gtag('config', 'AW-861451502'); Why Are Some Medical Tests Better Than Others?

Why Are Some Medical Tests Better Than Others?

May 24, 2020

Please note that this post is current as of the date published. 

 

There is a flurry of new tests that hit the market for COVID testing, and many already have been pulled. With the FDA having initially relaxed parameters for approval, they have had to enact stricter rules to stem the flood. But what makes some tests better than others? You may have heard terms like "Sensitivity" and "Specificity" thrown around, but are unsure what they mean. I recently was invited back on the Gut Check Podcast with Dr. Kenneth Brown and we explained these terms, and what it means to have a good test - both for positive and negative results. Click the video below to watch, or read on to learn more..... 

 

 

If you don't have the time to watch or listen to the podcast, let me try to briefly summarize. 

 

There are several terms to be familiar with:

 

Sensitivity – The percentage of patients that DO HAVE the disease that will also TEST POSITIVE

Specificity – The percentage of patients that DO NOT have the disease that will TEST NEGATIVE

 

 ****** These are the intrinsic qualities of the test *******

 

Positive Predictive Value – The percentage of positive tests that are accurately positive

Negative Predictive Value – The percentage of negative tests that are accurately negative

 

****** These relate to the accuracy of a test based on the prevalence of disease within the population  i.e. the post-test probability of having or not having the disease process*****

 

Incidence – The proportion of infected individuals or active cases

               •For COVID19, this is evaluated by using PCR, or soon by                    ANTIGEN testing

 

Prevalence – The proportion of individuals who have been (present and past tense) infected or exposed

               •For COVID19 this would be evaluated by IgG Antibodies

 

The very basic way tests are evaluated is by setting up what is called a 2 x 2 table

 

 How do you know who actually has the disease (For the top row of "disease positive and disease negative)? By comparing to a Gold Standard test. You can them use your new test and figure out from the positive and negative tests the sensitivity and specificity. Once you accomplish that, you can apply the sensitivity and specificity of the test to any population (in other words the sensitivity and specificity of the test won;t change in the disease if prevalent in 10% or 50% of your population since it is an intrinsic value of your test). 

 

 Contrast this to the Positive Predictive Value (PPV) and Negative Predictive Value (NPV) - they will change based on the disease prevalence. A disease that is MORE prevalent in a population will have a better PPV (better chance that a positive test means you actually have the disease), and a disease that has LOW prevalence will have a better NPV (so it is a better test to tell people that have a negative result that they don;t have the disease).

 

 Everything that has been mentioned to this point is generalizable to any test - Strep tests, Pregnancy tests, or COVID tests. How can you apply this information to all the new COVID testing in particular? 

 

•Many tests are currently available, all with different sensitivity and specificity

 

•You need to know the parameters of the specific test kit you are using in order to know how accurate the test is

 

•Many of the antibody tests have poor sensitivity and/or specificity and were recalled by the FDA

 

We do not yet know the true incidence/prevalence of the disease, which makes it challenging to calculate the Positive and Negative Predictive Values

 

•It appears the incidence may be differ based on geography - it stands to reason that an area like NYC will have a higher incidence and prevalence of disease than Dallas, therefore the same test may tell you different results (PPV & NPV) based on where the test was administered. A test in NYC may be better at suggesting a True Positive (better PPV) due to presumed higher prevalence and a test administered in Texas may be better at suggesting a True Negative (better NPV)

 

•This raises new concerns regarding standardization of results when baseline characteristics differ

 

New information and methods of testing are coming out every day. Our ability to test and our efficiency at testing continue to improve. This is a (inter)national conversation as we figure out together how best to live with Corona and be able to get back to school, get back to work, and enjoy all the things we like to do outside of our homes. Stay vigilant, stay informed, and stay safe. 

 

DISCLAIMER: Please note that this blog is intended for Informational Use only and is not intended to replace personal evaluation and treatment by a medical provider. The information provided on this website is not intended as substitute for medical advice or treatment. Please consult your doctor for any information related to your personal care. 

 

Please reload

Recent Posts

Please reload

Archive

Please reload

Tags