« Follow up questions, Group 3 -Lecture 3 (Jan 29, 2008) | Main | Paper Presentation Assignments »

follow up questions, Group 4 -Lecture 3 (Jan 29, 2008)

Group 4 - Example 16.7.4

This example is similar to those found when validating new clinical tests. A quick and less quantitative screening method identifies those clinical samples testing positive. Then, these samples are analyzed using a quantitative method. Considering that a person’s life may be at risk, what is more important: sensitivity, selectivity, or efficiency?

TrackBack

TrackBack URL for this entry:
https://blog.lib.umn.edu/cgi-bin/mt-tb.cgi/61307

Comments

I feel as if the efficiency shouldn't matter at all; efficiency is the percent correct results and since the first assay has already found the correct results the second, quantitative assay need not stress that aspect. The selectivity isn't as important in this case either as the first assay determined the positive samples already. In my opinion, sensitivity is the most important aspect as the detection of the exact amount of analyte is needed. To be able to determine an accurate amount of analyte present with a small LOD is the most important problem to address in this assay.

If time is of concern (i.e. results are needed quickly), I think that sensitivity is the most important trait of an assay because a quick detection of a drug in an over dose situation will lead to a quick medical response. However, if the person is not in critical condition, I believe that efficiency is most important. That is, if there is time, it is really important to have an accurate/correct assessment of the person’s conditions in order to properly treat that person. In any case, specificity is of least concern.

I would say that the sensitivity is more important considering that a person's life might be at risk. Becuase the more sensitive the method is, the more easily people with abuse drugs or other banned substances will be found out.

A person's life is in risk doesn't mean we can increase the chance to let him/her be false positive, since we don't want to treat people medicine if they don't have that kind of disease.
If an assay is highly sensitive, it will exclude many false negative while include some false positive; if an assay is highly selective, it will exclude many false positive while include some false negative. Considering a man's life is in risk, which means we should exclude false positive or false negative to give exactly right diagnostic result, thus efficiency is the most important factor. High efficiency means less mis-diagnostic result.

I think that there are really two questions at play here. What do you want your initial test to be and what should the second follow up test be. I think that in the first test you want something that is highly efficient, because as people have said before you want to get all true positives even if that means including some false positive. This is so you can start people on treatment immediately, and in general it is better to start people who do not need the treatment then miss people who do need the treatment. In the second test this is where you want it to be selective so that you can separate the false positives from the true positives, and then to make sure that people are not getting treatments that they do not need and could have harmful side effects.

I think the efficiency is most important. Actually, the quantitative assay is more accurate and can give a final result. Since a person's life is at risk, it is important that the first assay shows the right result which means the same result with the second assay. So in my opinion, the true positive and true negative results are what a good assay should give. So efficiency is the most important one.

It sort of depends on the intial quantitative methods chosen. I agree with Jon and Melissa, as you would not care so much about sensitivity initially, as you are dealing with a person's life. Specificity and efficiency matter most, and from a malpractice standpoint, the person or person's family will have less of a problem being treated prematurely than not at all. However, if you are considering false positives, efficiency does not matter, and selectivity is the chief consideration.

If a person's life was at risk, what's more important? In my opinion, the efficiency is most important. Efficiency means the percent of correct results. At this crucial time, correct is most important. We cannot treat patients wrong. We need to distinguish who really get that disease and who don't get it correctly and quickly. If we missed the people who get that disease, the risk that people take would increases. But if some people were at risk and the disease they take was not the disease we treat, it's really a terrible thing.The risk increases more. So I think efficiency is most important.

When someone's life is on the line, the only that matters is getting the answer right. This falls right in line with the definition of efficiency. Obviously a highly sensitive test will help increase the efficiency so those two elements seem to be closely tied.

If the purpose of this assay is to rescue a person’s life, I definitely agree with Melissa, Meng, Jon and Yixiao. Because you want to save a life, 100% efficieny would be the ideal assay for treatment. That means you can get 100 % correct results. As Yixiao mentioned above, doctors do not want give their patients wrong treatment. But I think there is no 100 % efficiency assay. So in addition to using high efficiency assay in the first test, we can further combine with other high sensitivity or specificity assays to increase the diagnostic accuracy.

SUMMARY AND COMMENTS

Several points were mentioned or implied when providing an opinion: (1) How critical is the patient's condition, (2) how nasty are the side effects, (3) how quickly a secondary method (quantitative) could be run.

An important point to consider is that assays are rarely 100% efficient. As sensitivity increases, selectivity decreases, and vice versa

Some possible situations are:

1. Critical condition, no side effects, delay in using a confirmatory method. I would select a method with HIGH SENSITIVITY.

2. Critical condition, nasty side effects, delay in using a confirmatory method. I would select a method with HIGH EFFICIENCY.

3. Not so critical condition, nasty side effects, delay in using a confirmatory method. I would select a method with HIGH SELECTIVITY.

4. If affordable,I would run several qualitative method simultaneously...

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

The views and opinions expressed in this page are strictly those of the page author. The contents of this page have not been reviewed or approved by the University of Minnesota.