Share Facebook
Crohn's Disease Forum » Forum Wiki » Diagnostic Test Sensitivity and Specificity


Diagnostic Test Sensitivity and Specificity

Contents


About Diagnostic Test Sensitivity and Specificity

Diagnostic Tests can vary widely in how trustworthy the test results are. Laboratory testing facilities typically publish how reliable their protocols are. This Wiki page explains the terms used to describe the Accuracy for specific Laboratory Tests.

Sensitivity

The Sensitivity of a Diagnostic Test is the Probability that the test will Report a Positive Test Result when a person should obtain a Positive Test Result = aka. TRUE POSITIVE

High Sensitivity = Good Diagnostic Test

For example, I have Crohn's Disease. A Highly Sensitive Test for Crohn's Disease will give me a Positive Test Result showing I have Crohn's Disease.

Test Result = You HAVE Crohn'sTest Result = You DO NOT have Crohn's
I HAVE Crohn's DiseaseTest Accurately Reports - You HAVE Crohn'sTest misreports you have Crohn's
I DO NOT have Crohn's DiseaseTest misreports you have Crohn'sTest Accurately Reports you do not have Crohn's

Specificity

The Specificity of a Diagnostic Test is the Probability that the test will Report a Negative Test Result when a person should obtain a Negative Test Result = aka. TRUE NEGATIVE

High Specificity = Good Diagnostic Test

For example, I do NOT have Ulcerative Colitis. A Highly Specific Test for Ulcerative Colitis will give me a Negative Test Result showing I do not have Ulcerative Colitis.

Test Result = You HAVE Ulcerative ColitisTest Result = You DO NOT have Ulcerative Colitis
I HAVE Ulcerative ColitisTest Accurately Reports - You HAVE Ulcerative ColitisTest misreports you have Ulcerative Colitis
I DO NOT have Ulcerative ColitisTest misreports you have Ulcerative ColitisTest Accurately Reports you do not have Ulcerative Colitis

False Positive

A False Positive Diagnostic Test Result Reports a Positive Test Result when a person should obtain a Negative Test Result.

High False Positive Rate = Bad Diagnostic Test

For example, I do NOT have Behcet's Disease. A False Positive Test Result will give me a Positive Result showing I have Behcet's Disease.

Test Result = You HAVE Behcet'sTest Result = You DO NOT have Behcet's
I HAVE Behcet's DiseaseTest Accurately Reports - You HAVE Behcet'sTest misreports you do not have Behcet's
I DO NOT have Behcet's DiseaseTest Misreports you HAVE Behcet's DiseaseTest Accurately reports you do not have Behcet's

False Negative

A False Negative Diagnostic Test Result Reports a Negative Test Result when a person should obtain a Positive Result.

High False Negative Rate = Bad Diagnostic Test

For example, I HAVE Clostridium difficile (C. diff) Infection. A False Negative Test Result will give me a Negative Test Result when I actually do have a Clostridium difficile Infection.

Test Result = You HAVE Clostridium difficileTest Result = You DO NOT have Clostridium difficile
I HAVE C. difficileTest Accurately Reports - You HAVE C. difficileTest misreports you do NOT have C. difficile
I DO NOT have C. difficileTest Misreports you HAVE C. difficileTest Accurately reports you do not have C. difficile

Positive Predictive Value (PPV)

The Positive Predictive Value (PPV) of a Test takes into account how many people in a population have the disease or condition in question (Disease Prevalence). Whereas Sensitivity and Specificity are a function of how good a particular Test is, the PPV goes a step further and looks at the probability a person has the disease.

PPV = (True Positive) / [(True Positive)+(False Positive)] X 100

Negative Predictive Value (NPV)

The Negative Predictive Value (NPV) of a Test takes into account how many people in a population have the disease or condition in question (Disease Prevalence). Whereas Sensitivity and Specificity are a function of how good a particular Test is, the NPV goes a step further and looks at the population based probability a person does NOT have the disease.

NPV = (True Negative) / [(True Negative)+(False Negative)] X 100

References

1. Sensitivity and Specificity. 1999. North Carolina School of Science and Mathematics: Statistics Leadership Institute. http://courses.ncssm.edu/math/Stat_I...pecificity.pdf

2. https://onlinecourses.science.psu.edu/stat507/node/71

Popular Threads Discussing Diagnostic Test Sensitivity and Specificity



Page Tools
No one has commented on this article.

Reply
Page Tools
Search this Page


All times are GMT -5. The time now is 03:03 PM.
Copyright 2006-2017 Crohnsforum.com