Quality Assurance Flashcards
Parameter
The whole population
Statistic
Sample of the population
Population
set of homogeneous items or events that is of interest for some question or experiment.
Sample
A sample is a smaller group of members of a population selected as a representative of the population.
Sample is biased
Random Sample
Randomly means that every member of the target population has an identical chance of being included in the sample. In other words, the method you use for selecting your sample can not be biased.
quality
Documented data use of statisticial calculations to compare lab data against other confirmed means of testing
what are the 2 medicine statistical tests
Reference test (Gold standard) Comparison test (Procedure to be validated)
when does a test become gold standard?
When it is the most consistent and most accurate test for a particular disease
What is a comparison test
Any other testing procedure, competing technology or variation from the gold standard
Making statistical indices more intuitive
Realize that all indices start with a presumption that the Gold Standard is perfect
Pay particular attention to the English of the statistical terminology
Pay close attention to the labels and layout of any table of data
Golden rule
For all purposes of statistical indices, the assumption is that the gold standard s perfect
If the results of a test match the results of a gold standard
Test=true
If the results of a test do not match the results of a gold standard test
Test=false
Both comparison and gold standard test are attempting to do what?
Predict the presence of disease
If the result is positive
it means that the test predicted the presence of disease
If the result was negative
Test predicted the absence of disease
True
is when the test results match the gold standard
False
is when the test results do not match the gold standard
True Positive
The test is correct (T) because it matches the gold standard and the test is positive for disease (P)
True Negative
The test is correct (T) because it matches the gold standard and the test is negative for disease (N)
False Positive
The test is incorrect (F) because it does not match the gold standard and the test is positive for disease (P)
False negative
The test is incorrect (F) because it does not match the gold standard and the test is negative for disease (N)
Gold standard
Assumed to be perfect
Number of tested Patients with disease
Have disease=TP+FN
Patients without disease
No disease=TN+FP
Number of tested patients in the population that the test predicts to have the disease
Test positive: TP+FP
Number of tested patients in the population that the test predicts do not have the disease
Test negative: TN+FN
Sensitivity
Implies the ability of a test to detect disease
Times the test correctly predicted disease divided by al of the the times the disease really does exist
Sensitivity equation
TP/TP+FN (*100%)
Specificity
Refers to the ablity of a test to detect the absence of disease relative to the number of people who in reality do not have disease
Specificity equation
TN/TN+FP (*100%)