Biostats/Bored=Review.

There’s nothing I can say to make this question less painful…

What statistical value will help you determine how much you should change your clinical decision based on the test's result (terrible hint: it is independent of prevalence)?

Likelihood ratio.

Yes. The LR+ gives you a tool to gage how much a positive or negative result can be used in your clinical judgement. This, along with a pretest probability helps you use your tools more wisely. A higher LR+, for instance, means that you have a higher probability to have a True Positive result.

How do you calculate it?

LR+= Sensitivity / 1-Specificity

LR-=1-Sensitivity / Specificity

What the hell is up with that image, you ask?

Taken by Kevin Frayer of the Associate Press, this is an Indian Muslim Sufi performing an act of self flagellation during a festival in Rajasthan celebrating one of their saints.