Tag Archives: pre-test probability

“Testing ad infinitum”

Take for example the patient who presents with “recent weight loss”, even though their recorded weight is exactly the same as it was 6 months ago.

…or the patient who “bounds” into the surgery, looking far healthier than the doctor will ever be, and then complains of “tiredness”.

…or the patient who has diagnosed themselves with “Chronic Lyme Disease” on the internet, even though they have never travelled to an area endemic for Lyme disease.

The temptation for the clinician in such cases above is to order a whole battery of tests in order to prove to, or reassure the patient that they have no organic pathology.

The patient then leaves the clinic with a whole list of (often expensive) laboratory investigations, and thinks to themselves;

“Wow, look at how many tests I am getting. The doctor must be worried for me. I really must be sick!”

And thus the cycle goes on. The tests come back negative, but the sick role is now reinforced. The patient then often comes back for more, or goes off elsewhere to seek a second opinion…

Worse still, if enough tests are performed, then one will eventually come back falsely equivocal or positive, confusing the issue even further for the clinician. And the positive reinforcement of the sick role in the patient has just gone through the roof!

Everyone is scared of missing something, of not diagnosing that long shot… But sometimes it is best just to trust good clinical acumen, and appreciate that laboratory testing can occasionally cause more harm than good…


“Dead Certs and Long Shots”

The more uncertain the result will be, the more useful the laboratory test generally is…

Sounds a little paradoxical, but it is absolutely true.

If we are looking to confirm something that is almost certain before the lab test is performed, then we need a “super-sensitive” test to fulfil this task. Otherwise we run the risk of giving false negative results.

For example, if we have a teenager with a sore throat and lymphadenopathy, a lymphocytosis and atypical lymphocytes on blood film, then the probability of this being EBV infection is about 90%. There is little point then in doing a confirmatory Monospot test with a sensitivity of 80-85%. This will only lead to giving negative results on patients who actually have EBV infection.

And if we are looking to diagnose a long shot (aka a very unlikely diagnosis) then we had better be sure our laboratory test is “super-specific”, otherwise we will run the risk of giving false positive results.

For example if we want to diagnose dengue fever in a patient with “flu like” symptoms returning from Mexico (an area of relatively low Dengue endemicity), then we need to think twice about performing Dengue serology testing which has a specificity rate of about 95%. You are just as likely to report a positive test in someone who doesn’t actually have Dengue.

What we are doing in actual practice here is taking our pre-test probability, and using it to give a prevalence rate (by proxy) in our tested population. Once we know this, then we can use our test sensitivity and specificity to calculate positive and negative predictive values, not always with the results we would like…

Laboratory specialists tend to be more aware of testing limitations such as these. Clinicians, in general,  tend to just take the laboratory results as gospel.

But I believe it is ultimately the laboratory’s responsibility to stress the limitations of using laboratory testing for “Dead Certs or Long Shots”, and either prevent such testing taking place, or put big disclaimers on the results.



“Measuring the odds….”

“Medicine is a science of uncertainty and an art of probability.” William Osler

This article is an extension of a previous article on sensitivity, specificity, and positive predictive value, this time looking at pre- and post-test probability.

Take the following hypothetical scenario:

A 15 yr old teenager and an 85 yr old Rest Home resident both present to their doctor with a sore throat and cervical lymphadenopathy.

An Epstein Barr Virus (EBV) screening test (with sensitivity and specificity both at 97%) is positive in both patients.

Without knowing any other information, what is the likelihood of each person having Infectious Mononucleosis/Glandular Fever due to EBV?

Let’s look at it before the EBV test is performed. The prevalence of glandular fever in 15 year olds with sore throats is many times greater than in 85 year olds. Thus the chance of the 15 yr old having glandular fever is much, much higher than the 85 yr old. (This is called pre-test probability).

After the test result is known, the post-test probability of glandular fever is extremely high in the teenager, but still relatively low in the 85 yr old, as the very low prevalence in this age cohort will lower the positive predictive value of the test. Therefore the chances of the EBV result being a false positive in the 85 year old is relatively high.

In conclusion, exactly the same result in different patients needs to be interpreted differently.

So what is the lesson from this?

I suspect that a lot of laboratory users don’t really think enough about pre and post-test probability when they see the laboratory result. They may well take the result at face value and diagnose the patient on the basis of it. (I have seen this happen many many times…)

It is our job in the laboratory to convince the requestors that on the basis of the factors described above, the results are not always perfect, however much we would like them to be……


p.s. I have added a quick powerpoint on the basics of Norovirus Infection to the website.