When I was born (OK, I’ll admit it, it was over a half-century ago), there were about 250 different laboratory tests available for physicians to order. Now, there are probably over 4,000. Certainly, in the intervening years, there have been many new diseases that required new tests for diagnosis, and many new drugs that needed new assays for monitoring therapy. But how many of these 4,000 tests belong in Jurassic Lab, either because they really are “dinosaurs” or because they really are useless? Given the huge problem of the cost of health care delivery, could we help save some money by eliminating tests that are really no longer needed?
Three years ago, the editors of CLN asked a number of laboratorians to nominate unnecessary tests. Five were mentioned by the majority of the respondents: the LE cell test; the Schilling test; free thyroxine index; prostatic acid phosphatase; and Bence Jones protein. Eliminating these would probably have little impact on the laboratory’s contribution to health care cost, because they are probably not very frequently ordered. It is similar to the current federal budget battle, where the bulk of the cuts may fall on a small portion of the overall budget because everyone’s afraid to tackle Medicare and the Department of Defense. If we really want to reduce unnecessary tests, let’s discuss whether we really need ANA screening rather than worry about the LE cell test. Forget about prostatic acid phosphatase; is continuing to offer PSA worthwhile?
Many new tests are probably much better than old ones. Consider anti-CCP vs. rheumatoid factor, or troponin vs. CK-MB. But do the old ones disappear or do they march on like zombie tests, still ordered by some physicians and still offered by laboratories afraid to say goodbye? At Mayo Clinic, they eliminated CK-MB in 2008 (see the discussion by Amy Saenger and Allan Jaffe in Circulation 118:2200-2206 titled, “Death of a Heavyweight”). Why haven’t we all done so? Some new tests are probably no better than the old ones. And yet, the new tests are often implemented anyway. Do we need assays for antibodies to deamidated gliadin peptide when anti-tissue transglutaminase antibody assays are already extremely sensitive and specific for celiac disease? Do we really need two types of natriuretic peptide assays?
I know that there are many aspects to the large expansion of laboratory tests available but wonder if we laboratorians fully understand our role in weeding out the tests that need to go. It is probably not just cost that is a factor. Consider cystatin C vs. creatinine. If cystatin C is a much better marker of glomerular filtration rate, why has it not replaced creatinine? It is easy to say that the reason is cost and convenience. Yet, if cystatin C were as cheap and as widely available as creatinine, how many people think that creatinine would disappear? We need a new paradigm. If a new laboratory test offers no advantages over an existing test, we should treat it like Charley the tuna and say, “Sorry!” But, if it is superior to an existing test, we should replace the existing one. What are your suggestions for candidates for Jurassic Lab?