Many tests are simply not that good. It's not that simple, like pay $1 for a test that with absolute certainty tells you if you have disease X or not. It would

Many tests are simply not that good. It's not that simple, like pay $1 for a test that with absolute certainty tells you if you have disease X or not. It would be good if that was the case because then it WOULD have been just a question of money. But if you have a test that is wrong 20% of the time, and you take this on all your patients every year, you will end up with a huge amount of mental suffering or possibly physical side-effects (unnecessary biopsies, or even surgery). You might end up giving people long runs of unnecessary meds. OTOH, you might end up missing some diagnoses if you don't do some tests. Balancing all of this is difficult. Of course this is not news to the medical profession, this is a big input to the standard flowcharts. And you can always find anecdotal edge-cases that think their particular case was crystal clear and should have immediately been diagnosed. But don't assume that medical diagnosis and treatment is as trivial as doing a blood test and some googling.