Statistical analysis
Statistical analysis was performed using Statistical program R 3.6
version and R Studio 1.2.1335, version. The sample size was calculated
by r package (”MKmisc”), with command power. diagnostic.test, for
diagnostic test with sensitivity of 0.95, specificity of 95 Significant
level = 0.05, delta = 0.1, power = 0.8 as suggested by Cohen. The sample
size required was 68 patients. We decided to remain with 60 patients as
it does not alter the statistical power. The distribution of continuous
data was assessed by Shapiro-Wilk’s test, a comparison of capillary
Hemoglobin, lactate, and shock index values between severe and
non-severe haemorrhage groups was done by t-Test in case of normal
distribution and Wilcox test for non-normal distribution data. A
Receiver Operating Curve analysis was done to determine cut points,
Sensitivity, specificity, and performance of each test at 0, 15, and 30
minutes from the diagnosis.