Interpreting Diagnostic Tests Flashcards
What are the two ways you may be able to diagnose a disease?
- Detection of the changes caused by the agent
- Detection of the agent/ cause of the disease
What is the accuracy of a test?
The accuracy of a test is its ability to give a true measure of the substance assessed
What is the precision of a test?
The precision of a test relates to how consistent the results are (i.e the proportion of true negatives)
What is the sensitivity of a test?
The ability of a test to detect true positives (for example the ability of a test in detecting how many diseased animals there are)
out of everyone who tested positive for a disease, how many were also positive in the test
What is the specificity of a test?
The ability of a test in detecting non diseased animals (e.g true negatives)
What is a ‘gold-standard’ test?
A test that is completely accurate in detecting a disease or infection
What is a false positive?
The test is positive when the individual does not have the disease
What is a false negative?
The test is negative when the individual does have the disease
What is the positive predictive value?
the probability that an animal testing positive is actually positive
What is the negative predictive value?
The probability that a test negative animal is actually negative
What is the ROC?
Receiver Operating Characteristic
What does the ROC curve plot?
True positive rate against False positive rate
What is the meaning of agreement in terms of testing?
This refers to how well two tests give similar results
What method would you use to measure agreement for continuous data?
The pearson correlation coefficient
What method would you use to measure agreement for categorical/ quantitative data?
Cohens kappa
What is meant by prevalence?
The proportion of diseased cases in the population
What is True prevalence?
The actual prevalence of a disease in the population
What is apparent prevalence?
An estimate of true prevalence based on the results of an imperfect test
What does it mean when an ROC curve is closer to 1
the better the models ability to distinguish between positive and negative classification
