Comment by adevine | Hacker News Reader

Comment by adevine | Hacker News Reader

Parent Full thread

Agreed. I thought this quote from the article was pretty lame:

> Because once the patient has those test results, more likely than not, they're going to need a doctor's opinion to decipher them and to know what those results mean. So why not keep the doctor involved from the beginning so that, first of all, there is a logic for getting the blood test, and then there's an expert opinion there when the test results arrive to help you interpret them.

I call BS on that. Many, if not most, tests would NOT need any doctor involved if the test came back normal, which in the majority of cases, it does. Heck, when I go to my doctor NOW and have to get a test, if everything is normal I just get a phone call: "Yep, everything was normal." Explain again why a doctor needs to be involved in that process?

As another example, look at STD tests. Anyone can walk into one of thousands of clinics around the country to get a suite of STD tests, no doctor required, and only if a test comes back positive would you need to see a doctor. So why is this so different from other types of tests?

If you got every blood test done you would no question get several false positives. If 1,000,000 people got every blood test possible they would on average have worse Heath outcomes than a control ground who got zero tests. This is who of he reasons mass screenings for conditions are often a bad idea.

Subsequent screenings can reduce the false positive rate. There's no reason to accept a single measurement as gospel.

The problem with false positives is they often show up because your body chemistry is X not because the test failed. So, subsequent screenings are often more invasive and involve risks. (radiation, being cut open etc.)

Blood tests vary from reasonably simple to extremely complex, extremely reliable to extremely variable and requiring no further action (as in your case) to requiring multiple possible further decisions.

Understanding the results of multiple medical blood tests requires some knowledge of multiple subjects including biochemistry, chemical pathology, microbiology and immunology.

There's also the question of what to do next - do you repeat the test, get additional blood tests, get other tests (such as radiology or biopsies) and most critically do you actually have a diagnosis and do you need treatment? (If so, what?)

Then there's the issue of false positives and false negatives which most people get wrong (especially if you're in a low prevalence population).

The doctor needs to be involved because there's no guarantee that things will be normal.

Also depending on the test you may or may not understand what "normal" means.

Here's a quick test - These are some of the most commonly ordered blood tests: Which of these tests would you be happy to interpret yourself?:

- CBC (Including HB, platelets, differential blood count, neutrophils, leucocytes)

- Urea and electrolytes (Sodium, potassium, urea, creatinine, U/Cr ratio, chloride, bicarbonate, Anion gap)

- Liver function tests (ALP, AST, ALT, CBr, UBr) If abnormal - Hepatitis serology (including HBsAg, anti-HBc, IgM anti-HBc, anti-HBs)

- CMP (Free calcium, bound calcium, protein, magnesium, phosphate)

Reply on news.ycombinator.com