I have this variable that is clinically worse the lower it is. Instead of interpreting, for example, a hazard ratio of 0.9 as per unit increase in variable x, there is a x% reduction in risk for mortality, I want to get a 1 HR (for ex 1.05) and say per unit decrease in variable x there is a 5% higher risk for mortality.
How do I handle this using SPSS?
Topic spss regression
Category Data Science
With a simple linear regression you have a dependent and independent variable, with an outcome model of:
y = - mx + c
If I understand your problem, y would be mortality, and x would be your clinical variable (i.e. when x increases y decreases). And it would be the same if you want to do your other example, because there is an inverse relationship between the two (i.e. when you decrease x, y increases).
It sounds like your hazard ratio is m (i.e. the relationship between the clinical variable and the mortality)
m = (y - c )/x
So to getting a HR>1 will depend on your data.
Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.