Calculate the Likelihood Ratio - USMLE Forums
 USMLE Forums         Your Reliable USMLE Online Community     Members     Posts
 Home USMLE Articles USMLE News USMLE Polls USMLE Books USMLE Apps
 USMLE Forums Calculate the Likelihood Ratio
 Register FAQs Members List Search Today's Posts Mark Forums Read

 USMLE Step 2 CK Forum USMLE Step 2 CK Discussion Forum: Let's talk about anything related to USMLE Step 2 CK exam

#1
08-05-2012
 USMLE Forums Master Steps History: 1+CK+CS Posts: 1,867 Threads: 149 Thanked 2,089 Times in 1,047 Posts Reputation: 2109
Calculate the Likelihood Ratio

The table below summarizes the results of a new test to identify patients with an acute myocardial infarction (AMI).
The test was performed on 400 subjects.

Acute Myocardial Infarction...... Present............ Absent................. Total

Test Result
...................Positive................ 120.................. 30....................... 150

..................Negative................. 80.................... 170 ................... 250

.......................Total............... 200..................... 200..................... 400

If one of the 400 subjects in the study is randomly chosen, what is the likelihood that an abnormal test indicates that the individual has an AMI?

A. 0.50

B. 0.60

C. 0.68

D. 0.80

E. 0.85
 The above post was thanked by: cingulate.gyrus (08-06-2012)

#2
08-05-2012
 USMLE Forums Addict Steps History: 1+CK+CS+3 Posts: 151 Threads: 48 Thanked 109 Times in 60 Posts Reputation: 119

Positive predictive value....C...0.68
#3
08-05-2012
 USMLE Forums Master Steps History: 1+CK+CS Posts: 1,867 Threads: 149 Thanked 2,089 Times in 1,047 Posts Reputation: 2109

Explanation:

The correct answer is D. The likelihood of an abnormal test indicating disease, or the positive predictive value of
a test (PV+), is expressed with the following formula:

PV+ = TP/(TP + FP)

where a true positive (TP) is an abnormal test result in an individual with disease, while a false positive (FP) is an
abnormal test result in a healthy person. Note that as the FP rate decreases, the PV+ approaches 100%. The FP
rate reflects the test's specificity, or the chance that a test is normal in a healthy individual. As the FP rate of a
test decreases, the test's specificity approaches 100%. Therefore, a test with 100% specificity always has a PV+
of 100%. This underscores why tests with high specificity are used to confirm disease, since an abnormal test is
more likely to represent a TP rather than a FP.

From the table we can see the following: TP = 120, FN = 80, TN = 170, and FP = 30. The likelihood of an
abnormal test indicating an AMI is therefore:

PV+ = 120/(120 + 30) = 0.80

Thus there is an 80% chance that an abnormal test is a TP and a 20% chance that it is a FP.

Choice A is the prevalence of AMI in this study. Prevalence refers to the number of subjects with disease (TP +
FN) divided by the total number of subjects in the study. Hence, the prevalence of an AMI in this study is (120 +
80)/(120 + 80 + 170 + 30) = 0.50, or 50%. As expected, the higher the prevalence of disease, the greater the
PV+ , since the number of individuals with disease (TPs) is increased.

Choice B is the test's sensitivity [TP/(TP + FN)], which is calculated as follows: 120/(120 + 80) = 0.60, or 60%.
The test's sensitivity is too low to be useful as a screening test for an AMI.

Choice C is the negative predictive value (PV-) = [TN/(TN + FN)], which is calculated as follows: 170/(170 + 80) =
0.68, or 68%. This percentage indicates that there is a 68% chance that a normal test is a TN and a 32% chance
that it is a FN. The higher the prevalence of disease, the lower the PV-, since a greater number of FNs are
present.

