Sponsored Links
Results 1 to 9 of 9
  1. #1
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Biostatistics Question

    Advertisements



    Eric King presents to the clinic with a new rash on his face. The rash itself does not cause him any particular discomfort, but he has been feeling rather unwell lately. When asked to elaborate he complains of nonspecific aches and pains, fever, fatigue, and headaches. After obtaining the relevant information from the history and physical examination, Dr. Ramirez suspects systemic lupus erythematosus and counsels her patient that ordering an antinuclear antibody test would be appropriate. Mr. King consents, but admits that he is a little confused about something. What he would like to know is why people sometimes say that a negative result is a good thing? Dr. Ramirez explains that if an investigation is very sensitive, then a negative test simply means that the patient probably does not have the condition. In a hypothetical study, 413 out of 415 positive tests and 169 out of 184 negative tests were correct. Based on this information, what proportion of people with systemic lupus erythematosus would Dr. Ramirez tell her patient are correctly identified by the test?

  2. #2
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    a. 169/184
    b. 413/428
    c. 413/415
    d. 169/171
    e. 169/415

  3. #3
    mayyo is offline Newbie 510 points
    Join Date
    Feb 2016
    Posts
    5
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    so what the question asking for PPV? Right
    If we have TP is 413 of people with the SLE then FN is 2. TN is 169 of people who dont have SLE, Then FP is 15.
    so PPV would be 413/428. Of course it is better when you draw 2/2 table but please correct me if I am wrong!!
    Thanks

  4. #4
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    The question asks for the sensitivity of the test, which is the proportion of patients WITH the condition that are correctly identified (i.e., test positive). PPV refers to the proportion of patients with a positive test that have the condition. The denominator in the two values is different. In sensitivity it is the number of patients with the condition (a+c), whereas in PPV it is the number of patients with a positive test (a+b).
    Last edited by Kruno; 02-26-2016 at 03:20 AM. Reason: Incomplete, grammar

  5. #5
    mayyo is offline Newbie 510 points
    Join Date
    Feb 2016
    Posts
    5
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    Thank you for clarification. Would the answer be 413/415 for sensitivity?

  6. #6
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    No, that would be PPV.

    413 is the number of patients with the condition and a positive test, while 415 is the total number of patients with a positive test.

    To solve the question you need to determine how many patients in total have the condition.

  7. #7
    mayyo is offline Newbie 510 points
    Join Date
    Feb 2016
    Posts
    5
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    sensitivity equal TP/TP+FN. TP is 413. and FN is what is left 415-413 in the presence of disease. therefore the answer should be 413/415.
    415 is the total numbers of patient with the disease not only with postive test!

  8. #8
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    That is the correct equation for sensitivity, but 415 is the total number of people with a positive test (true positives plus false positives). As you mentioned, 413 is the number of true positives (patients with the condition and a positive test).

    To calculate the number of false negatives you have to subtract the total number of patients with a negative test by the number of true negatives (patients without the condition who tested negative).
    Last edited by Kruno; 02-26-2016 at 04:28 AM. Reason: Clarification

  9. #9
    Kruno's Avatar
    Kruno is offline Junior Member 510 points
    Join Date
    Feb 2016
    Posts
    21
    Downloads
    0
    Uploads
    0
    Thanks
    0
    Thanked 0 Times in 0 Posts
    You got it all right the first time, you just mixed up PPV with sensitivity.

Similar Threads

  1. Biostatistics
    By elizibith11 in forum Medical University of Lublin
    Replies: 1
    Last Post: 10-22-2011, 04:36 PM
  2. Biostatistics
    By reshamu in forum AUC Medical School Classifieds
    Replies: 1
    Last Post: 01-14-2011, 02:48 PM
  3. biostatistics
    By felicia in forum USMLE Step 1 Forum
    Replies: 0
    Last Post: 12-30-2006, 11:05 AM
  4. Biostatistics
    By ob4life in forum St. Matthews University School of Medicine
    Replies: 9
    Last Post: 02-03-2006, 06:55 AM
  5. Biostatistics
    By Anonymous in forum USMLE STEP 3 Forum
    Replies: 1
    Last Post: 10-24-2004, 07:11 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •