Tanta Innovators Association

هنكمل كلامنا عن sensitivity and specificity
وعشان نفهم اكتر هنطبق اللي قولناه امبارح علي مثال عملي

A new test for diagnosis of hepatitis C is applied to 500 individuals, the test indicated that 53 of them are diseased and 400 of them are not.
Actually 70 patients are suffering from the disease while 430 are not.
(We know that: * Sensitivity = True +ve / (True +ve + False -ve)

* Specificity = True -ve / (True -ve + False +ve)
So, we have to identify the previous terms, and we will get the following: True +ve: 53
True -ve: 400
False +ve: 30
False -ve: 17
Total: 500
Sensitivity = 53 / (53+17) = 0.7
Specificity = 400 / (400+30) = 0.9
Finally,sensitivity and specificity which are used for evaluating the clinical tests and give us a statistical measurements of their values and compare them with the current Gold Standard test (a test gives the best diagnosis of a given disease

عيدنا واجزنا نرجع تاني للاحصاء بتاعتنا =D

In this week, we are going to talk briefly about sensitivity and specificity

Is the patient has the disease? Yes / No
What is the result of the test? Positive / Negative

and we will get one of the following:
1- The patient has the disease and the test result is positive, this is called a “True +ve”
2- The patient has the disease and the test failed to indicate that: False -ve.”

3- The patient doesn't have the disease and the test result is positive: “False +ve”
4- The patient doesn't have the disease and the test result is negative: “True -ve”

* Sensitivity = True +ve / (True +ve + False -ve)
This means that the sensitivity of a given test is the ability of that test to indicate 'correctly' the DISEASES individuals and the result rages from 0 to 1.
The perfect sensitivity is 1, in this case the test can indicate all patient with the disease correctly. In other words: High sensitivity | Negative test | role out.

* Specificity = True -ve / (True -ve + False +ve)
This means that the specificity of a given test is the ability of that test to indicate correctly the HEALTHY individuals and the result also results from 0 to 1.
A test with perfect sensitivity (which is 1) can indicate all healthy individuals correctly. In other words: High specificity | Positive test | role in.

★ FINER criteria of a good research question:

• Feasible: Is the question answerable?
•Interesting: The question has to be interesting to the investigator, but should also be interesting to others.
• Novel: Has this study been done before? Does it add to the current body of medical knowledge?
• Ethical: Can the study be done in a way that does not subject subjects to excess risks?
• Relevant: Will it further medical science?