Formula Help!

Hi I am doing a project in work looking at predicted mortality. This uses a scoring system known as P-POSSUM based on various physiological parameters.

As per the website riskprediction.org.uk the equation used is:
Ln R/1-R = -9.065 + (0.1692 x physiological score) + (0.1550 x operative severity score)

I already have all of the physiological and operative scores.

Extrapolating R from the above equation using X to represent -9.065 + (0.1692 x physiological score) + (0.1550 x operative severity score)

Ln R/1-R = X

R = e^[SUP]x[/SUP] /( 1 + e^[SUP]x[/SUP] )

I have got as far as:
=(-EXP(X)/(EXP(X)-1))

For whatever reason I cannot get the same scores with the exact same data as the above website calculator for P-POSSUM
Is this something to do with rounding errors or is my formula just wrong?! help!
Thanks in advance
Chris

Hi Chris,

I wish I could help you with this, but unfortunatley my math skills aren't good enough to be able to tell if your interpretation is correct or not. (Been 15+years since I've worked with math at that level!)

Look at the formula documentation. Maybe there are known "issues" that may affect the result.

Chris,


don't know if this would help you or not but here is a little different approach. use a formula to solve for your mortality based on below. say in cell A1 you solve for x then you write your formula to equate mortality% based around that. you say you already have your Phys Scores and Op Scores. I will assume you enter them in cells like c1 and c2 (or somewhere). in A1 write your formula to solve for x

X= (0.1692 x physiological score) + (0.1550 x operative severity score)



now put a formula in say A2 to solve the equation below. this will give you your mortality %

[SUP]1/(1+ e[SUP]-x[/SUP])


Example:
in A2 enter this formula. 　
　
=SUM(1/(EXP(-A1)+1))*100
[/SUP]

here is a file that uses this same approach. you can test the results against the website you referenced.