mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
Can anyone give a solid definition of the word sensitivity?
___________________
Smell the coffee! "Is That an Osler move??"
|
| PsychDr2B
Forum Elite
Topics: 35
Posts: 197
|
Sensitivity in screening tests refers to the proportion of actually diseased people whom correctly tested positive for the disease over all the cases of people actually with the disease, whether they tested positive or not. "Diseased people are sensitive"
|
| mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
Thx for the rock solid definition. :idea:
___________________
Smell the coffee! "Is That an Osler move??"
|
| digitaldoc
Topics: 35
Posts: 369
|
Can you further make it rock hardened by explaining the difference between sensitivity and positive predictive value ?
___________________
USMLEtoMD : MY FUN TIPS-N-TRICKS !
|
| rida
Forum Guru
Topics: 109
Posts: 721
|
Positive predictive value is the value of all the True positive out of the people who tested positive for the test. When prevalence increases, so does this value!!!
___________________
"If He takes you to it, He'll take you through it."
|
| PsychDr2B
Forum Elite
Topics: 35
Posts: 197
|
Positive Predictive Value is the probability that a person with a positive test result is a true positive actually having the disease. Which is why it is termed "predictive value of a positive test".
Sensitivity, on the other hand, is the probability that a truly diseased positive person is correctly tested positive. Which is why this can be called "true positive rate."
Summing up:
Sensitivity is TRUE POSITIVES OVER ALL PEOPLE ACTUALLY HAVING THE DISEASE (whether they tested positive or negative).
Sensitivity = TP / (TP + FN)
Positive Predictive Value is TRUE POSITIVES OVER ALL PEOPLE WHO TESTED POSITIVE FOR THE DISEASE (whether they actually have the disease or not).
PPV = TP / (TP + FP)
Here is a couple of examples to help make sense:
Given the screening test for Coronary Heart Disease below answer the following questions:
TP = 180
TN = 720
FP = 80
FN = 20
1. What is the sensitivity?
2. What should you tell the patient if his test came back positive and he asks, "How sure are you that this test means I have coronary Heart Disease?" Hint: You must calculate Positive Predictive Value.
I'll post the answer once you guys try yours here.
|
| mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
sensitivity=180/200=90%
PPV= 180/260=about 65%
___________________
Smell the coffee! "Is That an Osler move??"
|
| PsychDr2B
Forum Elite
Topics: 35
Posts: 197
|
Great! That was just warm up. So this means that this test senses the disease in 90% of people who have the disease. And You should tell the patient that he is only 69% (180/260 is closer to 69%) likely to have CHD based on his positive test result.
BONUS QUESTIONS:
What is the specificity of this test?
What is the Negative Predicitive Value?
|
| kalsam
Forum Elite
Topics: 31
Posts: 361
|
Specificity= TN/TN + FP = 720/720+80 =90%
Negative Predictive Value = TN/TN + FN = 720/720 +20 = 97%
___________________
"Read Repeat Recall Remember Recognise & Reproduce"
|
|
| |
| | | | | | | | | |
