Diagnostic testing unable to achieve 100 percent accuracy

Veterinarians often rely on clinical and laboratory tests to diagnose disease in livestock species.

Diagnostic tests can be as simple as listening to a rumen with a stethoscope or as complicated as a laboratory procedure that isolates and identifies a particular virus or bacteria or measures a specific enzyme in the blood of an animal.

In many situations, these diagnostic tests are an essential part of arriving at a more conclusive diagnosis or assisting with the development of a treatment plan.

However, it is important to recognize that most diagnostic tests are not perfect tools. Test results can be misleading and in some situations may actually give the wrong answer.

Let’s look at a common example in one diagnostic procedure that we commonly use in cattle.

The diagnostic test for tritrichomonas foetus (trich) in cattle is a relatively straightforward procedure. This microscopic-sized, single-celled parasite resides in the reproductive tract of cows and in the prepuce or sheath area of bulls. It is a venereal disease that is transmitted from bulls to cows and from cows to bulls by mating.

When cows become infected, the parasite can cause early embryonic death and subsequent infertility in the cow herd.

Veterinarians are often asked to test bulls for this parasite to make sure that we don’t bring an infected bull onto a breeding pasture.

The collection method is fairly simple: take a sample by scraping the inside of the bull’s sheath with a pipette and then culture this sample in an incubator for several days in a special culture media designed for the parasite.

The veterinarian or laboratory technician can then examine this sample under a microscope and look for the small, single-celled organism, which has typically multiplied in the culture media and now can be seen moving throughout the sample.

However, it turns out that this parasite is tricky to grow, and in spite of doing everything correctly, we may be able to isolate the organism from infected bulls only 80 to 90 percent of the time.

If I had a corral full of 100 bulls that were all infected with trich, I would identify only 85 of them as positive with this diagnostic test if I tested every bull. This is sometimes called the “sensitivity” of the diagnostic test.

The test’s sensitivity is defined as how well it is able to identify diseased individuals. In this case, the test for trichomoniasis would have a sensitivity of about 85 percent.

It also means that there is a 15 percent chance that we could incorrectly classify an infected bull as negative when using this test.

Veterinarians work around this potential diagnostic error by doing multiple tests on a bull to make sure it is negative.

The requirement to have three negative trich tests is designed to get around the problem of imperfect sensitivity of the diagnostic test.

We are sometimes asked to try to culture cows for this parasite, but the sensitivity is much lower and is probably less than 50 percent. We have so many false negatives that it really isn’t even worth the effort.

Therefore, all of the diagnostic efforts for trich are usually focused on the bull, where the test is a little more reliable.

It turns out that it is possible to make a mistake in the diagnostic process and also have a false positive diagnosis. Other parasites that normally reside in the feces can sometimes be mistaken for trich if manure contaminates the culture media.

A veterinarian or laboratory technician might mistake them for trich and call the bull positive when it is really negative.

It turns out that this diagnostic error is much more uncommon, probably occurring in less than one percent of uninfected bulls.

This diagnostic test characteristic is called “specificity” and refers to how well the test correctly diagnoses uninfected animals. In this case, the test for trich would have a specificity of 99 percent.

The diagnostic test for trich, with a sensitivity of 85 percent and a specificity of 99 percent, is quite a good one.

Some diseases, such as Johne’s disease in cattle, are true diagnostic challenges and our best laboratory tests barely approach a sensitivity of 50 percent.

There are lots of false negative tests with this disease and it is difficult to screen herds effectively because of the poor sensitivity of the diagnostic tests.

Sensitivity and specificity are important characteristics of all diagnostic tests. Every diagnostic test, from the clinical examination to the laboratory test, has a possibility of either a false positive or a false negative result.

These diagnostic tests are important tools for veterinarians and pro-ducers, but veterinarians are always trying to interpret the results, keeping in mind the characteristics and probabilities of sensitivity and specificity.

Like many things in life, although we would like to have a clear cut answer to questions, we are still forced to deal with some uncertainty.

John Campbell is head of Large Animal Clinical Sciences at the University of Saskatchewan’s Western College of Veterinary Medicine.

One Response

Post a response
  1. If you are referring to the level of sensitivity of elisa tet for JD you are correct. However this is not the case for qPCR which has a level of specificity and sensitivity close to 90%
    The only way to determine if an animal is infected with MAP is to test with molecular assay. When will people finally realize it?!!!!


You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>