Blood test that finds 50 types of cancer is accurate enough to be rolled out

Re:Blood test that finds 50 types of cancer is accurate enou... by captainmuon on Friday June 25, 2021 @04:23PM (#27635808)

I believe that's what the data says, but I have a really hard time reconciling it with common sense. More information is strictly better than less information, because you can always choose to ignore the extra information.

I would say most medical decisions are made either due to statistics, or due to experience. What treatment has the highest chance of making the patient better, extending their life, or giving the best quality of life? You'd "just" have to adjust the tables for the new test.

I mean in a contrived example, you could have the lab technician themself look up the numerical result (xyz > 100, abc < 10 whatever) in some table, and then there would be the rule to throw the result in the bin and report "don't treat" because this results in the best outcomes. I don't want to have all that extra diagnostic information, but I want my doctors to use it conditionally to improve my treatment if possible.

[ Parent | on HN ]

Re:Blood test that finds 50 types of cancer is accurate enou... by Fomite on Friday June 25, 2021 @06:49PM (#27637207)

"I believe that's what the data says, but I have a really hard time reconciling it with common sense."

Can you see why some people might be hesitant about basing medical treatment on "That's what the data says, but it doesn't conform to my priors, so it's probably wrong"?

People are terrible at ignoring information. Clinicians are people. We know this.

[ Parent | on HN ]

Re:Blood test that finds 50 types of cancer is accurate enou... by Jap2-0 on Friday June 25, 2021 @05:56PM (#27636754)

> you can always choose to ignore the extra information.

Can you?

That is to say: you can try, but I don't think most people are very good at consciously choosing not to think or worry about something that they know, if that's even possible.

[ Parent | on HN ]