As there was no similar previous study conducted using the time to an event data model, the sample size was determined by using single population proportion formula with the following assumptions; 52.8% proportion of death among head injury patients in the ICU [17 ], 95% confidence level and 5% margin of error. Based on this, the actual sample size for the study was computed by using the following formula: n=Z2P1-pd2=1.9620.5281-0.5280.052=383, where n = sample size, Z = critical value of 95% CI = 1.96, P = Proportion mortality among trauma patients in the ICU (52.8%) = 0.528, D = precision (marginal error) = 0.05.
Thus, by adding 10% for possible incomplete charts and lost medical records, the total sample size was 421.
Free full text: Click here