epinephrine
Forum Newbie
Topics: 1
Posts: 1
|
Hey people please could anyone clear the funda regarding which parameter increases or decreases when you lower the range of a particular disease, eg lower the blood pressure.that means which of the follwing would be affected, sensitivity, specificity, positive predictive value, negative predictive value.
|
| Ancylostoma
Forum Guru
Topics: 42
Posts: 642
|
great example of this is diabetes . the old diagnosis for diabetes was >140 mg/dl of fasting glucose. They decreased it to 126 mg/dl . by doing that they increased the sensitivity and decreased the specificity. Remember that sensitivity is a parameter that deals with the number fo diseased . If a test has 100% sensitivity then anyone screened with the disease will come up positive. This doesnt meant that all positives have the disease but that none of the negative will have the disease. Specificity deals mainly with non diseased. If a disease has 100 % specificity, then is will detect all the nondiseased . In this case anyone who is positve , will have the disease for sure. Anyone who is negative can have the disease or may not. High sensitivty is great to find the diseased people, then high specificity is good to confirm whether they have the disease or not.
summary, decreasing the threshold for htn diagnosis will cause a increase in sensitivity, and a decrease in specificity. the point of highest sensitivity is also the point of highest negative perdictive value. The point of 100% specificity is also the point of 100% positive perdictive value. Some may have a tought time dealing with this concept but work at it. Make sure you look at the specificy and sesitive chart curves. They are frequently tested.
|
| Dinie
|
You can try this Q with explanation (Beh SCi in Medicine Fadam Pg 486)
Research shows cut off value for PSA used to identify pts with prostate CA should be lowered from 4ng/ml to 3ng/ml.
This change would,
a. increase negative predictive value
b.decrease sensitivity
c.increase false negative rate
d.increase PPV
e.increase specificity
Answer:A
reference interval change can be expected to decrease the number of false negatives and increase the number of false positives. This alteration will increase sensitivity (TP/TP+FN) and negative predictive value (TN/TN+FN) and decrease specificity (TN/TN+FP) and PPV (TP/TP+FP).
|
| Ancylostoma
Forum Guru
Topics: 42
Posts: 642
|
A good trick with these type of questions is to remember the graph with diseased and non diseased , and then remember the points of highest sensitivtiy and negative perdictive value, and highest sensitivity and positive perdictive value. You will be able to see where movement of parameters would change the calculations.
|
|
| |
| | | | |
