Medical tests are characterized by validity and utility:

  • Validity is the probability that the test results are correct.
  • Utility is the usefulness of having correct results (e.g. how much life-time can be gained by taking a test, finding out that there is a disease, and treating it in time).

I am looking for tables which show the validity and utility (and, maybe, also the cost) of some commonly used clinical tests.

I don't need them for clinical use; I only intend to bring some examples to illustrate a theoretical paper about diagnostics. So, the tables don't have to be very accurate or up-to-date.

I found several research papers that try to assess the validity and utility of novel tests, e.g. some genetic or molecular tests for cancer, but they are too difficult for me because I don't have sufficient background in genetics or molecular biology. I need a simpler and approximate table.


NOTE: I asked this question in opendata.SE and got no reply. I hope this site is a better place for it.

| |
  • 2
    Hi there! I think this may not be the right place because it doesn't seem to be about health, but rather discussion on the cost of tests. Unfortunately, opendata.SE would probably still be a better place for your question. Did you consider putting a bounty to draw more attention to your question? Best of luck! – Dave Liu Jun 3 '15 at 18:32
  • 1
    Statistics courses aimed at biologists or medics will frequently use examples from real studies when introducing ROC curves. You could probably look through textbooks and courses (maybe look at MOOC platforms?) and thus find out the validity of some tests. Utility may be mentioned there, or more likely in papers which introduced or reviewed the tests (which should be easy to find once you have leads). The cost will have to come from somewhere else, though. Maybe from public information on how much an insurance have to pay the doctors for the test. – rumtscho Jun 3 '15 at 18:44
  • 1
    Rather than validity, I think the technical terms are sensitivity(how many false negatives) and specificity(how many false positives). Try searching for breast cancer screening or HPV screening tests. – jiggunjer Jun 4 '15 at 9:24
  • 2
    Would The Validity and Utility of Selection Methods in Personnel Psychology help? It's old and it's about HR psychology as I can see, but there is a table with quantitative values for validity... – Lucky Jun 5 '15 at 2:52
  • 1
    I accidentally came across a paper discussing tests for one specific condition, it is of course not as convenient as having a table, but at least it is one example. ncbi.nlm.nih.gov/pubmed/20828509 – rumtscho Jun 7 '15 at 19:11