Data Interpretation Flashcards
How do we get reference intervals?
Fluctuations in a healthy population
What is the definition of health in veterinary medicine?
Absence of disease
Wellbeing of the animal
What are the categories to define a population?
Species or breed
Age
Location
Climate
Season
In general, how much of the healthy population falls OUTSIDE the reference interval?
5%
(2 standard deviations from the mean)
What are the 3 phases errors occur in data interpretation?
Pre-analytical: before the sample gets to the analyzer. (most common source or error)
Analytical: issue with the machine
Post-analytical: uploading, transcription, interpretation
Instrument function, knowledge and training on the instrument, and manuals of standard operating procedure are all factors that can influence what phase of error in data interpretation?
Analyzer error
The probablility that a diseased patient tests positive for the disease is determined by the test’s ______.
Sensitivity
If a test is 99% sensitive, how many patients with the disease will test positive?
How many patients with the disease will test negative?
99 patients with the disease will test positive (true positive)
1 patient with the disease will test negative (false negative)
If 93 or 100 animals known to have a disease test positive for the disease, and 7 of 100 animals known to have a disease test negative for the disease, what is the sensitivity of the test used?
Sensitivity = TP/(TP + FN)
= 93/(93+7)
= 93% sensitivity
*True positives divided by all patients that HAVE the disease (true positives and false negatives)
The probability that a healthy patient tests negative for a disease refers to what feature of the test?
Specificity
If 87 of 100 healthy patients test negative for a disease, or 13 out of 100 healthy animals test positive, what is the specificity of the test?
Specificity = TN/(TN+FP)
=87/(13+87)
=87%
*Healthy patients (true negatives) / All animals that don’t have the disease (true negatives and false positives)
What is the formula for determining the the prevalence of a disease in a population?
Prevalence = (TP + FN)/ Entire Population
*All positives (true positives and false negatives) divided by the entire population.
Confidence in a positive test result is high is that test has a high ____ _____ ____.
Positive predictive value
*This test will have very few false positives
Confidence in a negative test result is high when the test has a high _____ _____ _____.
Negative predictive value
*This test will have very few false negatives