. Model 1 with Glasgow Coma Scale (GCS) score predicted the mortality/vegetative event at a rate of 56.5%, and the conscious survival event at 98.7%. Model 2, without the GCS score, predicted the mortality/vegetative event at 55.1% rate and the conscious survival event at 99.2%.Conclusion: Both models could be used for informing the patient and relatives and helping them to understand the severity of HI in busy working conditions of emergency departments.

Keywords

• Computer-based model
• head injury
• outcome
• prediction