Sensitivity and Specificity

Validity (accuracy): Extent to which a test measures what it is supposed to measure.

Sensitivity:      1. Ability of test to correctly classify an individual as diseased.

                        2. Probability of being test positive when disease is present.

	D+	D-
T+	A	B
T-	C	D

SnNOUT: Highly sensitive test if negative rules out the disease

Specificity:      1. Ability of test to correctly classify an individual as disease free.

                        2. Probability of being test negative when disease is absent.

                                                                                                

SpPIN: Highly specific test if positive rules in the disease.

PPV: Positive predictive value:

1.     % of patients with positive test who actually have the disease

2.     Probability of patient having disease when test is positive

NPV: Negative predictive value:

1.     % of patients having disease when test is positive

2.     probability of patient having disease when test is positive

Bayes Theorem:

PPV: Highly dependent on prevalence of disease

Parallel testing:

A-test or B-test: (A, B) sensitivity or specificity

Combined sensitivity: Sn= A+B-AB

Combined specificity: Sp=A*B

Sensitivity will increase and specificity will decrease

Series testing:

A-test or B-test: (A, B) sensitivity or specificity

Combined sensitivity: Sn= A*B

Combined specificity: Sp=A+B-AB

Sensitivity will decrease and specificity will increase

PSM / COMMUNITY MEDICINE by Dr Abhishek Jaiswal is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Based on a work at learnpsm@blogspot.com.
Permissions beyond the scope of this license may be available at jaiswal.fph@gmail.com.