policy for established norms and ranges used to interpret and report clinical la!
Question:
Answers:
Every medical laboratory is to some extent unique. This lab uses that reagent supplier for blood tests, and some other supplier for tissue tests, and so on. Another lab uses a different collection of suppliers. And, different labs use different procedures to some extent (eg, different workloads of facility layouts). And there are often several different equipment vendors, whose equipment works slightly differently, using different operational sequences and check points.
So, the very same sample (at least taken from a single larger original sample anyway) will give slightly different results in every lab for most any particular test. In order to lessen the chance of confusion, labs report a 'norm' value for many tests to tell those interpreting the test that our values run <a little higher than the literature reports.> or something similar. They also usually supply a 'normal range' for each test as a courtesy to whoever has to look at the reports; people vary quite a lot, and this value might be acceptable in you but not in someone else quite apart from any laboratory norm differences from other labs.
Since the values in clinical lab reports are often used to decide questions of diagnosis and treatment, it's pretty important (actually vital) to be certain the lab's results are as claimed. And certainly that any errors which can be caught are actually identified. There is a small industry of calibrated sample suppliers and operational testing to ensure this.
==========
blood glucose testing example
Take, for instance, the home blood glucose monitors used by many diabetics. These generally come with a small bottle of test solution (or control solution, depending on who is choosing names). This is carefully concocted 'fake blood' with a pre-determined concentration of glucose. By using it (if it hasn't partially evaporated or deteriorated), one can test the operation of the glucose meter/test strip combination, qualitatively -- and to some extent quantitatively. A similar dodge is used in actual laboratories. To avoid 'fudging' in the labs themselves (a kind of Caesar's wife's reputation sort of thing), test reagents are generally not made locally, but rather in a central facility.
======
regulation, licensing, accreditation
Labs are also supervised by industry accrediting bodies rather like colleges, and almost always have statutory and regulatory supervision (mandatory licensing, for example).
======
malfeasance
Despite all of this, it's still possible for mendacious lab operators to perform poorly. Especially in circumstances which require some subjective evaluation. A recent example in the US, though not exactly a laboratory issue, is the exposure of what seems to have been a litigation fraud involving lung diseases caused by mining or industrial operations. Attorneys have advertised for plaintiffs, provided testing for them, and organized suits for damages based on the results. The tests (X-rays and the interpretation thereof, primarily) were performed by independent contracting physicians and X-ray labs, but it seems that several folks noticed that business depended on the findings. There were an awfully large number of folks found to have lung disease who may not have had it all, or at least not the disease for which damages were being sought. Just who arranged for it all, beyond a small number of doctors who made quick diagnoses from medical files, usually without actually meeting patients, is not clear. The lawyers certainly benefited in that they found clients for whom ot work. The patients benefited, though often not very much as individual settlements were often not large. As the situation is clarified, we may learn more as to who organized it, if there is indeed a single organizing brain.
Psychological testing is not very lab based, but the same (or something similar) kind of problem has been noticed in exert evaluations there as well. There is even a Texas psychiatrist (Dr Grigson) who earned the nickname Dr Death for his consistent testimony that the defendant did not qualify for the legal classification of 'not responsible for his actions', which might preclude conviction or certain sentences (death, mostly, hence the nickname) if established. He was exceptionally popular with prosecutors, so much so that his career lasted 40 years in the Texas courts. And he was instrumental in convicting a very large number of defendants, and winning harsher sentences for many.
=============
crime labs
Perhaps more important (if it were possible) is crime lab operation. There have been several cases in recent years, again in the US, in which crime labs have been alleged (in the case of the FBI in the mid 90s and again in 2003) to have been sloppy or pressured to produce certain results (even resulting in false testimony in court some cases), or shown (in the case of several
state crime labs) to have falsified results. When discovered, these situations often lead to the reopening of many cases in which a defendant was convicted and sentenced on evidence from the lab (eg, DNA evidence).
It is not the case, as portrayed in assorted TV shows, that lab results are definitive.
*First, there may simply be mistakes (eg, differing mormal values from lab to lab not properly taken into account).
*Second, the crime lab might be more interested in satisfying its client (ie, police and prosecutors) than in 'simple' reporting of results without regard to their consequences for the case under investigation.
*Third, for various reasons including incompetence or ego gratification, the lab (or a worker there) may claim more certainly or significance for the results reported than they deserve. On a different level, incompetence in the lab can ruin samples (eg, contamination), produce wrong results, etc. Results reported under these circumstances are, of course, not worth relying on.
Answers: