Suppose you are involved in an automobile accident and in the aftermath you are required by law to submit to a drug test. The test results are sent off to a lab and they come back saying your blood alcohol level was .12. There is a b sample of your blood alcohol performed and it comes back at a sober .02. You're about to stand trial and the prosecution has a lot riding on this case so they, go against the law, have your b sample re-tested and wala, a .12 blood alcohol is returned.
That's quite hypothetical, as blood alcohol tests are very accurate. However, if the B sample re-tested at the exact same level as the A sample, then this person was drunk. People love the idea of getting off on technicalities, but tell that to the family of a person killed by a drunk driver set free over one poor lab test. Or, tell that to the guy who races clean for years and loses an Olympic medal because one or more of the medalists doped. Career is threatened, contract iffy, no sponsorship dollars, etc. Your example is what was used by Landis, chasing technicalities, but in his case, every single time they tested the samples, they got exactly the same numbers. Yet, people think he deserves to get off on a procedural technicality, ignoring the other facts.
However, in the US, many employers force drug tests on their workers and applicants.
The standard opiate test hawked by some labs is advertised as 99.9% accurate with little/no false positives, which is total BS. The false positive rate for this test is as high as 30%, and several journalists have shown they can test positive by eating a poppy seed bagel. Meanwhile, people are getting fired/not hired based on their breakfasts and branded heroin junkies.
Riis was referred to as "Mr. 60%" for years, and from experience, most rumors in the pro peloton are right on. Hematocrit above 50% is rare, even in mountain dwellers, but it can be adjusted back down to below 50% with a simple blood donation.