Interesting analogy and I see the parallel, mostly. There are different kinds of mechanics. Most mechanics will "diagnose" by plugging in the scanner, reading the code, and replacing whatever the book says to replace. Replacing parts is analogous to prescribing drugs: when the first attempt fails to solve the problem, they swap more parts (prescribe more drugs). Often it can take several trips to the shop to get the problem correctly fixed. Likewise with typical docs: they try different drugs until the symptoms go away and sometimes takes several trips to make the symptoms go away for good.

There are a few mechanics that can actually figure things out. They go beyond the code reader and book and try to figure out what's really going on. Instead of swapping parts until the problem goes away, they think it through, find the actual cause of the problem (symptom) and usually fix it for good. Sadly it seems there are fewer and fewer doctors like this. Seems most simply push pills and shots until something happens.

Of course humans don't have on-board diagnostics (or do we?) with ODB3 interfaces. When the individual patient doesn't fit the guidelines, it seems too often the docs toss their hands in the air and punt. Well, we tried all the drugs I have for this and it didn't work, sorry. Even worse sometimes is when it seems to work, but really only masks the underlying cause. Which doesn't go away.

One of the interesting things I read up on recently is how liver function affects cholesterol numbers. Consistently studies show that your liver is the source of most of what shows up as cholesterol, and that lifestyle choices such as diet have less impact. This is contrary to what "everybody knows" for sure. Most docs use statin drugs to get cholesterol numbers "where we want it" and may in fact be masking liver function problems. Which undetected tend to get worse, leading to liver failure. Which often does prevent the patient who dies from liver failure from developing heart disease.