Can calculate the revised probability (RP) from the initial probability (IP) of the diagnosis by knowing what 3 things? What is this called?
The prior probability of the disease
The probability of a positive test result with the disease
The probability of a positive test result without the disease
Called Bayes' Theorem
Good as screening test in disease with low prevalence; low false-negative (FN) rate; e.g. cancer screening tests, PSA, ELISA, HIV, etc.
High Sensitivity
A test with a low false-positive (FP) rate; good for confirming a positive obtained on a screening test; e.g. a positive ELISA HIV test must then be confirmed with a Western Blot assay
High Specificity
The percentage of patients with a "positive" result from a population with the disease (TPR)
Sensitivity of True Positive Rate (TPR)
The percentage of people with a "negative" test in a population without the disease (TNR)
Specificity of True Negative Rate (TNR)
False positive rate = ?
FPR = 1 - specificity
False negative rate = ?
FNR = 1 - sensitivity
SNOUT
SeNsitivity "N" test rules out or Sn-N-Out disease:
High sensitivity + Negative result --> Rules out diagnosis
SPIN
SPecificity "P" rules IN or Sp-P-IN disease:
High specificity + Positive result --> Rules in diagnosis
The ratio of the probability of finding a positive test in the presence of disease to the probability of a positive test in the absense of disease