|Prep for USMLE|
|         Forum      |     Resources||New Posts   |   Register   |   Login||»  |
I am in the process of making video lectures for epidemiology and biostatistics. They are animation based to aid in visual learning and they finish with a couple practice questions. I just uploaded a lecture about sensitivity onto youtube and would appreciate any advice or criticism. You can find the video by searching Sensitivity Epistats on youtube, or by going to the link below...
Thank you in advance
Edited by Kruno on Feb 23, 2016 - 7:51 AM
Here is a transcript of the video... though the animations do help.
Sensitivity is the ability of a test to detect a condition. The sensitivity of a test is determined by the percentage of patients with a condition that test positive for it. Or in other words, the percentage of cases that are detected.
The higher sensitivity of the test, then the lower the percentage of patients with the condition that will test negative. A low false negative rate is a characteristic of highly sensitive tests, which makes it useful for screening purposes. This makes sense, because you do not want a screening test to incorrectly tell patients that they do not have the condition when in fact they do. When a test is very sensitive, a negative test result means that presence of the condition is very unlikely and therefore can be confidently ruled out.
A patient presents with signs and symptoms suggestive of primary syphilis. The doctor orders a rapid plasma reagin test, but it comes back negative. Knowing that the test is very sensitive and has a low false negative rate, the doctor realizes that the patient probably does not have syphilis. Further testing indeed shows that the patient had another condition.
This is an example of the ability of a very sensitive test to rule out a disease based on its very low false negative rate. Another example would be the ability of a negative d-dimer test to rule out a deep venous thrombosis.
While a sensitive test is good at detecting a condition, it may not be good at identifying individuals without the condition. That is, some individuals without the condition may test positive as well. These cases are referred to as false positives. A patient without the condition who tests negative is considered a true negative. So although a negative test result suggests the absence of the condition, a positive test result does not necessarily mean that the condition is present. Therefore, if a highly sensitive test comes back positive the patient should undergo further testing in order to determine whether the test was correctly or falsely positive.
A woman is eight weeks pregnant and undergoes routine screening. Rather unexpectedly, the rapid plasma reagin test comes back positive. Knowing that screening tests are better at ruling out a condition than they are at ruling in, her doctor decides to order a more specific test to confirm. When the FTA-ABS test comes back negative, the doctor considers the possibility that the initial test was a false positive. Further testing shows that the woman did not have syphilis and that the cause of the false positive was anti-phospholipid syndrome.
This is an example of a false positive that can occur with the rapid plasma reagin test. Other tests that more commonly produce false positives include the fecal occult blood test when screening for colon cancer, electrocardiography when looking for left ventricular enlargement, and the ELISA test for detecting Lyme disease.
Sensitivity is calculated by dividing the total number of patients that have the condition and have a positive test by the total number of patients that have the condition. When using a sample table, sensitivity can be calculated by dividing the number of patients in Box A (condition present, test positive) by the number of patients in Box A + C (total condition present).
The video follows with a couple of questions and explanations.
Edited by Kruno on Feb 23, 2016 - 7:23 AM
I also helped make an app, called EpiStats. It has a calculator, punnett square flashcards, and more. Currently, its only on android. If you're interested you can check it out on the google play store.
| Similar forum topics|
Epidemiology & Biostatistics
UW Biostatistics and Epidemiology
Epidemiology and biostatistics
| Related resources|
NMS Clinical Epidemiology and Biostatistics
Kaplan Lecture Notes: Psychiatry and Epidemiology
Epidemiology, Biostatistics & Preventive Medicine