Choice E is the test's specificity [TN/(TN + FP)], which is calculated as follows: 170/(170 + 30) = 0.85, or 85%.
Although this test's specificity is higher than its sensitivity, its specificity is not high enough for it to be useful as a
confirmatory test for an AMI.
#4
08-05-2012
 USMLE Forums Addict Steps History: 1+CK+CS+3 Posts: 151 Threads: 48 Thanked 109 Times in 60 Posts Reputation: 119

seems am sleepy calculated the wrong thing...its 0.8....still PPV as i mentioned earlier....
#5
08-06-2012
 USMLE Forums Master Steps History: 1+CK+CS Posts: 720 Threads: 217 Thanked 551 Times in 308 Posts Reputation: 561

Quote:
 Originally Posted by Novobiocin The table below summarizes the results of a new test to identify patients with an acute myocardial infarction (AMI). The test was performed on 400 subjects. Acute Myocardial Infarction...... Present............ Absent................. Total Test Result ...................Positive................ 120.................. 30....................... 150 ..................Negative................. 80.................... 170 ................... 250 .......................Total............... 200..................... 200..................... 400 If one of the 400 subjects in the study is randomly chosen, what is the likelihood that an abnormal test indicates that the individual has an AMI? A. 0.50 B. 0.60 C. 0.68 D. 0.80 E. 0.85
LR+ = sens / (1-spec) = 60 / 15 = 4
 The above post was thanked by: riya rai (08-06-2012)
#6
08-06-2012
 USMLE Forums Addict Steps History: Not yet Posts: 108 Threads: 15 Thanked 22 Times in 13 Posts Reputation: 32

Quote:
 Originally Posted by Novobiocin The table below summarizes the results of a new test to identify patients with an acute myocardial infarction (AMI). The test was performed on 400 subjects. Acute Myocardial Infarction...... Present............ Absent................. Total Test Result ...................Positive................ 120.................. 30....................... 150 ..................Negative................. 80.................... 170 ................... 250 .......................Total............... 200..................... 200..................... 400 If one of the 400 subjects in the study is randomly chosen, what is the likelihood that an abnormal test indicates that the individual has an AMI? D. 0.80

2X2 Statistics / Epidemiology Table (Punnett Square)
#7
08-06-2012
 USMLE Forums Addict Steps History: Not yet Posts: 108 Threads: 15 Thanked 22 Times in 13 Posts Reputation: 32

Quote:
 Originally Posted by cingulate.gyrus LR+ = sens / (1-spec) = 60 / 15 = 4
?????? please explain this for me
#8
08-07-2012
 USMLE Forums Master Steps History: 1+CK+CS Posts: 720 Threads: 217 Thanked 551 Times in 308 Posts Reputation: 561

Quote:
 Originally Posted by riya rai ?????? please explain this for me
Sorry I was calculating The Likelihood Ratio . Seeing the word likelihood i jumped into conclusion of calculating likelihood ratio..

Sorry in this case we are supposed to calculate positive predictive value.

Likelihood ratio is the likelihood that a given test result would be expected in a patient with the target disorder compared to the likelihood that that same result would be expected in a patient without the target disorder.

Likelihood ratio for +ve result and -ve result are calculated by the formula.

LR+ = sensitivity / (1-specificity) = (a/(a+c)) / (b/(b+d))
LR- = (1-sensitivity) / specificity = (c/(a+c)) / (d/(b+d)
 The above post was thanked by: riya rai (08-07-2012)

 Tags Biostatistics-Epidemiology, Step-2-Questions

Message:
Options

## Register Now

In order to be able to post messages on the USMLE Forums forums, you must first register.
User Name:
Medical School
Choose "---" if you don't want to tell. AMG for US & Canadian medical schools. IMG for all other medical schools.
 AMG IMG ---
USMLE Steps History
What steps finished! Example: 1+CK+CS+3 = Passed Step 1, Step 2 CK, Step 2 CS, and Step 3.

Choose "---" if you don't want to tell.

 Not yet Step 1 Only CK Only CS Only 1 + CK 1 + CS 1+CK+CS CK+CS 1+CK+CS+3 ---
Favorite USMLE Books
 What USMLE books you really think are useful. Leave blank if you don't want to tell.
Location
 Where you live. Leave blank if you don't want to tell.

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.